Evidence for the Plausibility of an
Endosymbiotic Origin of Eukaryotic Organelles

In her book The Origin of Eukaryotic Cells, microbiologist Lynn Margulis outlined her hypothesis that eukaryotic cells (cells with nuclei, mitochondria, and sometimes chloroplasts) originated from the symbioses of prokaryotic cells. This is now known as  the endosymbiotic theory for the origin of eukaryotic cells.  Student Joshua A. Bond has done a very nice introduction to Margulis' Endosymbiotic Theory. Another, slightly more technical discussion can be found at Endosymbiosis and The Origin of Eukaryotes  from Kimball's Biology Pages  (an excellent resource for biological information).

The purpose of this page is to present some intriguing evidence supporting Margulis' endosymbiotic theory that is mentioned by Josh in his paper. Microbiologist Kwang Jeon has spent the last two or three decades working with strains of Amoeba proteus that have been infected with bacteria. In some cases, although many of the infected amoeba were killed by the infection, a few survived with populations of the bacteria remaining alive within the amoebas' cells. Over time, some of these surviving but infected amoebae became dependent on the bacteria within their cells ( K.W. Jeon 1972K.W. Jeon 1987 ).

Kwang and others showed that the amoebae could not survive without the internal bacteria using surgical methods ( K.W. Jeon 1972K.W. Jeon et al. 1976 ) . These methods consisted of removing the nucleus of an infected cell and placing it into another cell that had previously had its nucleus removed. A nucleus from an infected amoeba would only survive if the cell it was inserted into had the bacterial endosymbionts in its cytoplasm. If the endosymbionts were not present, the previously infected nucleus (and hence the entire cell) died.

Kwang also showed that the infected amoeba required the endosymbiotic bacteria using chemical methods (  K.W. Jeon et al. 1977 ). This was done by treating the infected amoebae with the antibiotic chloramphenicol (CAP). The antibiotic reduced the number of bacteria living in the amoeba to less than 10% of the level found in control lines (not treated with CAP). None of the amoebae that had all of their endosymbiotic bacteria killed were able to survive. (Note that the  amoeba's mitochondria were damaged by the CAP, but at least some of the amoebae retaining their symbionts were able to survive, whereas none of the amoebae that lost all of their endosymbiotic bacteria survived.)

Further evidence for the amoebae's dependence on the endosymbiotic bacteria was provided by K.W. Jeon 1995 and J.Y. Choi et al. 1997 . The researchers demonstrated that infected amoebae no longer produced a protein (an enzyme in this case) required for survival. However, the infected amoebae had some activity for that enzyme (about half the level found in normal, uninfected amoebae). It was shown that it was the bacteria providing the enzyme that the amoeba was no longer producing. Furthermore, if the bacteria were removed from the amoebae, the amoebae's nucleoli were damaged, apparently due to the loss of the enzyme formerly provided by the bacteria. It is interesting that the amoebae were no longer able to produce the enzyme because of interference from the infecting bacteria, but because the amoebae would be damaged if the bacteria were removed, the endosymbiotic bacteria were now required for the amoebae's survival.

Although none of this provides proof that the origin of the eukaryotic cell was due to endosymbiosis, it has been demonstrated that such an event can happen naturally even with modern cells.  This is evidence for the plausibility of Margulis' endosymbiotic theory of the origin of eukaryotic cells.


Science 1970 Mar 20;167(925):1626-7 Reassembly of living cells from dissociated components.

Jeon KW, Lorch IJ, Danielli JF

Science 1972 Jun 9;176(39):1122-3  Development of cellular dependence on infective organisms: micrurgical studies in amoebas.

Jeon KW

J Cell Physiol 1976 Oct;89(2):337-44 Endosymbiosis in amoebae: recently established endosymbionts have become required cytoplasmic components.

Jeon KW, Jeon MS

A strain of large, free-living amoeba that became dependent on bacterial endosymbionts which had infected the amoebae initially as intracellular parasites, was studied by micrurgy and electron microscopy. The results show that the infected host cells require the presence of live endosymbionts for their survival.Thus, the nucleus of an infected amoeba can form a viable cell with the cytoplasm of a noninfected amoeba only when live endosymbionts are present. The endosymbiotic bacteria are not digested by the host amoebae and are not themselves used as nutritional supplement. While the host amoebae are dependent specifically on the endosymbionts, the latter can live inside amoebae of different strains, indicating that their dependence on the host cells is not yet strain specific.

J Protozool 1977 May;24(2):289-93 Effect of chloramphenicol on bacterial endosymbiotes in a strain of Amoeba proteus.

Jeon KW, Hah JC

The effect of chloramphenicol (CAP) on the bacterial endosymbiotes of a strain of Amoeba proteus was studied by growing the symbiotic amebae in media containing 0.5-1.6 mg/ml CAP for up to 4 weeks. Treatments with CAP caused such ultrastructural changes as expansion of the nuclear zone and deformation of symbiotes. The CAP treatment also damaged the mitochondria, e.g. disappearance of central and protrusion of peripheral cristae. Number of bacteria per ameba decreased to less than 10% of control in CAP-containing media, but no viable amebae became completely free of symbiotes. The results supported previous studies that amebae were dependent on endosymbiotes.

J Cell Physiol 1979 Jan;98(1):49-57 Growth and electron microscopic studies on an experimentally established bacterial endosymbiosis in amoebae.

Ahn TI, Jeon KW

A strain of nonsymbiotic A. proteus was infected with endosymbiotic bacteria isolated from another strain of amoeba which had become dependent on the symbionts after a few years of spontaneously established symbiosis. In the newly infected amoebae, the bacteria avoided digestion and multiplied at a faster rate than the hosts, reaching the maximum carrying number (about 42,000 per amoeba) in fewer than ten cell generations of the hosts. The experimentally infected amoebae were also examined under the electron microscope, and the development of bacteria-containing vesicles was followed. The results show that the infective bacteria that were initially harmful to host amoebae have become harmless and that they have changed in their mode of multiplication during the course of establishing a stable symbiosis with their hosts.

Ann N Y Acad Sci 1987;503:359-71

Change of cellular "pathogens" into required cell components.

Jeon KW

J Eukaryot Microbiol 1995 Jan-Feb;42(1):1-7

                       The large, free-living amoebae: wonderful cells for biological studies.

                       Jeon KW.

                       Department of Zoology, University of Tennessee, Knoxville 37996, USA.

                       The large, free-living amoebae have been widely used as model cells for studying a variety of biological phenomena, including cell
                       motility, nucleocytoplasmic interactions, membrane function, and symbiosis. Results of studies by our group on amoebae as moving
                       cells, as material for micrurgical manipulations, and as hosts for intracellular symbionts are summarized here. In particular, our recent
                       studies of the amoeba as a microcosm, in which spontaneously infecting foreign microbes have become integrated as necessary cell
                       components, are described in some detail. These processes have involved an initial microbial infection, mutual adaptation by the host
                       and symbionts, and development of obligatory symbiosis. Evidence is presented to show that symbiont-derived macromolecules are
                       involved in the protection of symbionts from digestion, the symbionts have acquired regulatory elements on their chromosomal genes to
                       enhance production of beneficial gene products, and symbionts apparently utilize host-derived macromolecules to their benefit. These
                       studies involved morphological observations both at light and electron microscopic levels, physiological and genetic studies, production
                       and use of poly- and monoclonal antibodies, and molecular-biological approaches including gene cloning and sequencing. It is shown
                       that amoebae are uniquely suited as model cells with which to study these phenomena.

J Eukaryot Microbiol 1997 Sep-Oct;44(5):412-9

Evidence for symbiont-induced alteration of a host's gene expression: irreversible loss of
SAM synthetase from Amoeba proteus.

Choi JY, Lee TW, Jeon KW, Ahn TI

Department of Biology Education, Seoul National University, Korea.

Symbiont-bearing xD amoebae no longer produce a 45-kDa cytoplasmic protein that functions as S-adenosylmethionine synthetase in symbiont-free D amoebae. The absence of the protein in xD amoebae is attributable to xD amoeba's failure to transcribe the corresponding gene as a result of harboring bacterial symbionts. However, xD amoebae have about half the level of enzyme activity found in D amoebae, indicating that they use an alternative source for the enzyme. xD amoebae originated from D amoebae by bacterial infection and now depend on their symbionts for survival. xD amoebae exhibit irreversible nucleolar abnormalities when their symbionts are removed, suggesting that X-bacteria supply the needed enzyme. A monoclonal antibody against the 45-kDa protein was produced and used as a probe in cloning its corresponding cDNA. The product of the cDNA was found to have S-adenosylmethionine synthetase activity. These results show how symbiotic X-bacteria may become essential cellular components of amoeba by supplementing a genetic defect for an amoeba's house-keeping gene that is brought about by an action of X-bacteria themselves. This is the first reported example in which symbionts alter the host's gene expression to block the production of an essential protein.

J Eukaryot Microbiol 1997 Nov-Dec;44(6):614-9

                       A symbiont-produced protein and bacterial symbiosis in Amoeba proteus.

                       Pak JW, Jeon KW.

                       Department of Biochemistry, University of Tennessee, Knoxville, USA.

                       Gram symbiotic X-bacteria present in the xD strain of Amoeba proteus as required cell components, synthesize and export a large
                       amount of a 29-kDa protein (S29x) into the host's cytoplasm across bacterial and symbiosome membranes. The S29x protein
                       produced by E. coli transformed with the s29x gene is also rapidly secreted into the culture medium. Inside amoebae, S29x enters the
                       host's nucleus as detected by confocal and immunoelectron microscopy, although it is not clear if S29x is selectively accumulated inside
                       the nucleus. The deduced amino-acid sequence of S29x has a stretch of basic amino acids that could act as a nuclear localization
                       signal, but there is no signal peptide at the N-terminus and the transport of S29x is energy independent. The functions of S29x are not
                       known, but in view of its prominent presence inside the amoeba's nucleus, S29x is suspected to be involved in affecting the expression
                       of amoeba's nuclear gene(s).