No one knows for sure why only eucarya form large multicellular organisms (bacteria can form small ones), but we can make a few observations that limit the scope of plausible answers:
- Large multicellular organisms are a fairly recent phenomenon. The eucaryal lineage is thought to go back perhaps 2.7B years[1] , but the first large multicellular organisms we have evidence for are the Ediacarian biota[2] , which emerged some 600M years ago. The actual lineages of plant and animals are much older, perhaps 1.5B years[3] . My point here is that eukaryotes in general, and plants and animals in particular, existed for a very long time before large organisms emerged.
- Large multicellular organisms are not common among the eucarya. We are large multicellular organisms, and we tend to notice other large multicellular organisms, but in truth we are rare, even among the eucarya, as this eucaryal phylogenetic tree shows:
Plants and animals are there alright (keep looking, you will find them eventually), but most eucarya are unicellular or are simple microscopic multicellular organisms no more complicated than a slime mold.
So having mitochondria, or a cytoskeleton or internal membranes may be necessary for the development of large multicellular organisms, but these features certainly are not sufficient and do not lead inexorably to the development of largeness. Most eucarya throughout most of history, including the present day, have been unicellular. Some other factor must be necessary also.
I’ll suggest that this missing ingredient is endogenous viruses – viruses that integrate into host DNA and persist to be passed down to its descendants. Endogenous viruses seem to be rare in the genomes of unicellular eukaryotes[4] , but are very common in animals, comprising up to 1% of the genome.
Integrated viruses can contribute to macroevolution through a number of mechanisms. They provide a source of novel genes, and can transfer genes between species. They facilitate genomic rearrangements, such as gene duplications, that provide opportunities for new genes and gene networks to evolve. They affect the expression of nearby genes – indeed the development of the placenta is controlled by endogenous viruses to a very surprising degree[5] . An excellent review of these mechanisms, along with their likely impact on evolution, can be found here[6] .
No doubt many other factors are involved in the evolution of large multicellular organisms. But I think a good case can be made that endogenous viruses facilitate large evolutionary jumps, and thus are likely to be a key contributor to this evolution. It’s possible that the ancestors of animals and plants were just not very good at keeping free-loading viruses out of their genomes. If so, it may be this apparent defect that made plants and animals possible. Evolution is not always efficient, but it is sometimes ironic.
Footnotes
[1] The Origin and Evolution of Cells
[5] Retroviruses facilitate the rapid evolution of the mammalian placenta
[6] Genomic creativity and natural selection: a modern synthesis