There are over 50 different protein units (called subunits) spread out over seven different proteins needed to read a gene. The uninspiring names of these proteins are TFIIB, TFIID, TFIIE, TFIIF, TFIIH , mediator and RNA polymerase II (pol II). Don't blame me, I didn't name them.
The TFII part means transcription factor for RNA polymerase II. Remember, transcription is what happens when the gene in the DNA is read into RNA. The letters are just an artifact from the way these proteins were originally discovered years and years ago. Now that we know their names, let's see how all of these proteins work together to read a gene.
Imagine a piece of DNA at the start of a gene. Roger Kornberg and his group have done a lot of work on figuring out how the cell decides whether or not this gene will be read. We aren't going to deal with any of this here though.
Once the cell has "decided" to read a gene, the next step is to build a train station for the train to leave from. Of course this station isn't made of bricks. Instead it is made of proteins. These proteins have to be put together as carefully as any bricks at a train station.
The train station gets built mostly using TFIID. TFIID is a massive protein with at least thirteen different protein subunits.
TFIID lands on a spot close to the start of the gene (often at a place called the TATA box). Once there it bends the DNA about 90 degrees and gets the DNA ready for the train.
Now this next part is less clear. The train might come in one piece. Or it might be built up at the station, the cars all getting connected there. Scientists are still trying to work this out.
Either way, the train ends up sitting between the bent DNAsort of like an animal in DNA jaws. The train is at the main part of the station, waiting for its signal to start.
The most important part of the train is pol II. This is the protein responsible for making RNA from the DNA.
TFIIB connects TFIID (the station) and pol II (the train). And the other proteins all rest on top of the train and some of them touch the DNA in front of the train, sort of like a cow catcher. Roger Kornberg and his group have helped work out some of the finer details of this complex.
For example, some recent work he did showed in detail how TFIIB is a bridge between the TFIID platform and the train, pol II. And how TFIIB works with pol II to keep the DNA track pried open to read.
The train at this point isn't actually touching the track. He also worked out when the train touches the track and starts moving.
All of this is good solid work. But his major work has been on the moving train.