Symbiogenetic evolution of complex cells and complex plastids

Europ.j. Protisto!' 29, 131-143 (1993) May 28,1993

Review

Symbiogenetic EVolution of Complex Cells and Complex Plastids* Peter Sitte Institut tiir Biologie II, Zellbiologie, Universitat Freiburg, Freiburg, Germany

SUMMARY It isgenerallyaccepted todaythat mitochondria andplastidsofeukaryoticcells("eucytes")have their phylogenetic origins in prokaryotic cells ("protocytes") that had been taken up into urkaryotic host cells as intracellular symbionts.This concept, stronglysupported (among other evidence) by comparisons of rRNA and protein sequence data, has many important consequences for understanding both cellular evolution and cellular cornpartmentation. Accordingto theSerial Endosymbiont Theory (SET), the eucyte came aboutby the formation of stable intracellular symbioses ofquitedifferentcells.Theformation ofsuch symbioses is referred to here as intertaxoniccombination (ITC). In addition to mutation and genetic recombination, ITC emerges as a third progressive power in evolution. The situation can be complicated by repeated ITC. This is discussed in detail by taking the evolution of complex plastids as an example.Plastids of this kind, possessing 3 or 4 enveloping membranes insteadof2,arewidespread inalgae.Theyappear to be remnantsofeukaryotic, and phototrophic, endocytobionts in phagotrophic host cells. The phylogeny of complex plastids could recently be fully reconstructed in the case of cryptomonads, and partly alsoin the case of Chloraraclmion.

Introduction Incosmic dimensions, the biosphere of this planet is little more than a flimsy layer on its surface. The vast number of individuals of far more than one million species are crowded within that layer. Consequently, living organisms constitute a considerable part of the environment of other organisms. This proximity has led to the evolution of many varied relationships. They are particularly intimate between symbionts (or para sites) and their respective hosts. Contrary to mutual symbioses, parasitic systems are often less stable, and many para sitic systems have eventually been transformed into mutualistic ones (ef., e.g., [103]). An impressive example of stable cellular symbiosis is provided by the lichens [92). Morphologically, physiologically, and ecologically they appear as single organisms. * Adapted from a lecture presented at the Spring Meeting of the German Society of Cell Biology, Konstanz, March 16, 1992. © 1993 by Gustav Fischer Verlag, Stuttgart

Although their true nature as symbiosesof fungi with algae was discovered long ago [112), even today these "superorganisms" are treated as a class of organisms of their own in the textbooks, despite their compound character. Symbiosis in lichens is extracellular, or epicytobiotic. A more intimate form of a companionship of two taxonomically different cells (or more than two) is intr acellular symbiosis, or endocytobiosis [110]. In this case, one cell (the microsymbiont) is entirely enclosed in a larger host cell, the cellular macro symbiont, that is, one cell is encapsulated by a completely different other one. Endocytobiosis Amazing as the phenomenon of endocytobiosis is, it is by no means rare [1,5 7,95, 124, 135, 139]. Some better known examples are the nitrogen-fixing Rhizobium bacteroids in endopolyploid root cortex cells of leguminous plant s [10, 85, 127). Sulphur bacteria, living within the trophosome cells of Pogonophora such as Riftia, help to 0932-4739/93/0029-0131$3 .50/0

132 . P. Sitte

power the biocoeno ses aro und "black smokers" in the deep sea [15,56] and in gutless oligochaetes of other sulfide biotopes [27, 33]. Saprope1ic protozoa contain endocytic methanogens which supplement the anaerobic metaboli sm of their hosts (d. [128, 138]). Many insects with unbalanced nutrition have up to 7 different species of endocytic bacteria [9, 89]. Geosiphon, a zygomycete living in soil, contains Nostoc cells and these cyanobacteria funct ion as "chloroplasts" [62, 102]. The zoochlorellae of green amoebae, the green species of Paramecium, and the green Hydra (d . [59]) render their hosts at least partl y ph ototrophic. All larger foriminifera are hosts of endosymbiotic algae - chlorophytes, rhodophytes, diatoms or dinoflagellates [72]. It is only thr ough the photosynthetic activity of endocytic Symbiodinium cells [4, 98] that the massive aragonite skeletons of the corals can be deposited so rapidly [64]. There are, of course, many more examples of endocytobiotic systems (cf., e.g., [48, 119, 120]). A comparative overview yields some trends which apply to all of them. In our conte xt, two of these trend s appear relevant: (1) The formation of endocytobiotic systems is an interspecific (heterologous) phenomenon . Homologous (intraspecific) associations of cells lead typically to cytoplasmic fusion (formation of syncytia) and sometimes to nuclear fusion (syngamy; Fig. 1). Whereas in the laboratory, under experimental conditions, cells of very distantl y

related organisms can be forced to fuse, under natural conditions only cells belonging to the same taxon may fuse. On the oth er hand, however, homologous cells apparently do not form stable endo cytobioses. Th is is not really surprising. Many kind s of phagocytes take up smaller cells. On the other hand , cells of any kind remain separated from each other due to their negative surface charges, and this short-range repulsion can be abolished only by species-specific adhesion molecules (or, in the laboratory, by artificial means such as polyethylene glycol in high concentration, hemolytic viruses, and electroporati on), (2) Endocytobionts are excluded from gametes. Consequently, vertical transmission of endocytobionts from one sexual generation to the next does not occur, apparently to prevent vertical spreading of endoc ytic parasites. The re-infection of egg cells or develop ing embryos with indispensable endosymbionts is ensured by particular, sometimes sophisticated, mechanisms.

The Serial Endosymbiont Theory (SET), Part I The basic tenet of the SET is well known tod ay and broadl y accepted [20, 40, 81, 119, 130, 14 1]. It state s that all DNA-containing comp artm ents of a eucyte out side the nucleus are xenosomes [16], that is descendant s of once

1

2

+

•

Fig. 1. Under natural conditions, cell fusion (1) is restricted to cells of the same species (formation of zygotes, or syncytia). Cells belonging to different taxa may form endocytobiotic systems (2) in which they however remain separ ated by a non-plasmatic space. No such systems can be formed by identical cells (1"), nor will alien cells fuse under natural conditions (2* ) (from [116]).

Evolution of Complex Cells and Plastids . 133

free-living cells. These were incorporated into ho st cells as endocytobionts and have since been integrated to become organelles in the course of an intimate co-evolution of guest and host. The genomes of xenosomes are the remain s of genomes that once evolved independently from their present surroundings. It is thought that part of these genomes was transferred to the host cell nucleus. Eventuall y a situation was reached wherefrom the xenosomes/organelles were no longer capable of independent existence. Their vertical transmission was rendered possible since they were no longer regard ed as "non-self" but as "self" by the host organism and consequently were no longer exclud ed from forming gamet es. The typ ical eucyte is not actually a single cell but a compound " mosaic" cell, and hence a genetic chimaera. Whene ver the phylogeny of such a "supercell" is to be worked out , the question cannot just be where "it" came from , but rather, where did the diverse genomes come from. Th ere will be as many answers as there are different sorts of organelles in a given cell. For each geno me has had its own past, its own evoluti on, and consequently has its non-interchangeable position in a phylogenetic tree. (It neverthel ess seems advisable, for obvious reasons, to give host cell characters prim e importance in all questions of taxonomy.) For more than 100 years the endosymbiont theory posed as one of the most da ring and pro vokin g hypothe ses in both cell and evolutionary biology. Since it has been confirmed for plastids and mitochondria by sequence dat a it now app ears well established. It should not be forgotten, however , that many facts in favour of it were know n already before sequence data became available [116].

Some Corollaries of the SET In 1978, Schwartz and Dayhoff publ ished a first "combined phyletic tree" , based on sequence comparisons of ferredoxins, 5S rRNAs, and c type cytochromes [108]. In this tree, plastids of green plants (chlorobionts) show up amidst the cyano bacterial cluster, far distant from chlorobionts. Likewise, mitochond ria are located among aerobic purple bacteria (a-s ubdivision), remote from their present car riers. Even if one takes into account the considerable uncertainties of this early wo rk, it nevertheless provided an impre ssive verification of the SET. Since then, compelling evidence has been gath ered from numerous sequence comparisons showing that plasrids as well as mitoch ondria are descendant s of eubacterial, endocytobiot ic prokaryotes [36,38,41,52,65,66,90,91,93, 101, 104, 109, 147]. This docs not mean , of course, th at there were not also serious conceptual difficulties. Three main problems are still acute: (1) The majority of organellar proteins are not coded for in the orga nellar DN A but in the nucleus so that the proteins arc translated on 80S cytoribosomes and need to be translocated to their final places post-translationally. (2) H ow could the endocytobionts have been "p ersuaded" to give away energy-rich compo unds such as ATP,

PEP, or triose phosph ates to their hosts (and possibly also, how could the host s have ad apted to release nutrients to the symbionts)? (3) Into what kind of cells were the progenitors of plastids and mitochondria incorporated ? At least preliminary answers to these questions can now be provided. As to the first question, extensive gene transfer from symbionts to their hosts has been envisaged as a possible explanation. In recent years it could be demon str ated that intracellular gene transfer is not just an ad hoc postulate but reality [2, 23, 28, 67, 73, 99, 107, 132, 133, 146]. Although the molecular mechan ism of gene tra nsfer is still an issue of speculati on [8, 106, 114], proof of interspecific gene transfer in some quite different systems [49,5 4, 126] and the possibilit y of achieving gene tran sfer by relat ively simple ar tificial means [74] leaves little doubt that gene transfer has played an important role in the evolution of complex cells. Recently, yeast systems have been found in which actual gene transfer can be studied experimentally [50, 51, 131]. Extensive gene transfer in an endocytobiotic system has been demonstrated convincingly in the case of Cyanophoraparadoxa [61]. The endocytic cyanelles of this organism still possess remnants of a peptidoglycan wall, so their eubacterial origin is indubitable. Thei r circular genome is, however, much smaller than the one of free-living cyanobacteria, and even smaller than typical plastid DNA s [5, 70]. Consequently, it cannot code for all cyanellespecific proteins. By use of inhibitors it has been shown that about 90 % - both qualitatively and quantitati vely of the soluble proteins of cyanelles are nucleus-encoded [100]. Mo re specifically, ferredoxin-Nafrl'f -reducrase, a cyanellar enzyme, is nucleus-encod ed in Cyanophora [3, 99]. But the sequence of this protein exhibits 72% similarity to its counterparts in cyanobacteria, and only 32 to 42 % similarity to corresponding plant enzymes. The demonstration of gene transfer among different compa rtments of eucytes, mainly in the direction of the cell nucleus, raises the question of why there still are any remain s of genomes of form er endosymbionts in xenosomic organelles. According to a hypoth esis proposed by von Heijne [137], only genes coding for prot eins not containing any potenti al hydroph ob ic expo rt signals could have been successfully transferred from the mitochondrial to the nuclear genome. (A different hypoth esis has been put forward by Jacobs [55].) On the other hand, it has been speculated that peroxisomes (micro bodies) could be xenoso mic compartments, the genome of which has been partly deleted, partl y transferred to the nucleus [14, 71], so that peroxisomes (like the QO-mutants of yeast mitochondria ) are devoid of ow n DN A. Peroxisomes thu s obtain all of their specific proteins by import mechanisms resembling tho se of mitochondria and plastids. Concerning the second issue, that of the secretion of energy-rich compounds by microsymbionts, it is to be rememb ered that small cells in general are leaky. Furth ermore, certain strains of Chiorella secrete sugars even in an apo symb iotic state [26, 60]. In Dictyostelium, cAMP is used as an extra cellular chemical signal [7].

134 . P. Sitte As to the third question, the oldest known eucyte microfossils are not much older than 2,000 millions of years [45,63]. However, according to sequence data eukaryotes are supposed to have evolved early in the Archaic. Since then they must have evolved separately, in parallel along with the archae a and the eubacteria [21, 143]. The early eucytes presumably already had some eukaryotic characters such as relatively large dimensions, endomembranes, linear chromosomes with telomeres and centromeres, a cytoskeleton and a spindle apparatus, they supposedly possessed the ability of phagocytosis [17], meiosis and syngamy, so that they were able to reproduce sexually. Yet they did not contain mitochondria nor plastids, and they derived their energy from fermentation. These unicellular or plasmodial "urkaryotes" [142], or "proto-eukaryotes" [125], were supposedly the predators of the ancient biosphere. A few years ago, Cavalier-Smith expressed his conviction that some urkaryotic organisms still exist in the present-day biosphere [11]. These are protoctists which do not contain mitochondria nor plastids, and this apparently not due to secondary loss. In fact, the "primitiveness" of these protoctists (including Polobiontida, Diplomonadida, Retortomonadida, Trichomonads, and Oxymonads) is endorsed through sequence studies. Among these presently living urkaryotes ("archezoa", d. [12, 13]) is the large amoeboflagellate Pelomyxa palustris, which lives in anaerobic and micro aerobic freshwater habitats. It characteristically contains three different types of endocytic bacteria [140], amongst them two methanogens [136]. On the whole, the former endosymbiont hypothesis became transformed into an acknowledged theory with extensive relevance, an enormous heuristic potential, and a correspondingly large explanatory capacity. Some otherwise puzzling facts can be understood on the basis of that theory. One example is provided by the ion pumps [37,88]. These membrane ATPases are classified as P, V, and F ATPases. Proton-translocating V ATPases (which are found exclusively on eukaryotic cytomembranes) are large complexes (> 400 kDa), and - contrary to the smaller P ATPases - they are not phosphorylated during the action cycle. Membrane ATPases of the F type likewise translocate protons without being phosphorylated, and they too are complexes of more than 400 kDa. The physiological significance of FATPases stems from the fact that they function, under physiological conditions, not as ATPases but as ATP synthases. However, these synthases only evolved in the realm of eubacteria. In eucytes they are exclusively found at the inner membranes of mitochondria and at the thylakoids of chloroplasts. The urkaryotel eukaryote transition is, among other things, also marked by the acquisition of F ATPases by endocytobiosis with eubacteria as microsymbionts.

General Consequences New biological entities can be created by establishing stable endocytobioses [82, 111, 116, 129]. Phyletic trees not only ramify but may also coalesce to form networks.

Entirely different genomes are combined within a single cell so that they henceforth have a common fate and evolve further in strict coevolution. Processes of this kind are, since the days of Mereschkowsky [86], collectively referred to under the term "symbiogenesis". For a long time, mutation and recombination have been known as the driving forces of evolution. But there is a third one, which I have called intertaxonic combination (ITC; [116, 118]). ITC is the prerequisite of symbiogenesis (as syngamy is the prerequisite for commencing ontogeny). lTC, ~:l,e the other two expansive forces of evolution, is based on chance and singularities. ITC might be looked upon as an exaggerated kind of recombination. By recombination, non-identical genomes of a given species are, by syngamy, brought together in a single cell. In lTC, fusion of gametes is replaced by the formation of a stable intracellular symbiosis. In that case, too, different genomes are put together in a single cell. However, they remain located in different compartments. Recombination in a more strict sense can then be enacted only if and when gene transfer has taken place.

Endosymbiont Theory, Part II Eucytes are, with the exception of the amitochondrial taxa, mosaic cells. This suggests that there is an enormous selective advantage conveyed by symbiogenesis. However, evolutionary successful ITC apparently came about only seldom since many favourite singularities need to happen concomitantly. How often did ITC lead to the formation of plastids or mitochondria? Are these organelles monophyletic or polyphyletic? And what do these terms actually mean in the present context? It appears appropriate to regard as "monophyletic" all organelles that phyletically go back to one single event of II'C. In contrast, "polyphyletic" evolution of xenosomes would be based on multiple processes ofITC where diverse prokaryotes have become incorporated into similar or different hosts, thus creating different progenitors of diverse evolutionary lineages, each of which then might have radiated further in a monophyletic fashion (Fig. 2) [94, 119]. The following discussion will be concentrated on plastids. (For mitochondria, see [6, 38, 39, 41]). It appears likely that there have been several instances of ITC that eventually led to phototrophic eucells [39, 76, 94]. Solid evidence in favour of this assumption is provided by the existence of "complex" plastids in many protoctists [115, 116].

The Evolution of "Complex" Plastids Complex plastids are plastids with more than two enveloping membranes (as they are typical for green algae and all plants), three in phototrophic euglenids and dinoflagellates, and four in all of the (otherwise quite heterogeneous) "chromophyte" algae [42]. More than 50% of all algal species possess complex plastids [19].

Evolution of Complex Cells and Plastids . 135

•• Fig. 2. Monophyletic (right) and polyphyleticorigin (left) of xenosomes and DNA-containing organelles. In the case of monophyletic evolution, diversification of the genomes of xenosomes/organelles came about within diversifying host cells/organisms, whereas polyphyletic evolution is based upon several endosymbioticevents with different symbionts (and hosts).

Som e years ago, Tomas and Cox put forward a hypoth esis [134] that complex plastids are descendants of eukaryotic endocytobionts th at had been taken up into protozoan host cells. In mo st cases, th e endosymbionts ha ve since been reduced to th e only organelle the host was ori ginally devoid of, that is the plastid(s) (Fig. 3). According to this view (substanciated further mainly by Gibb s [29, 30, 32, 35]) there were two successive stages in the evolution of plastid-containing cells. Whereas " regular" plastids are the result of " prima ry" lT C, that is the incorpo ration of cyano bacteria into urkar yoti c cells, complex plastid s derive from "seco ndary" lT C, consisting of an upt ak e of eukaryotic photot rophs into protozoan host cells [115, 116 , 117, 123]. In most cases it is only supernumera ry envelope membranes (derived from the symbiont's cell membrane and/o r th e phagosome membr ane) th at point to a second process of ITe. Th ere are several lines of evidence suppo rting the above con cept . M any protoctists are known to be composed of different cells. Paramecium bursaria, for exa mple, is a mosaic of five sorts of cells [96] . It acco rdingly contai ns five different kinds of genomes (even if centrioles and/or basal bodies are not pr esumed to cont ain DNA, d. [44,58 ]). Euglenids ("Euglenozoa") and dinoflagellat es (" Dinozoa ") in almost every respect behave as protozoa [116]. However, the best proof of a phyletically secondary ITC is pro vided by protoctists endowed with eukaryotic end osymbionts that are partially, but not fully, reduced to

I

o complex plastids. Such cryptomonads [34].

IS

appa rently the situation

In

Cryptomonads as "Connecting Links" Th e crypto monads are phototrophi c biflagellates po ssessing a flagellar pocket and associated ejectisomes (Figs. 4, 5). Th e plastid s are complex, and there is a narrow plasma tic space between the two pairs of envelop e membran es. This interspace does not contain mitochondria no r dictyosomes, but it contains some 80S ribosomes [83] and a small nucleus-like or gan elle, th e nucleomorph [43]. Th e nucleomorph contains DN A, if only in minimal amounts. According to Gibb s [31], the nucleomorph is the vestigial nucleus of a eukary otic endosy mbiont . Thus the cells of cryptom on ad s are apparently composed of (at least) four different cells, two prokaryoti c and two euka ryot ic, and they acco rdingly contai n four different geno mes. Nuclear and mitochondrial geno mes belong to th e original host cell, whereas the genomes of th e nucleomorph and the plastid (s) stem fro m th e endo cytobiont . In the last few years, th e four genomes could be characterised and their respective phylogenies elucidated [24,77, 78, 79, 80, 121]. In the author 's laboratory, pure nuclear fractions could be obtain ed from mass cultures of Pyrenomonassalina [46] (Fig. 6 a). From total DNA, three fractions could be recovered the mino r peaks of which are

136 . P. Sitre

a

I

b

I

Fig. 3. "Primary" and "secondary" intertaxonic combin ation (ITC). By primary lT C, bacter ial cells were taken up into urkaryotic (a) or eukaryotic cells (b). By secondary lTC, a phototrophic eukaryote symbiont was taken up into a phagotrophic eucyte (c). In this case, the symbiont may become reduced to its plastid (s), thus ending up as a "complex" plastid in its host cell (e). An intermediate stage of the reductive evolution (d) is exemplified by the cryptomonads which still cont ain the vestigial nucleus ("nucl eomorph" ) of the original endosymbiont (from [116]).

mitochondrial and plastid DNA. Both of them are circular [47]. In Pyrenomonas, the nucleomorph is inserted into the pyrenoid of the plastid and can be isolated together with this suborganelle that is devoid of DNA [46]. From a pure fraction of pyrenoid/nucleomorph complexes (Fig. 6 b), three very small linear chromosomes in a 1: 1 : 1 stoichiometry could be isolated. The length of these chromo-

Fig. 4. A stylized cryptomonad cell with ejectisomes (E), nucleus (N), mitochondria (M), and a large, cup-shaped complex plastid (P). The compartment between the two pairs of plastid envelope membrane s contains starch grains (5) and some 805 ribosomes, together with a "nucleornorph" (N m),

somes is between 195 and 250 kb. Thus the genome of the nucleomorph, comprising not more than 660 kb in the haploid state, is the smallest eukaryotic genome known toda y [25]. After sequencing the genes of the ssu rRNAs from all four genome s the phyletic ancestry of all four genomes could be established [24, 77- 80]. It is for the first time that the build-up of a compound mosaic cell could be entirely cleared up in a phyletic, and evolutionary, context. The nuclear genome of the cryptomonads clusters near the green algae (Fig. 7), and the mitochondrial genome too is closest to the ones of chlorobionta. That means that the host cell of the cryptomonads descended from the same taxonomic range as did the ancestor s of the green algae. On the other hand, both plastid DNA and nucleomorph DNA as the endosymb iont's DNAs are derived from (or close to) the red algal lineage. Both results agree with what has been found for Cryptomonas ep by the Halifax group [22].

Evolution of Complex Cells and Plastids . 137
Fig. 5. Pyrenomonas salina, longitudinal section, same orientation and abbreviations as in Fig. 4 (flagella not visible). The nucleomorph (Nm) is almost enclosed in the pyrenoid (Py) (Electron micrograph: courtesy by Heinz Falk).

Future Problems In the case of the cryptomonads, there has presumably been gene transfer from the nucleomorph to the host cell nucleus. Which genes remained in the nucleomorph, and why? What proteins, if any, are synthesised on the relatively few 80S ribosomes in the remaining cytoplasm of the reduced symbiont? Only their own ribosomal proteins (so that the nucleomorph simply would be a rudimentary organelle)? And how can plastid-specific proteins, the genes of which have been transferred from the plastid to the nucleomorph and finally to the host cell nucleus, cross the 3, 4, or 5 membranes between the place of their synthesis and the one of their final destination? Have some genes perhaps been relocated to plastid DNA? We do not know yet. Still more intriguing is the question why nucleomorphs are not encountered more frequently. We know of many animal cells that harbour complete algal cells, and of many "algal" cells with complex plastids. However, apparent intermediate stages of reduction of endocytic algae to complex plastids are found only seldom. As yet nucleomorphs have only been found in cryptomonads and in Chlorarachnion reptans [53, 75]. Presumably, the integration of an endocytic symbiont goes on rather quickly once it has commenced. If so, the nucleomorph must be

Fig. 6. Isolated pyrenoid-nucleomorph complexes (a, b) and nuclei (c) of Pyrenomonas salina. a, phase contrast and DAPI fluorescence, demonstrating the DNA content of the nucleomorphs; bar = 10 urn. b, c, electron micrographs; bars = 1 urn (Courtesy by Stefan Eschbach). •

138 . P. Sitte

Distance 0.1

1----100 100

NC Gallus gallus Ch NC M us musculus Ch

100 -

NC Drosophila melanogaster Ar NC Saccharomyces cerevisiae Fu

~

NC Neurospora crassa

64

-

NC Glycine max Di

100

NC Zea mays Mo

100

-

100 '- 90 100 I

I

NC PYRENOMONAS SALINA Kr NC Cryptomonas phi Kr NC Tribonema aequaIe Cr

~ .--

NC Volvox carteri Cl NC Chlamydomonas reinhardtii Cl

50 -

82

Fu

NC OChromonas danica Cr NC Tetrahymena australis ci

'-5~

NC Paramecium tetraurelia ci

- 58

.

NC Plasmodium malariae Ac

100 I

I

58 100

NC Plasmodium falciparum Ac

NC Crypthecodinium cohnii Df NM PYRENOMONAS SALINA Kr

100 I

I

100

NM Cryptomonas phi Kr NC Palmaria palmata Rh

100 I

I

NC Gracilaria 1emaneiformis Rh NC Eugl ena gracilis Eu

100 I 100 100

I

PT

Pisum sativum Di

Anacystis nidulans Cy Escherich'la coli Ba Halococcus morrhuae Arch

Evolution of Complex Cells and Plastids . 139 On the other hand, however, con vergent evolution, leading to such complicated and yet almos t identical stru ctures appears extremely unlikely, as well. So far no idea can be offered on how to solve thi s parad ox. There are of course some more funda mental problems concerni ng the evolution of complex cells. To mention just one, the question has been raise d recentl y of whether the proro-eukaryoti c cell itself was a product of symbiogenetic events [105, 125, 151]. According to rRNA trees, euk aryotes are more distan t from both euba cteria and archa ea (archebacteria) than these two pr okaryotic superkingdoms are from each other [143,145] (see, however, also [68, 97]). Conversely, according to protein sequence comparison s and some other criteria, the eukaryotes show grea ter affinities to the lineage of sulphur-metabolizing archaea (acido-thermophiles; also termed "eocytes" by Lake [69], or "Crenarcbaeota" by Woese et al. [145]) tha n th e archaea ro eubacteria [87,1 13, 144,148,149,150]. To solve th is parad ox, a symbiogenetic origin of the euka ryote cell nucleus has been prop osed [125, 151]. App arentl y, much mor e sequence (and other?) dat a are necessar y to answer questio ns of th is kind in a satisfying way . Ackno wledgements Fig. 8. Part of a plasmodium of Cblorarachnion reptans with mitochondria (M), and chloroplasts (CP) with pyrenoids ("), Nucleomorphs (arrows) are deeply inserted into the pyrenoids. Bar = 1 urn (Electron micrograph: Courtesy by Volker Speth). somewha t of a " metas ta ble sta te" in the ongoing process of nuclea r reduction. Th e pla stids in cryptomonad s are not at all related to th e ones of Chlorarachnion as judged from fine structure and pigment complement. Still, th e EM morphology of the nucleomorphs of Chlorarachnion is identical to the one observed in the cryptomonads [53]. As in Pyrenomonas, the nucl eomorphs of Chlorarachnion are inserted into the plastidal pyrenoids (Fig. 8). Since Chlorarachnion grows slow ly it is difficult to wo rk with. Nevertheless, recently it coul d be show n that the nucleom orph of Chlorarachnion, as the one of cryptornonads, harbou rs three very small chromosomes hybrid ising against euka ryotic rDNA [79, 84, 122]. So an almos t ident ical situation has been reached in two evolutionar y lineages that, according to all known facts, ha ve different phyletic origins, so that homology seems to be excluded.


Work supported by Deutsche Forschungsgemeinschaft (Schwerpunktprogramm "Intrazellullare Symbiose"). I am indebted to Dr. Uwe-Gallus Maier for critically reading the manuscript, and to Dr. John A. Thompson for checking the English.

References 1 AhmadjianV.and Paracer S. (1986):Symbiosis. Anintroduction to biological associations. Univ. Press of New England, Hanover. 2 Baldauf S. L. and Palmer J. D. (1990): Evolutionary transfer of the chloroplast tufA gene to the nucleus. Nature, 344, 262-265. 3 Bayer M. G., Maier T. L., Gebhart U. B. and Schenk H. E. A. (1990): Cyanellar ferredoxin-NADP+-oxidoreductase of Cyanophora paradoxa is encoded by the nuclear genomeand synthesized on cytoplasmic 80S ribosomes. CurroGenet., 17, 265-267. 4 Blank R. J. and Huss V. A. R. (1989): DNA divergency and speciation in Symbiodinium (Dinophyceae). Plant Syst, Evol., 163, 153-163.

Fig. 7. Phyletictree derived fromsmallsubunit rRNAsequences. For sequencesother than the onesof Pyrenomonasand corresponding references d. [18]. Construction of the tree was done with the program packageTREECON 2.1 [Yves Van de Peer, Antwerp]. Matrix calculation was done with theJukes-Cantor correction for multiple mutations. The trees were constructed with the neighbour-joining method. An alignment including ssu rRNA sequences from species of all kingdoms with a total of 3023 positions was used for the calculation. Bootstrap analysis has beenperformed as a pseudo-empirical test to check for the reliability of the tree topology. Numbers at the forks indicate how often the group to the right of the fork did appear in 100 trees. - Abbreviations: NC = nucleus; NM = nucleomorph; PT = plastid. Arch = Archaea. Ba = Eubacteria, including Cy = Cyanobacteria. Protoctista: Eu = Euglenida; Ac = Apicornplexa; Of = Dinoflagellata; Ci = Ciliophora; Kr = Cryptophyta; Rh = Rhodophyta; Cr = Chrysophyra; CI = Chlorophyta. Spermatophyta: Mo = Monocotyledonae; Di = Oicotyledonae. Fu = Ascomycota. Animalia: Ar = Arthropoda; Ch = Chordata.- The different respective positions of nuclear and nucleomorph sequences in cryptomonads show up quite clearly: Whereas the nuclear rRNAsequences are closeto the chlorobionta cluster, the corresponding sequencesof nucleomorph RNAs are affiliated to the ones of red algae (Courtesy by Stefan Rensing and Uwe-G. Maier).

140 . P. Sitte 5 Bohn ert H. J. and Loffelhardr W. (1992) : Mo lecular genetics of cyanelles fro m Cyanopboraparadoxa. In: Reisser W. (ed.): Algae and symbioses: plants, animals, fungi, virus es, inter action s explore d, pp . 379 -397. Biopress Ltd. , Bristol. 6 Bonen L. (199 1): The mito chondrial geno me: so simple yet so complex. Curr. Opinion Genet . Developm., 1, 5 15-522. 7 Bonn er J. T . (196 7): The cellular slime mold s. Princeton Univ. Press, Princeton. 8 Brenn icke A. (199 1): Genverschiebung iiber RNA . Bio!. in uns. Ze it, 21, 284. 9 Buchn er P. (1965): Endosymb ioses of animals with plant microorganisms. Inter science, N ew York. 10 Caeta no-Anolles G. and GresshoffP . M. (1991): Plant genetic contro l of nodulation. Annu . Rev. Microbiol., 45, 345 -3 82. 11 Cavalier-Smith T. (1987) : Euk aryotes with no mitochondria. Natu re, 326, 332-333. 12 Cavalier-Smith T. (1988) : Euka ryote cell evolution. In: Gre uter W. and Zimmer B.: Proc, XIV Intern. Bot. Congr. Berlin 1987, pp. 203-223. Koeltz Sci. Book s, Konigstein, 13 Cava lier-Smith T. (1989) : Archaebacteria and archezoa . Nature, 339, 100-1 0 1. 14 Cava lier-Smith T. (1990) : Symbio tic origin of pero xisomes. In: N ord on P. (ed.): Endocy to biology IV, pp. 5 15-521. INR A, Paris. 15 Cava na ugh C. M . (1985): Symbioses of chemoautotrophic bacte ria and marine invertebra tes from hydrothermal vents and reducing sediments. Bull. Bio!. Soc. Wash., 6, 373 -388 . 16 Corliss J. O. (1985): Concept, definition, pr evalence, and host-interactions of xenoso mes (cytoplasmic and nuclear endosy mbionts). J. Prot ozoo!., 32, 373-376. 17 De Duve C. (198 9): Th e pr imitive ph agocyte. In: N ard on P. (ed.): Endocytobiology IV, pp . 5 11- 51 4. INRA , Paris. 18 Derijk P., Neefs J. M. , Van der Peer Y. and De Wachter R. (199 2): Compilation of sma ll ribosomal subunit RN A sequences. Nucl. Acids Res., 20, Suppl., 20 75-2090. 19 Dodg e J. (1973) The fine struc ture of algal cells. Acad em. Press, London. 20 Doolittle W. F. (1980): Revoluti on ary concepts in evolutionary cell biology. Trends Biochem. Sci., 5, 146-149. 21 Doolittle W. F. (1987): Th e evolutio nary significance of the archae bacteria. Ann. N .Y. Acad . Sci., 503, 72-77. 22 Dou glas S. E., Murphy C. A., Spencer D. F. and Gra y M. W. (199 1): Crypto mona d algae are evolutiona ry chimaera s of two ph ylogeneti cally distinct unicellular eukary otes . Nature,

30

31

32

33

34

35

36

37

38 39 40 41

42

43

350, 148-1 51. 23 Ellis J. (1982) : Promi scuous DN A - chloroplast genes inside plant mitochondria. N ature, 299 , 678 -679. 24 Eschbach S., Wolters J. an d Sitte P. (199 1): Primary and seconda ry stru ctu re of th e nuclear sma ll subunit ribo som al RN A of the cryptomo na d Pyrenomonas salina as inferred fro m the gene sequence: evolutionary implications. J. M ol. Evol., 32, 247-252. 25 Eschbach S., Hofmann C. J. B., Maier U.-G., Sitte P. and Han smann P. (199 1): A eukaryotic geno me of 660 kb: electro pho retic karyotype of nucleom orph and cell nucleus of the cryptom on ad alga, Pyrenomonas salina. Nucl. Acids Res., 19, 177 9-1 781. 26 Fischer A., Meindl D. and Loos E. (1989): Glucose excretion by th e symbiotic Chlorella of Spongillafluvia tilis. Planta, 179, 25 1- 256. 27 Fliigel H. J. and Langhof I. (1983): A new hermaphroditic pogonophore from the Skagerr ak. Sarsia, 68, 131-138. 28 Gellisen G. and Michaelis G. (1987): Gene transfer. Mitochondria to nucleus. Ann. N .Y. Acad. Sci, 503, 391-401. 29 Gibbs S. P. (1978): Th e chloroplast of Euglena may have

44

45

46

47

48 49 50

evolved from symbiotic green algae. Can. J. Bot. , 56, 2883 -2889. Gibbs S. P. (1981): Th e chloro plasts of some algal gro ups may have evolved fro m endosy mbiotic euka ryotic algae. Ann. N.Y. Acad. Sci., 361, 193-208. Gibbs S. P. (1983) : Th e cryptom on ad nu cleom orph: is it the vestigia l nucleu s of a euka ryotic endosymbiont? In: Schwe mmler W. and Schenk H. E. A. (eds.): Endocytobi ology II, pp. 987-992. de Gru yter, Berlin. Gibbs S. P. and Gillott M. A. (1980): H as the chloroplast of crypto monads evolved from an euka ryo tic symbiont? In: Schwe mmler W. and Schenk H . E. A. (eds.): Endocyto biology I, pp . 73 7-743. de Gru yter, Berlin. Giere O. and Langgheld C. (1987) : Structural organisation, transfer and biological fate of endosymbiotic bacteria in gutless oligochaetes. Mar. Biol., 93, 641-650. Gillott M. A. (1990): Phylum crypto phyta (cryptomonads). In: M argulis L., Corlis s J. 0., Melkonian M . and Chapman D. J. (eds.): Handbook of pr otoctista, pp. 139-151. Jones & Barlett, Boston . Gillott M. A. and Gibbs S. P. (1980) : Th e crypto monad nucleomorph: its ultrastructure and evolutionary significance. J. Phycol., 16, 558-568 . Giova nno ni S. ]., Turner S., Olsen G. J. , Barns S., Lan e D. J. and Pace N. R. (1988): Evolutionary relation ships among cyano bacteria and green chlorop lasts . J. Bacteriol., 170, 3584-3592. Gogarten J. P., Kibak H. , Ditt rich P., Taiz L., Bowman E.]., Bowman B. J., Manolson M. F., Poole R.]., Date T., Oshima T., Koni shi]., Denda K. and Yoshida M. (1989): Evolution of the vacuolar H +-ATPase: implications for the origin of euka ryotes. Proc. Natl. Acad. Sci. USA, 86, 666 1-6665. Gray M . W. (1988) : O rga nelle origi ns and ribosom al RN A. Bioch em. Cell Biol., 66, 325 -348. Gray M . W. (1989): Th e evolutionary orig ins of organelles. T rends Genet., 5, 294-299. Gray M . W. and Doolittle W. F. (198 2): Ha s the endos ymbiont hypothesis been proven ? M icrobiol. Rev., 46, 1-42. Gra y M. W., Cedergren R., Abel Y. and Sankoff D. (1989) : On the evolutionary origin of the plant mitochondrion and its geno me. Proc. Natl. Acad . Sci. USA, 86, 2267-2271. Green J. c., Leadbeater B. S. C. an d Diver W. L. (1989 ): Th e chromophyte algae. Probl ems and perspectives. Clarendon Press, Oxford. Greenwood A. D., Gri ffith H. B. and Santo re U. J. (1977): Chloroplasts and cell compart ments in Cryptophyceae . Brit. Phycol.]., 12, 119. Hall J. L., Ram anis Z. and Luck D. J. L. (1990): Basal body, centrio lar DN A: molecular genetic studies in Chlamydomonas. Cell, 59, 121-132. Han T. -M. and Runnegar B. (1992 ): Megas copic euk aryoti c algae from the 2.1-billion-year -old Nega unee Iron -Form ation , Michigan . Science, 25 7, 232-235. Han sman n P. and Eschbach S. (1990): Isolat ion and prelimina ry characterization of th e nucleus and the nucleomorph of a crypto mo nad, Pyrenomonas salina. Eur . ]. Cell BioI., 52, 373 -3 78 . Han smann P., Maerz M. and Sitte P. (1987): Investigati on s on geno mes and nucleic acids in crypto mona ds. Endo cyt. Cell Res., 4, 289 -295 . Han smann P., Maerz M . and Sitte P. (198 9): Cytosymbiosis. Progr. Bot., 51, 21-47. Heinemann J. A. (1991 ): Genetics of gene transfer betw een species. Trends Genet., 7, 181-1 85. HeinemannJ. A. and Spragu e G. F.J r. (1991a): Transmission of plasmid DNA to yeast by conjugation with bacteria. Meth. Enzymol. , 194, 187-195.

Evolution of Complex Cells and Plastids . 141 51 Heinemann J. A. and Sprague G. F. Jr. (1991b): Bacterial conjugative plasmids mobilize DNA transfer between bacteria and yeast. Nature, 340, 205-209. 52 Hemmingsen S. M., Woolford c., van der Vries S. M., Tilly K., Dennis D. T., Georgopoulos C. P., Hendrix R. W. and Ellis R. J. (1988): Homologous plant and bacterial proteins chaperone oligomeric protein assembly. Nature, 333, 330-334. 53 Hibberd D. J. and Norris R. E. (1984): Cytology and ultrastructure of Chlorarachnion reptans (Chlorarachniophyta divisio nova, Chlorarachniophyceae classis nova). J. Phycol., 20,310-330. 54 Houck M. A., Clark J. B., Peterson K. R. and Kidwell M. G. (1991): Possible horizontal transfer of Drosophila genes by the mite Proctolaelaps regalis. Science, 253, 1125-1129. 55 Jacobs H. T. (1991): Structural similarities between a mitochondrially encoded polypeptide and a family of prokaryotic respiratory toxins involved in plasmid maintenance suggest a novel mechanism for the evolutionary maintenance of mitochondrial DNA. J. Mol. Evol., 32, 333-339. 56 Jannasch H. W. and Mottl M. J. (1985): Geomicrobiologyof deepsea hydrothermal vents. Science, 229, 717-725. 57 Jeon K. W. (ed.) (1983): Intracellular symbiosis. Int. Rev. Cytol., Suppl. 14. Academic Press, New York. 58 Johnson K. A. and RosenbaumJ. L. (1990): The basal bodies of Chlamydomonas reinhardtii do not contain immunologically detectable DNA. Cell, 62, 615-619." 59 Kessler E., Huss V. A. R. and Rahat M. (1988): Speciesspecific ability of Chlorella strains (Chlorophyceae) to form stable symbioses with Hydra viridis. Plant Syst. Evol., 160, 241-246. 60 Kessler E., Kauer G. and Rahat M. (1991): Excretion of sugars by Chlorella species capable and incapable of symbiosis with Hydra viridis. Bot. Acta, 104, 58-63. 61 Kies L. (1992): Glaucocystophyceae and other protists harbouring procaryotic endocytobionts. In: Reisser W. (ed.): Algae and symbioses: plants, animals, fungi, viruses, interactions explored, pp. 353-377. Biopress Ltd., Bristol. 62 Knapp E. (1933): Dber Geosiphon pyriforme Fr. Wettst., eine intrazellulare Pilz-Algen-Symbiose. Ber. Deutsch. Bot. Ges., 51, 210-216. 63 Knoll A. H. (1992): The early evolution of eukaryotes: a geological perspective. Science, 256, 622-627. 64 Kremer B. P. (1985): Symbiose von Algen und Korallen. Naturwiss. Rundschau, 38, 508-516. 65 Kuntzel H. and Kochel H. G. (1981): Evolution of rRNA and origin of mitochondria. Nature, 293, 751-755. 66 Kuhsel M. G., Strickland R. and Palmer J D. (1990): An ancient group I intron shared by eubacteria and chloroplasts. Science, 250, 1570-1573. 67 Lacoste-Royal G. and Gibbs S. P. (1985): Ochromonas mitochondria contain a specific chloroplast protein. Proc. Natl. Acad. Sci. USA, 82, 1456-1459. 68 Lake J A. (1988): Origin of the eukaryotic nucleus determined by rate-invariant analysis of rRNA sequences. Nature, 331, 184-186. 69 Lake J A. (1991): Tracing origins with molecular sequences: metazoan and eukaryotic beginnings. Trends Biochem. Sci., 16,46-50. 70 Lambert D. H., Bryant D. A., Stirewalt V. L., Dubbs J M., Stevens S. E. and Porter R. D. (1985): Gene map for the

* The negative results of these authors are in agreement with unpublished results obtained in the author's laboratory by H. Falk, P. Hansmann and V. Speth, with basal bodies of

Pyrenomonas salina.

Cyanophora paradoxa cyanelle genome. J. Bacteriol., 164, 659-664. 71 Lazarow P. B. and Fujiki Y. (1985): Biogenesis of peroxisomes. Annu. Rev. Cell Biol., 1, 489-530. 72 Lee J. J. (1992): Symbiosis in foraminifera. Taxonomy of algae symbiotic in foraminifera. In: Reisser W. (ed.): Algae and symbioses: plants, animals, fungi, viruses, interactions explored, pp. 63-92. Biopress Ltd., Bristol. 73 Liaud M.-F., Zhang D. X. and Cerff R. (1990): Endosymbiotic transfer of genes into the nucleus: evolutionary and genetic implications inferred from the nuclear genes encoding subunits GAPA and GAPB of chloroplast glyceraldehyde3-phosphate dehydrogenase from pea. Proc. Natl. Acad. Sci. USA, 87, 8918-8922. 74 Liebler E. M. and Weber H. (1989): Genetic manipulation of animal and plant cells. Endocyt. Cell Res., 6, 149-169. 75 Ludwig M. and Gibbs S. P. (1987): Are the nucleomorphs of cryptomonads and Chlorarachnion the vestigial nuclei of eukaryotic endosymbionts? Ann. N.Y. Acad. Sci., 503, 198-211. 76 Luttke A. (1991): On the origin of chloroplasts and rhodoplasts: protein sequence comparison. Endocyt. Cell Res., 8, 75-82. 77 Maerz M. and Sitte P. (1991): Isolation, physical map and gene map of mitochondrial DNA from the cryptomonad Pyrenomonas salina. Plant Mol. Biol., 16, 593-600. 78 Maerz M., Wolters J, Hofmann C. J B., Sitte P. and Maier U.-G. (1992): Plastid DNA from Pyrenomonas salina (Cryptophyceae): physical map, genes, and evolutionary implications. Curr. Genet., 21, 73-81. 79 Maier U.-G. (1993) The four genomes of the alga Pyrenomonas salina (Cryptophyta). BioSystems (in press). 80 Maier U.-G., Hofmann C. J B., Eschbach J., Wolters J and Igloi G. L. (1991): Demonstration of nucleomorph-encoded eukaryotic small subunit ribosomal RNA in cryptomonads. Mol. Gen. Genet., 230, 155-160. 81 Margulis L. (1981): Symbiosisin cell evolution. Freeman, San Francisco. 82 Margulis L. and Fester R. (1991): Symbiosis as the source of evolutionary innovation. MIT Press, Cambridge, Mass. * 83 McFadden G. 1. (1990): Evidence that cryptomonad chloroplasts evolved from photosynthetic eukaryotic endosymbionts. J Cell Sci., 95, 303-308. 84 McFadden G. 1., Gilson P. R., Hofmann C. J B., Adcock G. J and Maier U.-G. (1993): Evidence that a phylum of photosynthetic amoebae arose by eukaryote/eukaryote endosymbiosis (submitted). 85 Mengel K. and Pilbeam D. J (eds.) (1992): Nitrogen metabolism of plants. Clarendon, Oxford. 86 Mereschkowsky C. (1910): Theorie der zwei Plasmaarten als Grundlage der Symbiogenesis, einer neuen Lehre von der Entstehung der Organismen. BioI. Centralbl., 30, 278-303, 321-347,353-367. 87 Nakayama M. and Kohiyama M. (1985): An a-like DNA polymerase from Halobacterium halobium. Eur. J. Biochem., 152, 293-297. 88 Nelson N. and Taiz L. (1989): The evolution ofH+-ATPases. Trends Biochem. Sci., 14, 113-116. 89 Oertel B. und Schaller G. (1984): Symbiose von Insekten mit Mikroorganismen. BioI. Rundschau, 22, 303-329. ,. See also Scannerini S. (1989): Is symbiosis the source of new trends in evolution? Endocyt. Cell Res., 6, 241-246. - Smith J. M. (1989): Generating novelty by symbiosis. Nature, 341, 284-285. - Charlesworth B. (1991): Living together. Nature, 352, 391.

142 . P. Sitte 90 Pace N. R., Olsen G.]. and Woese C. R. (1986): Ribosomal RNA phylogeny and the primary lines of evolutionary descent. Cell, 45, 325-326. 91 Parthier B. (1982): Transfer RNA and the phylogenetic origin of cell organelles. BioI. Zbl., 101, 577-596. 92 Peveling E. (1988): Beziehungen zwischen den Symbiosepartnern in Flechten. Naturwiss., 75, 77-86. 93 Piechocki R. und Borner T. (1981): Die evolutionare Verwandtschaft pro- und eukaryotischer 5S rRNA. BioI. Zbl., 100, 129-153. 94 Raven P. H. (1970): A multiple origin for plastids and mitochondria. Science, 169, 641-646. 95 Reisser W. (ed.) (1992a): Algae and symbioses: plants, animals, fungi, viruses, interactions explored. Biopress Ltd., Bristol. 96 Reisser W. (1992 b): Endosymbiotic associations of algae with freshwater protozoa and invertebrates. In: Reisser W. (ed.): Algae and symbioses: plants, animals, fungi, viruses, interactions explored, pp. 1-20. Biopress Ltd., Bristol. 97 Rivera M. C. and Lake]. A. (1992): Evidence that eukaryotes and eocyte prokaryotes are immediate relatives. Science, 257, 74-76. 98 Rowan R. and Powers D. A. (1991): A molecular genetic classification of zooxanthellae and the evolution of animalalgal symbioses. Science, 251, 1348-1351. 99 Schenk H. E. A. (1992): Cyanophora paradoxa, identification and sequencing of nucleus-encoded cyanellar proteins. A proof of gene transfer. Endocyt. Cell Res., 8, 215-222. 100 Schenk H. E. A., Bayer M. G. and Zook D. (1987): Cyanelles. From symbiont to organelle. Ann. N.Y. Acad. Sci., S03, 151-167. 101 Schlegel M. (1991): Protist evolution and phylogeny as discerned from small subunit ribosomal RNA sequence comparisons. Europ. ]. Protistol., 27, 207-219. 102 Schnepf E. (1964): Zur Feinstruktur von Geosiphon pyriforme. Ein Versuch zur Deutung cytoplasmatischer Membranen und Kompartimente. Arch. Mikrobiol., 49, 112-131. 103 Schnepf E. (1992): From parasitism to symbiosis: the dinoflagellate example. In: Reisser E. (ed.): Algae and symbioses: plants, animals, fungi, viruses, interactions explored, pp. 699-710. Biopress Ltd., Bristol. 104 Schon A., Kannangara C. G., Gough S. and Soli D. (1988): Protein biosynthesis in organelles requires misaminoacylation of tRNA. Nature, 331,187-190. 105 Schubert I. (1988): Eukaryotic nuclei of endosymbiontic origin? Naturwiss., 75, 89-91. 106 Schuster W. and Brennicke A. (1987): Plastid, nuclear and reverse transcriptase sequencesin the mitochondrial genome of Oenothera: is genetic information transferred between organelles via RNA? EMBO J., 6, 2857-2863. 107 Schuster W. and Brennicke A. (1988): Interorganellar sequence transfer: plant mitochondrial DNA is nuclear, is plastid, is mitochondrial. Plant Sci., 54, 1-10. 108 Schwartz R. M. and Dayhoff M. O. (1978): Origins of prokaryotes, eukaryotes, mitochondria, and chloroplasts. Science, 199, 395-403. 109 Schwarz Z. and Kossel H. (1980): The primary structure of 16S rDNA from Zea mays chloroplast is homologous to E. coli 16S rRNA. Nature, 283, 739-742. 110 Schwemmler W. (1979): Endocytobiose und Zellforschung. Naturwiss., 66, 366. 111 Schwemmler W. (1989): Symbiogenesis: macro-mechanism of evolution. de Gruyter, Berlin. 112 Schwendener S. (1860-68): Untersuchungen uber den Flechtenthallus. Nageli's Beitr. wiss. Bot., 2, 109-186; 3, 127-198; 4, 161-202.

113 Searcy D. G. (1987): Phylogenetic and phenotypic relationships between the eukaryotic nucleocytoplasm and thermophilic archaebacteria. Ann. N.Y. Acad. Sci., 503, 168-179. 114 Sitte P. (1981): Die Endosymbionten-Hypothese - eine kritische Betrachtung zur Zell-Evolution. Nova Acta Leopold., NF 56,41-58. 115 Sitte P. (1987): Zellen in Zellen: Endocytobiose und die Folgen. In: Liist R. (Hrsg.): Beobachtung, Experiment und Theorie in Naturwissenschaft und Medizin, S. 431-446. Wiss. Verlagsges., Stuttgart. 116 Sitte P. (1990): Phylogenetische Aspekte der Zellevolution. BioI. Rundschau, 28, 1-18. 117 Sitte P. (1991): Die Zelle in der Evolution des Lebens.BioI.in uns. Zeit, 21, 85-92. 118 Sitte P. (1993): "Intertaxonic combination": introducing and defining a new term in symbiogenesis. Proc. Endocytobiology V (in press). 119 Sitte P. and Eschbach S. (1992): Cytosymbiosis and its significance in cell evolution. Progr. Bot., 53, 29-43. 120 Sitte P. and Hansmann P. (1986): Cytosymbiosis. Progr. Bot., 48, 30-55. 121 Sitte P. and Maier U.-G. (1992): Evolution of "complex plastids" from eukaryotic endosymbionts. Endocyt. Cell Res., 8, 223-225. 122 Sitte P. and Maier U.-G. (1993): Cryptomonad evolution as a paradigm for cell evolution. Proc. Endocytobiology V (in press). 123 Sitte P., Eschbach S. and Maerz M. (1992): The role of symbiosis in algal evolution. In: Reisser W. (ed.): Algae and symbioses: plants, animals, fungi, viruses, interactions explored, pp. 711-733. Biopress, Bristol. 124 Smith D. C. and Douglas A. E. (1987): The biology of symbioses. Edward Arnold, London. 125 Sogin M. L. (1991): Early evolution and the origin of eukaryotes. Curr. Opinion Genet. Developm., 1, 457-463. 126 Sprague G. F. Jr. (1991): Genetic exchange between kingdoms. Curr. Opinion Genet. Developm., 1, 530-533. 127 Stacey G., Burris R. H. and Evans H. ]. (eds.) (1992): Biological nitrogen fixation. Chapman & Hall, New York. 128 Stumm C. K. and Zwart K. B. (1986): Symbiosisof protozoa with hydrogen-utilizing methanogens. Microbiol. Sci., 3, 100-105. 129 Taylor F. ]. R. (1974): Implications and extensions of the Serial Endosymbiosis Theory of the origin of eukaryotes. Taxon, 23, 229-258. 130 Taylor F. ]. R. (1987): An overview of the status of evolutionary cell symbiosis theories. Ann. N.Y. Acad. Sci., 503, 1-16. 131 Thorsness P. E. and Fox T. D. (1990): Escape of DNA from mitochondria to the nucleus in Saccharomyces cerevisiae. Nature, 346, 376-379. 132 Timmis]. N. and Scott N. S. (1984a): Homologies between nuclear and plastid DNA in spinach. Theor. Appl. Genet., 67, 279-288. 133 Timmis J. N. and Scott S. N. (1984b): Promiscuous DNA: sequence homologies between DNA of separate organelles. Trends Biochem. Sci., 9, 271-273. 134 Tomas R. N. and Cox E. R. (1973): Observations on the symbiosisof Peridinium balticum and its intracellular alga. I. Ultrastructure. ]. Phycol., 9, 304-323. 135 Trench R. K. (1979): The cell biology of plant-animal symbioses. Ann. Rev. Plant Physiol., 30, 485-531. 136 van Bruggen].]. A., van Rens G. L. M., Geertman E.]. M., Zwart K. B.,Stumm C. K. and VogelsG. D. (1988): Isolation

Evolution of Complex Cells and Plastids . 143 of a methanogenic endosymbiont of the sapropelic amoeba Pelomyxa palustris Greef. ]. Protozool., 35, 20-23 . 137 von Heijne G. (1986): Why mitochondria need a genome. FEBS Letters, 198, 1-4. 138 Wagener S., Bardele C. F.and Pfennig N. (1990): Functional integration of Methallobacterium [ormicium into the anaerobic ciliate Trimyema compressum. Arch. Microbiol., 153, 496-50l. 139 Werner D. (1987): Pflanzliche und mikrobielle Symbiosen. Thieme, Stuttgart. 140 Whatley J. M. (1976): Bacteria and nuclei in Pelomyxa palustris: comments on the theory of serial endosymbiosis. New Phytol., 76, 527- 569. 141 Whatley J. M. and Whatley F. R. (1984): Evolutionary aspects of the eukaryotic cell and its organelles. In: PirsonA. and Zimmermann M. H. (eds.): Encyclopedia of plant physiology, N.S. 17, pp. 18-5 8. Springer, Berlin. 142 WoeseC. R. (1981): Archabakterien- Zeugen aus der Urzeit des Lebens. Spektrum d. Wiss., 1981/8, 75- 91 (Scientif. Amer., 244/6, 94-1 06). 143 Woese C. R. (1987): Bacterial evolution. Microbiol. Rev., 51,221-27l. 144 Woese C. R., Magrum L. J. and Fox G. E. (1978): Archaebacteria. J. Mol. Evol., 11, 245-252.

145 Woese C. R., Kandler O. and WheelsM. L. (1990): Towards a natural system of organisms: proposal for the domains archaea, bacteria, and eucarya. Proe. Natl. Acad. Sci. USA, 87, 4576- 4579. 146 WolfeK. H., Morden C. W. and Palmer] . D. (1991): Ins and outs of plastid genome evolution. Curr. Opinion Genet. Developm., 1, 523-529. 147 Wolters J. and Erdmann V. A. (1984): Comparative structural analysis of 5S rRNA - a powerful tool for studying chloroplast evolution. In: GlaeserP. S. (ed.): The role of data in scientific progress, pp. 103-109. Elsevier, North Holland. 148 Zillig W. (1987): Eukaryotic traits in archaebacteria: could the eukaryotic cytoplasm have arisen from archaebacterial origin? Ann. N.Y. Acad. Sci., 503, 78- 82. 149 ZilligW., Schnabel R. and Stetter K. O. (1985): Archaebacteria and the origin of the eukaryotic cytoplasm. Curr. Topics Microbiol. Immunol., 114,6- 18. 150 Zillig W., Schnabel R. and Tu ] . (1982): The phylogeny of archaebacteria, including novel anaerobic thermoacidophilesin the light of RNA polymerase structure. Naturwiss., 69, 197-204. 151 Zillig W., Klenk H.-P., Palm P., Leffers H., Piihler G., Gropp F. and Garrett R. A. (1989): Did eukaryotes originate by a fusion event? Endocyt. Cell Res., 6, 1-25.

Key wor ds: Symbiogenesis - Endosymbiosis - Interta xonic combination - Plastid evolut ion - Crypto monads Peter Sitte, Institut fur Biologic II, Zellbiologie, Universirat Freiburg, Schanzlestralie 1,0-7800 Freiburg, Germany