The endosymbiosis theory has been debunked for mitochondria

Ajith Harish and Charles Kurland at the Uppsala and Lund Universities in Sweden have in the article “Mitochondria are not captive bacteria” shown that the probability that mitochondria originated from proteobacteria is 44 orders of magnitude lower than the probability they were created locally. That increases considerably the probability that the Organelle Escape Theory describes the real origin of cellular structures. For more, see here.

Novelties and evolvability

Margulis did not only posit endosymbiosis as a source of organelles. REF? She posited that all novelties in eukaryotes are caused by endosymbiosis. In her opinion, bacteria are involved in an intelligent network covering the whole planet. In addition to working to the best for eukaryotes in a teleological manner it also supports our planet and maintains it in best possible ways. This theory was worked out in cooperation with James Lovelock. It is the "Gaia theory".  The intelligence in her model is thereby Gaia herself.

This way to regard bacteria can be seen as a solution to a serious problem in her scenario for cellular evolution. The problem is the origin of bacteria. According to the endosymbiosis theory their fusion starts evolution in a colossal tempo in eukaryotes. But the bacteria hardly evolve at all. That could be explained by the combination being so powerful. But there is another major problem. With Darwinian evolution even the bacteria must have evolved from single molecules. If Gaya created bacteria, then the problem is solved. Margulis criticized Darwinism especially because it is based on random mutations, which as she said:

 “Neo-Darwinism, which insists on (the slow accrual of mutations), is in a complete funk.”

I have illustrated Margulis´ way of thinking in the figure below: 

and my way in this figure: 

Organisms are illustrated by circles where complexity is the size of the circles. Evolvability is illustrated by the thickness of the arrows that show the direction of evolution. My theories do not rely on any intelligence. Instead they rely on evolvability, which is present in eukaryotes today, and which has been present almost from the beginning. There are a range of mechanisms that are typically present in eukaryotes and which are parts of the evolvability apparatus. They are: sex, diploidy, introns and protection against too high selection pressure. These mechanisms give organisms access to ingenious variants (e.g. through new combinations) and avoids stagnations at suboptimal solutions.

Evolution of energy production adapted for RNA world in a "pre-RNA world"

Most of the metabolic processes are today driven by enzymes, i.e. proteins. This was known long before the genetic systems, with transcriptions and translation were discovered. Today we know that genetics is necessary for protein synthesis, but in the first half of the twentieth century a common thinking about molecular biology was that proteins could somehow self-replicate. Most scientists today agree that that is not taking place and has never taken place. But there are still some strange hypotheses claiming that such processes in early life periods existed, or that proteins could somehow spontaneously appear in specific sequences.

William Martin and Michael J. Russell have postulated generation of energy in the form GTP or ATP, and they have formulated this suggestion:

"Thermodynamic considerations related to formyl pterin synthesis suggest that the ability to harness a naturally pre-existing proton gradient at the vent–ocean interface via an ATPase is older than the ability to generate a proton gradient with chemistry that is specified by genes."

They have suggested that specific peptide chains survive better than others, and that enzymes were formed from these. This suggestion is most probably not possible at all, and it is totally unnecessary, because life got its energy for formation and handling of RNA molecules in much simpler ways. They suggest that large parts of the metabolic apparatus was already present, based on these, when the first RNA molecules appeared. A lot of enzymes are suggested, forming the reverse Krebs cycle and the Wood-Ljungdahl pathway. But the most improbable and quite teleological suggestion is that the highly complex membranous ATPase was present before RNA. This complex changes an RNA molecule from a high energetic to a low energetic variant or vice versa. The high energetic variant was on the young earth readily present as pyrophosphate and other phosphate chains. RNA molecules were built from these.