We will compare E. coli with H. sapiens. I will assume the volume of the bacteria to be 1 cubic micron, and the volume of the human cell as around 1000 cubic micron. There are larger human cells as well smaller ones but this should work. The volume of the nucleus is around 400 – 500 cubic microns – yes they are that big. That leaves us with 500 cubic micron to work with.
Macromolecules: Bacteria are estimated to have around 2 million cytosolic proteins whereas human cells have around 1 billion proteins copies per cell. In addition, bacteria have around 77 fg of RNA and 18.3 fg of DNA. Human cells have around 10 pg RNA. Assuming an average RNA size of around 1000 bp in bacteria, it gives us 77000 mRNA molecules. In addition, there are 375,000 tRNA molecules and 0.1 million rRNA molecules.
In contrast, human cells have around 7 milion rRNAs, 30 million rRNAs, and 0.4 million mRNA molecules in total.
However, not much data is available about intracellular distribution of small molecules. Assuming ionic concentration is pretty comparable, bacteria have around 2.5 million macromolecules packed in 1 cubic micron space as opposed to a human cell with 1.04e+9 macromolecules in 1000 cubic micron volume. That makes E. coli the winner.
Caveats: The shape and volume of proteins definitely affect cell crowding. We have also ignored the contribution of other organelles such as mitochondria/golgi/ER etc. Factoring those in could probably give us a completely different winner 😉 Peusner (1974) estimates 0.5 billion molecules per bacterial cell – much much higher than what we have above. Also, if we exclude the volume of nucleioid in a bacteria, it gets superdense in bacteria.