All I have is a list of about 500 eukaryotic genera, and I need a neighbor-joining tree to show their relationship. I want to cast on top of this tree the loss of a particular enzyme. Is the best way to do this to make a 16S phylogeny?
16S phylogeny would only work for bacteria and archaea unless you are talking about mitochondrial or chloroplast 16S rRNA. Your phylogenetic marker choice should depend on which phyla you are looking at. I suggest you try using Ef-tu if your genera are highly diverse http://www.scs.illinois.edu/schulten/tutorials/ef-tu/eftu_tutorial.pdf
Once you have a tree in Newick format, iTOL has some nice web tools for overlaying data on trees http://itol.embl.de/itol.cgi