The production of a library as just described is sometimes referred to as “shotgun” cloning because the experimenter clones a large sample of fragments and hopes that one of the clones will contain a “hit”—the desired gene. The task then is to find that particular clone, considered next.

A library might contain as many as hundreds of thousands of cloned DNA fragments. This huge collection of fragments must be screened to find the recombinant DNA molecule containing the gene of interest to a researcher. Such screening is accomplished by using a specific probe that will find and mark only the desired clone. There are two types of probes: (1) those that recognize a specific nucleic acid sequence and (2) those that recognize a specific protein.

368

Probes for finding DNA Probing for DNA makes use of the power of base complementarity. Two single-stranded nucleic acids with full or partial complementary base sequence will “find” each other in solution by random collision. After being united, the double-stranded hybrid so formed is stable. This approach provides a powerful means of finding specific sequences of interest. Probing for DNA requires that all molecules be made single stranded by heating. A single-stranded probe labeled radioactively or chemically is sent out to find its complementary target sequence in a population of DNAs such as a library. Probes as small as 15 to 20 base pairs will hybridize to specific complementary sequences within much larger cloned DNAs.

The identification of a specific clone in a library is a multistep procedure. In Figure 10-12, these steps are shown for a library cloned into a fosmid vector. The steps are similar for libraries of plasmids or BACs. For libraries of phages, plaques are screened rather than colonies. First, colonies of the library on a petri dish are transferred to an absorbent membrane by simply laying the membrane on the surface of the medium. The membrane is peeled off, colonies clinging to the surface are lysed in place on the membrane, and the DNA is simultaneously denatured so that it is single-stranded. Second, the membrane is bathed with a solution of a single-stranded probe that is specific for the DNA sequence being sought. Generally, the probe is itself a cloned piece of DNA that has a sequence that is complementary to that of the desired gene. The probe must be labeled with either a radioactive isotope or a fluorescent dye. Thus, the position of a concentrated radioactive or fluorescent label will indicate the position of the positive clone. For radioactive probes, the membrane is placed on a piece of X-ray film, and the decay of the radioisotope produces subatomic particles that “expose” the film, producing a dark spot on the film adjacent to the location of the radioisotope concentration. Such an exposed film is called an autoradiogram. If a fluorescent dye is used as a label, the membrane is exposed to the correct wavelength of light to activate the dye’s fluorescence, and a photograph is taken of the membrane to record the location of the fluorescing dye.

Where does the DNA to make a probe come from? The DNA can come from one of several sources.

369

Probes for finding proteins If the protein product of a gene is known and isolated in pure form, then this protein can be used to detect the clone of the corresponding gene in a library. The process, described in Figure 10-13, requires two components. First, it requires an expression library, made by using expression vectors that will direct the host cell to produce the protein. To make the library, cDNA is inserted into the vector in the correct triplet reading frame with a bacterial promoter, and cells containing the vector and its insert produce a translation of the cDNA insert. Second, the process requires an antibody that binds to the specific protein product of the gene of interest. (An antibody is a protein made by an animal’s immune system that binds with high affinity to a given molecule.) The antibody is used to screen the expression library for that protein. A membrane is lain over the surface of the medium and removed so that some of the cells of each colony are now attached to the membrane at locations that correspond to their positions on the original petri dish (see Figure 10-13). The imprinted membrane is then dried and bathed in a solution of the antibody, which will bind to the imprint of any colony that contains the fusion protein of interest. Positive clones are revealed by a labeled secondary antibody that binds to the first antibody. By detecting the correct protein, the antibody identifies the clone containing the gene that must have synthesized that protein and therefore contains the desired cDNA.

We can see how this type of probe works in practice by returning to the human insulin example. To clone a cDNA corresponding to human insulin, we first synthesize cDNA using mRNA isolated from pancreas cells as the template. The cDNA molecules are then inserted into a bacterial expression vector and the vector is transformed into bacteria. Bacterial colonies containing insulin cDNA will express insulin protein. The insulin protein is identified by its binding with an insulin antibody as described above.

In many cases, we don’t have a probe for the gene to start with, but we do have a recessive mutation in the gene of interest. This gene could be a mutant gene in a bacterium or yeast or even a plant or mouse. The goal of this approach is to identify the clone containing the gene of interest by the fact that it will restore the function eliminated by the recessive mutation. In practice, one first generates a genomic or cDNA library from an organism that has the wild-type allele of the gene of interest. The gene of interest is one of thousands represented in the library. However, only the gene of interest has the ability to complement the mutant organism and restore the wild-type phenotype. Thus, if we are able to introduce the library into the species bearing the recessive mutation (see Section 10.6), we can detect specific clones in the library through their ability to restore the function eliminated by the recessive mutation. This procedure is called functional complementation or mutant rescue. The general outline of the procedure is as follows:

370

So far, we have described techniques to transform only bacterial cells. You will see later in this chapter that DNA can be introduced into many genetic model organisms, including Saccharomyces cerevisiae (yeast), Caenorhabditis elegans (nematode worm), Arabidopsis thaliana (plant), and Mus musculus (mouse).

371

After you have amplified your PCR product or selected a clone of interest from a genomic or cDNA library, the next step is to find out more about the DNA. Let’s say that you have recovered the insulin cDNA from an expression vector and want to determine the restriction sites in the genomic copy of the insulin gene. Perhaps you want to see whether these sites differ among diverse human populations. You might also want to know whether the size of the insulin mRNA varies among human populations. Alternatively, you might want to determine whether a similar gene is present in the genome of a related organism. In the section below, you will see that these important questions can be answered by using relatively simple techniques. In these techniques, complex mixtures of DNA or RNA are sorted by size and then probed by hybridization to detect DNA molecules related to some other DNA molecule.

The most extensively used method for detecting a molecule within a mixture is blotting, which starts with gel electrophoresis to separate the molecules in the mixture. A mixture of linear DNA molecules is placed into a well formed in an agarose gel. The gel is oriented in a box with electrodes at either end so that the wells are at the cathode end (negatively charged) and the DNA, because of its negative charge, migrates to the anode end (positively charged). The speed of migration of DNA molecules in the gel is inversely dependent on their size because the agarose acts as a sieve through which small molecules move more easily and quickly than larger fragments (Figure 10-14). Therefore, the fragments in distinct size classes will form distinct bands on the gel. The bands can be visualized by staining the DNA with ethidium bromide, which causes the DNA to fluoresce in ultraviolet light. The absolute size of each fragment in the mixture can be determined by comparing its migration distance with a set of standard fragments of known sizes. If the bands are well separated, an individual band can be cut from the gel, and the DNA sample can be purified from the gel matrix. Therefore, DNA electrophoresis can be either diagnostic (showing sizes and relative amounts of the DNA fragments present) or preparative (useful in isolating specific DNA fragments).

372

Genomic DNA digested by restriction enzymes generally yields so many fragments that electrophoresis produces a continuous smear of DNA and no discrete bands. A probe can identify one fragment in this mixture, with the use of a technique developed by E. M. Southern called Southern blotting (Figure 10-15a). Like clone identification (see Figure 10-12), this technique entails getting an imprint of DNA molecules on a membrane by using the membrane to blot the gel after electrophoresis is complete. The DNA must be denatured first, which allows it to stick to the membrane. Then the membrane is hybridized with a labeled probe. An autoradiogram or a photograph of fluorescent bands will reveal the presence of any bands on the gel that are complementary to the probe. To detect the insulin gene, we can apply this protocol to human genomic DNA digested with restriction enzymes on the membrane, using insulin cDNA as the labeled probe.

The Southern-blotting technique can be modified to detect a specific RNA molecule from a mixture of RNAs fractionated on a gel. This technique is called RNA blotting or, more commonly, Northern blotting (thanks to some scientist’s sense of humor) to contrast it with the Southern-blotting technique used for DNA analysis. The RNA separated by electrophoresis can be a sample of the total RNA isolated from a tissue or from an entire organism. In the example shown in Figure 10-15b, the gel was run with RNA isolated from the seeds of various plants. Unlike DNA that is loaded onto a gel, there is no need to digest the RNA sample as it is produced in discrete transcript-size molecules. RNA gels are blotted onto a membrane and probed in the same way as DNA is blotted and probed for Southern blotting. One application of Northern analysis is to determine whether a specific gene is transcribed in a certain tissue or under certain environmental conditions. Another is to determine the size of the mRNA and whether an RNA of similar size can be detected in closely related plants (as in Figure 10-15b).

We started this section on blot analysis by posing questions about the insulin gene and its mRNA in human populations and in related species. Based on the techniques above, can you design Southern- and Northern-blot experiments to answer these questions? You can assume that you have access to samples of the required genomic DNAs and RNAs.

Hence, we see that cloned DNA finds widespread application as a probe used for detecting a specific clone, a DNA fragment, or an RNA molecule. In all these cases, note that the technique again exploits the ability of nucleic acids with complementary nucleotide sequences to find and bind to each other.

Probing for a specific protein Probing for proteins is generally performed by using antibodies as probes. An antibody is a protein made by an animal’s immune system; it binds with high affinity to a molecule such as a specific protein (which acts as an antigen) because the antibody has a specific lock-and-key fit with it. For protein detection, a protein mixture extracted from cells is separated into bands of distinct proteins by electrophoresis and then blotted onto a membrane (this is a Western blot). The position of a specific protein of interest on the membrane is revealed by bathing the membrane in a solution of antibody obtained from a rabbit or other host into which the antigen has been previously injected. The position of the protein is revealed by the position of the label that the antibody carries.

373

374