The Trp repressor protein can bind to the operator of the Trp operon, which contains the tryptophan biosynthetic genes. When tryptophan is in abundance, it binds to the repressor and induces a change so that the repressor can bind to DNA. When tryptophan levels are low, the tryptophan falls off the repressor, and the repressor goes back to its original conformation, losing its ability to bind to the DNA. The operator is now free for RNA polymerase and transcription proceeds, making tryptophan biosynthetic genes and replenishing the cell's supply of tryptophan.
This kind of feedback inhibition of transcription is very common. Ribosomal RNA's can also act to repress their own synthesis.
Met-Thr-Arg-Val-Gln-Phe-Lys-His-His-His-His-His-His-His-Pro-Asp-When this sequence begins transcription, the RNA comes of the DNA and ribosomes hop onto it to start translation. However, if there is little histidine in the cell, the ribosome stalls because there are no aminoacyl tRNA's that are charged with histidine. This leaves a long stretch of RNA (for RNA ploymerase is still transcribing it) with no ribosomes bound to it. The sequence of this RNA allows it to form a terminator loop only when ribosomes are bound to it, at which point the RNA is cleaved and the RNA polymerase stops transcribing the genes. Thus, the terminator only functions when the ribosome is not stalled; that is, when there is already plenty of histidine in the cell. The site at which the potential terminator loop forms is called the attenuation site.
Note that many amino acid synthetic operons are also controlled by some form of attenuation. The tryptophan operon has attenuation control as well as the repressor control described above.
There exist two competing proteins in the lambda bacteriophage. One protein, CI, promotes the lysogenic cycle. The other protein, Cro, promotes the lytic phase. These two proteins are in direct competition to each other for sites on the "right" promoter of lambda:
For an image of cro binding to the DNA site, students at MIT can type the following at the athena prompt:
When Vibrio fischeri is not in the squid's light organ, it does not need to be making luciferase, since glowing will not help it or anything else. On the other hand, when inside the light organ, it is to the bacteria's advantage to glow, because then the squid will not get eaten and will feed it, away from competition from any other kinds of bacteria. So how can the bacteria know that it is in a light organ in order to turn on synthesis of luciferase?
The answer lies in the concept of quorum sensing. When the density of bacteria is very high, the bacteria know that they must be inside a light organ instead of floating around in the ocean. Each bacterium is continuously secreting a unique small molecule called VAI (Vibrio fischeri autoinducer) that can diffuse readily through the cell membrane. Thus, there is a declining concentration of the small molecule in a growing circumference around the bacterium. When there are many bacteria around, the local concentration of the small molecule will be very high.
The genes for making luciferase are contained in the lux operon. A DNA binding site (luxO) near the lux promoter (luxP) binds a protein called luxR. This protein somehow calls RNA polymerase over when it is bound to the DNA, thus increasing transcription of the DNA and making mre polymerase. Thus, luxR is a transcriptional activator of the lux operon. When the local concentration of VAI is very high, it binds to luxR, enabling it to bind to the operator and turn on transcription. On the other hand, if VAI is low, luxR is in a conformation such that it cannot bind to the operator, and not very much luciferase can be made. In this fashion, the bacteria only make luciferase when there are lots of other bacteria around. LuxR is consistently transcribed at a low level so that there is always some around to affect regulation.
It is important to note that LuxI is the gene that encodes for the enzyme that synthesizes VAI. When a bacterium undergoes the transition from not making luciferase to making luciferase, it needs to have the autoinducer around in order to promote binding of LuxR to the operator. But before the operon is turned on, how can LuxI be made so that there is a continous level of autoinducer being made? The answer is that operons, in general, are never completely turned off. There is always some basal level of transcription going on, but, for example in this case, the uninduced LuxR protein still has a minimal affinity for the DNA binding site so that some DNA can be transcribed to make enough LuxI so that autoinducer is continuously made.
We tend to think of bacteria as single cell organisms living completely independently from each other. This isn't true. bacteria often communicate with other bacteria in the community. For example, what would be the point of making proteins to carry out bacterial conjugation if there were no other bacteria around? Quorum sensing is used to determine whether or not there are enough bacteria around to make it worth it to turn on the machinery for conjugation.
Mark Ptashne A Genetic Switch Blackwell Scientific Publications and Cell Press, 1992 (2nd Ed).