The foregoing descriptions are generic approaches to creating recombinant DNA. However, a geneticist is interested in isolating and characterizing some particular gene of interest, so the procedures must be tailored to isolate a specific recombinant DNA clone that will contain that particular gene. The details of the process differ from organism to organism and from gene to gene. An important initial factor is the choice of an appropriate vector for the job at hand.
Vectors must be relatively small molecules for convenience of manipulation. They must be capable of prolific replication in a living cell, thereby enabling the amplification of the inserted donor fragment. Another important requirement is that there be convenient restriction sites that can be used for insertion of the DNA to be cloned. Unique sites are most useful because then the insert can be targeted to one site in the vector. It is also important that there be a mechanism for easy identification and recovery of the recombinant molecule. There are numerous cloning vectors in current use, and the choice between them often depends on the size of the DNA segment that needs to be cloned. We will consider several commonly used types.
As described earlier, bacterial plasmids are small circular DNA molecules that are distinct from, as well as additional to, the main bacterial chromosome. They replicate their DNA independently of the bacterial chromosome. Many different types of plasmids have been found in bacteria. The distribution of any one plasmid within a species is generally sporadic; some cells have the plasmid, whereas others do not. In Chapter 9, we encountered the F plasmid, which confers certain types of conjugative behavior to cells of E. coli. The F plasmid can be used as a vector for carrying large donor DNA inserts, as we shall see in Chapter 12. However, the plasmids that are routinely used as vectors are those that carry genes for drug resistance. The drug-resistance genes are useful because the drug-resistant phenotype can be used to select not only for cells transformed by plasmids, but also for vectors containing recombinant DNA. Plasmids are also an efficient means of amplifying cloned DNA because there are many copies per cell, as many as several hundred for some plasmids.
Two plasmids designed as vectors for DNA cloning, showing general structure and restriction sites. Insertion into pBR322 is detected by inactivation of one drug-resistance gene (tetR), indicated by the tetS (sensitive) phenotype. Insertion into pUC18 is detected by inactivation of the β-galactosidase function of Z′, resulting in an inability to convert the artificial substrate X-Gal into a blue dye.
on the following
page. These vectors are derived from natural plasmids, but both have been
genetically modified for convenient use as recombinant DNA vectors. Plasmid
pBR322 is simpler in structure; it has two drug-resistance genes,
tetR and
ampR. Both genes contain unique restriction
target sites that are useful in cloning. For example, donor DNA could be
inserted into the tetR gene. A successful
insertion will split and inactivate the tetR
gene, which then will no longer confer tetracycline resistance, and the cell
will be sensitive to that drug. Therefore, the cloning procedure would be to
mix the samples of cut plasmid and donor DNA, transform bacteria, and select
first for ampicillin-resistant colonies, which must have been successfully
transformed by a plasmid molecule. Of the AmpR colonies, only
those that prove to be tetracycline-sensitive have inserts; in other words,
the AmpR TetS colonies are the ones that contain
recombinant DNA. Further experiments are needed to find the clones with the
specific insert required.
The pUC plasmid is a more advanced vector, whose structure allows direct visual selection of colonies containing vectors with donor DNA inserts. The key element is a small part of the E. coli β-galactosidase gene. Into this region has been inserted a piece of DNA called a polylinker, or multiple cloning site, which contains many unique restriction target sites useful for inserting donor fragments. The polylinker is in frame translationally with the β-galactosidase fragment and does not interfere with its translation. The transformation protocol uses recipient cells that contain a β-galactosidase gene lacking the fragment present on the plasmid. An unusual type of complementation occurs in which the partial proteins coded by the two fragments unite to form a functional β-galactosidase. A colorless substrate for β-galactosidase called X-Gal is added to the medium, and the functional enzyme converts this substrate into a blue dye, which colors the colony blue. If donor DNA is inserted into the polylinker, the enzyme fragment borne on the vector is disrupted, no complete β-galactosidase protein is formed, and the colony is white. Hence, selection for white AmpR colonies selects directly for vectors bearing inserts, and such colonies are isolated for further study.
Plasmids that contain large inserts of foreign DNA tend to spontaneously lose the insert; therefore, plasmids are not useful for cloning DNA fragments larger than 20 kb.
Cloning in phage λ. A nonessential central region of the phage chromosome is discarded and the ends ligated to random 15-kb fragments of donor DNA. A linear multimer (concatenate) forms, which is then stuffed into phage heads one monomer at a time by using an in vitro packaging system. (From J. D. Watson, M. Gilman, J. Witkowski, and M. Zoller, Recombinant DNA, 2d ed. Copyright © 1992 by Scientific American Books.)
). A
second useful property of a phage vector is that recombinant molecules are
automatically packaged into infective phage particles, which can be
conveniently stored and handled experimentally.
Cloning by cosmids. The cosmid is cut at a BglII site next to the cos site. Donor genomic DNA is cut by using Sau3A, which gives sticky ends compatible with BglII. A tandem array of donor and vector DNA results from mixing. Phage is packaged in vitro by cutting at the cos site. The cosmid with insert recircularizes after it is in the bacterial cell. (From J. D. Watson, M. Gilman, J. Witkowski, and M. Zoller, Recombinant DNA, 2d ed. Copyright © 1992 by Scientific American Books.)
.
Some phages contain only single-stranded DNA molecules. On infection of bacteria, the single infecting strand is converted into a double-stranded replicative form, which can be isolated and used for cloning. The advantage of using these phages as cloning vectors is that single-stranded DNA is the very substrate required for the Sanger method DNA sequencing technique currently in widespread use (page 324). Phage M13 is the one most widely used for this purpose.
One way of detecting a specific cloned gene is by detecting its protein product expressed in the bacterial cell. Therefore, in these cases, it is necessary to be able to express the gene in bacteria; that is, to transcribe it and translate the mRNA into protein. Most cloning vectors do not permit expression of cloned genes, but such expression is possible if special vectors are used. However, because bacteria cannot process introns, the cloned sequences must be stripped of introns. The cloned gene is inserted next to appropriate bacterial transcription and translation start signals. Some expression vectors have been designed with restriction sites located just next to a lac regulatory region. These restriction sites permit foreign DNA to be spliced into the vector for expression under the control of the lac regulatory system.
We have learned that the most important goal of recombinant DNA technology is to clone a particular gene or other genomic fragment of interest to the researcher. The approach used to clone a specific gene depends to a large degree on the gene in question and on what is known about it. Generally, the procedures start with a sample of DNA such as eukaryotic genomic DNA. The next step is to obtain a large collection of clones made from this original DNA sample. The collection of clones is called a DNA library. This step 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.
There are different types of libraries, categorized, first, according to which vector is used and, second, according to the source of DNA. Different cloning vectors carry different amounts of DNA, so the choice of vector for library construction depends on the size of the genome (or other DNA sample) being made into the library. Plasmid and phage vectors carry small amounts of DNA, so these vectors are suitable for cloning genes from organisms with small genomes. Cosmids carry larger amounts of DNA, and other vectors such as YACs and BACs (see Chapter 12) carry the largest amounts of all. Ease of manipulation is another important factor in choosing a vector. A phage library is a suspension of phages. A plasmid or a cosmid library is a suspension of bacteria or a set of defined bacterial cultures stored in culture tubes or microtiter dishes.
The synthesis of double-stranded cDNA from mRNA. A short oligo(dT) chain is hybridized to the poly(A) tail of an mRNA strand. The oligo(dT) segment serves as a primer for the action of reverse transcriptase, which uses the mRNA as a template for the synthesis of a complementary DNA strand. The resulting cDNA ends in a hairpin loop. When the mRNA strand has been degraded by treatment with NaOH, the hairpin loop becomes a primer for DNA polymerase I, which completes the paired DNA strand. The loop is then cleaved by S1 nuclease (which acts only on the single-stranded loop) to produce a double-stranded cDNA molecule. (From J. D. Watson, J. Tooze, and D. T. Kurtz, Recombinant DNA: A Short Course. Copyright © 1983 by W. H. Freeman and Company.)
). Because it is
made from mRNA, cDNA is devoid of both upstream and downstream regulatory
sequences and of introns. This means that cDNA from eukaryotes can be translated
into functional protein in bacteria—an important feature when expressing
eukaryotic genes in bacterial hosts.
The choice between genomic DNA and cDNA depends on the situation. If a specific gene that is active in a specific type of tissue in a plant or animal is being sought, then it makes sense to use that tissue to prepare mRNA to be converted into cDNA and then make a cDNA library from that sample. This library should be enriched for the gene in question. A cDNA library is based on the regions of the genome transcribed, so it will inevitably be smaller than a complete genomic library, which should contain all of the genome. Although genomic libraries are bigger, they do have the benefit of containing genes in their native form, including introns and regulatory sequences. If the purpose of constructing the library is a prelude to cloning an entire genome, then a genomic library is necessary at some stage.
In some cases, it is possible to narrow down the genomic fraction used in library construction, to more easily detect the desired gene. This approach is possible if the experimenter already knows which chromosome contains the gene. One technique used in mammalian molecular genetics is to sort the chromosomes with an instrument called a flow cytometer. A suspension of chromosomes is passed through the apparatus, which sorts the chromosomes according to size (this procedure is discussed in more detail in Chapter 12). The appropriate chromosomal fraction is then used to make the library.
Another technique possible in organisms with small chromosomes is to fractionate whole chromosomes by using pulsed field gel electrophoresis (PFGE). Electrophoresis is a general technique that fractionates nucleic acids or proteins according to size on gels under the influence of a strong electric field. This type of procedure separates shorter DNA fragments. PFGE is a specialized type of electrophoresis useful for very long DNA molecules. It uses several oscillating electric fields oriented in several different directions. This enables large DNA molecules such as whole chromosomes to snake through the gel to different positions according to their size. The appropriate chromosome can be identified on the gel by probing with a chromosome-specific probe (see the next subsection). Then the desired chromosome can be cut out, eluted from the gel, and used to make a chromosome-specific library.
How can an experimenter determine whether a library is large enough to contain any one unique sequence of interest with a reasonable degree of certainty? There are formulas for calculating the minimum number of clones needed, but a rough idea of the general order of magnitude of the library can be obtained simply by taking the total genome size and dividing by the average size of the inserts carried by the vector being used. Generally, this number will be at least doubled, but it does provide a rough estimate of the magnitude of the job of library construction.
The library, which might contain as many as hundreds of thousands of cloned fragments, must be screened to find the recombinant DNA molecule containing the gene of interest. Such screening is accomplished by using a specific probe that will find and mark the clone for the researcher to identify. Broadly speaking, there are two types of probes: those that recognize DNA and those that recognize protein.
These probes depend on the natural tendency of a single strand of nucleic acid to find and hybridize to another single strand with a complementary base sequence. A probe that is itself DNA, when denatured (made single-stranded by unwinding the two halves of the double helix), will therefore find and bind to other similar denatured DNAs in the library.
Facing page: (a) A genomic library can be made by cloning genes in λ bacteriophages. When a lawn of bacteria on a petri plate is infected by a large number of different hybrid phages, each plaque in the lawn is inhabited by a single clone of phages descended from the original infecting phage. Each clone carries a different fragment of cellular DNA. The problem now is to identify the clone carrying a particular gene of interest (dark blue) by probing the clones with DNA or RNA known to be related to the desired gene. (b) The plaque pattern is transferred to a nitrocellulose filter, and the phage protein is dissolved, leaving the recombinant DNA, which is then denatured so that it will stick to the filter. The filter is incubated with a radioactively labeled probe; that is, a DNA copy of the messenger RNA representing the desired gene. The probe hybridizes with any recombinant DNA incorporating a matching DNA sequence, and the position of the clone having the DNA is revealed by autoradiography. Now the desired clone can be selected from the culture medium and transferred to a fresh bacterial host, so that a pure gene can be manufactured. (After R. A. Weinberg, “A Molecular Basis of Cancer,” and P. Leder, “The Genetics of Antibody Diversity.” Copyright © 1983, 1982 by Scientific American, Inc. All rights reserved.)
). First, colonies
or plaques of the library on a petri plate are transferred to an absorbent
membrane (often nitrocellulose) by simply laying the membrane on the surface
of the medium. The membrane is peeled off, and colonies or plaques clinging
to the surface are lysed in situ and the DNA denatured. The next step is to
bathe the membrane with a solution of a probe that is specific for the DNA
being sought. The probe must be labeled either with radioactivity or a
fluorescent dye. Generally, the probe is itself a cloned piece of DNA that
has a sequence homologous to the desired gene. The probe DNA must be
denatured; it will then bind only to the DNA of the clone being sought. The
position of a positive clone will become clear from the position of the
concentrated label, often as a spot on an autoradiogram.
Where does the DNA to make a probe come from? The DNA can be from one of several sources. One source is cDNA from tissue that expresses the gene of interest. The idea is that, because the mRNA of a gene is abundant, many of the cDNAs made from this tissue and inserted individually into vectors will very likely be for the desired gene. For example, in mammalian reticulocytes, 90 percent of the mRNA is known to be transcribed from the β-globin gene, so reticulocytes would be a good source of mRNA for making a cDNA probe to find a genomic globin gene. In this case, a genomic library would be probed. The need for this kind of analysis depends on which questions are to be asked about the gene. If only the transcribed sequence is of interest, then the cDNA clone itself could provide that information just as well. However, if introns and control regions are needed, the genomic clone must be obtained.
Another source of DNA for a probe might be a homologous gene from a related organism. For example, if a certain gene has been cloned in the ascomycete fungus Neurospora, then it is very likely that this gene can be used as a probe to find the homologous gene in the related fungus Podospora. This method depends on the evolutionary conservation of DNA sequences through time. Even though the probe DNA and the DNA of the desired clone might not be identical, they are often similar enough to promote hybridization. The method is jokingly called “clone by phone” because, if you can phone a colleague who has a clone of your gene of interest but from a related organism, then your job of cloning is made relatively easy.
A short sequence of a protein is used to design a set of redundant oligonucleotides for use as a probe to recover the gene that encoded the protein. One of the set of probes will be a perfect match for the gene. (From H. Lodish, D. Baltimore, A. Berk, S. L. Zipursky, P. Matsudaira, and J. Darnell, Molecular Cell Biology, 3d ed. Copyright © 1995 by Scientific American Books, Inc.)
shows an example in which there are five
positions of redundancy, showing 2, 3, 2, 2, and 2 alternatives,
respectively. The reaction would make 2 × 3 × 2 × 2 × 2 = 48 oligonucleotide strands at the
same time. This “cocktail” of oligonucleotides would be used as a probe. The
correct strand within this cocktail would find the gene of interest. Twenty
nucleotides embody enough specificity to find one unique DNA sequence in the
library.
Additionally, free RNA can be radioactively labeled and used as a probe. This is possible only when a relatively pure population of identical molecules of RNA can be isolated, such as rRNA or fractionated tRNAs.
Finding the clone of interest by using antibody. An expression library made with phage derivative λgt11 is screened with a protein-specific antibody. After the un-bound antibodies have been washed off the filter, the bound antibodies are visualized through binding of a radioactive secondary antibody. (From J. D. Watson, M. Gilman, J. Witkowski, and M. Zoller, Recombinant DNA, 2d ed. Copyright © 1992 by Scientific American Books.)
. An antibody to the protein is prepared, and this antibody
is used to screen an expression library. These libraries are made by using
expression vectors designed to express high levels of a specific bacterial
protein. To make the library, cDNA is inserted into the vector in frame with
the bacterial protein, and the cells will make a fusion protein. A membrane
is laid over the surface of the medium and removed with an imprint of
colonies. It is dried and bathed in a solution of the antibody. Positive
clones are revealed by making an antibody to the first antibody; the second
antibody is labeled by a radioactive isotope or a chemical that will
fluoresce or become a colored dye. By detecting the correct protein, the
antibody effectively identifies the clone containing the gene that must have
synthesized that protein.
At the beginning of the chapter, we asked how it might be possible to find the gene for human albinism. It was in fact cloned by using an antibody to the enzyme that is known to be defective in this condition, the enzyme tyrosinase. This enzyme, like any protein, can be purified by standard biochemical procedures, and subsequently an antibody to the enzyme was prepared in rabbits. From tyrosinase-producing cells, mRNA was isolated and used to make cDNA. This cDNA was used to make an expression vector library. The library was probed with the antibody to tyrosinase, and several positive clones were detected. The cDNA in the positive clones was sequenced and found to contain a gene whose exons total 1590 nucleotide pairs. The cDNA was used to probe a library of human genomic DNA, and, in this process, the intact tyrosinase gene was found. It proved to have five exons and four introns.
Specific clones in a bacterial or phage library can be detected through their ability to confer a missing function on a mutant line of the donor organism, which acts as the transformation recipient. This procedure is called functional complementation. Here the protocol is:
This method depends on the ability to transform the donor organism, often a eukaryote. We have already considered transformation in prokaryotes (Chapters 2 and 9), but eukaryotes can be transformed, too. The procedure differs among eukaryotes, but generally some special treatment of recipient cells is required. For example, to transform fungi, generally the cell walls must be removed enzymatically. Let’s assume that we have isolated a mutant that is relevant to some biological process that interests us. For the present purpose, we will assume that it is an auxotrophic mutation in a fungus. We shall use DNA from the library to transform the auxotrophic mutant strain and then plate these recipient fungal cells on minimal medium. Fungal cells that contain the wild-type allele (from the wild-type culture used to make the library) will transform the auxotroph to prototrophy and allow growth on minimal medium.
The reason that this transformation method works is that the transforming fragment functionally complements the deficiency caused by the mutant allele in the recipient. It might seem at first that this view of complementation is not the same as the one developed in Chapter 6; that is, the production of a wild-type phenotype from the union of two mutant genomes. However, the transforming vector contributes something that the recipient genome lacks (the wild-type allele being sought), and the recipient genome contributes something that the vector lacks (the entire remainder of the genome), so a type of complementation is involved.
Finding a cloned gene by using progressively smaller pooled DNA samples in transformation. In this example, the quest is for the trp3 gene of Neurospora (After J. R. S. Fincham, Genetic Analysis. Copyright © 1995 by Blackwell.)
, using as an example the
Neurospora trp3 gene discussed at the beginning of the
chapter. In this case, a cosmid library was used. The cosmid must also carry a
marker gene that can be used to select for successful transformants of the
fungus. A gene for hygromycin resistance is commonly used in fungi, which are
normally sensitive to this drug. Subsets of the cosmid library made from
wild-type Neurospora DNA were used to transform
trp3 mutant cells, and trp3+
clones were selected by plating transformed cells on medium containing
hygromycin but lacking tryptophan. Colonies that grow are likely to contain the
trp3+ allele and are isolated from the
plate.
The possible fates of transforming DNA. A donor wild-type allele A+ (cloned in a bacterial vector) transforms an A− recipient by one of three different types of insertion. note: The recipient is generally of the same species as the donor DNA, either prokaryotic or eukaryotic. Two recipient chromosomes are shown, I and II.
). Less commonly, the transforming wild-type allele replaces the
resident auxotrophic mutation by a double-crossover-like process.
If a eukaryotic gene is cloned on a prokaryotic vector but a specific eukaryotic sequence is known that can act as an origin of replication, this sequence can be added to the vector. Then the vector will be able to replicate in both bacterial and eukaryotic cells, and insertion into the chromosome is not essential. These types of vectors are called shuttle vectors because they can be moved back and forth between different hosts. Without an origin of replication, the donor DNA must integrate into the eukaryotic chromosome to effect stable transformation.
Specific cloned donor genes can be selected by using their DNA to transform and complement null alleles in recipient cells of the donor organism.
Chromosome walking. One recombinant phage obtained from a phage library made by the partial EcoRI digest of a eukaryotic genome can be used to isolate another recombinant phage containing a neighboring segment of eukaryotic DNA, as described in the text. (From J. D. Watson, J. Tooze, and D. T. Kurtz, Recombinant DNA: A Short Course. Copyright © 1983 by W. H. Freeman and Company.)
summarizes the procedure of chromosome walking. End fragments
of a clone of the linked marker are used as probes to select other clones from
the library. These probes will detect clones of DNA regions that overlap with
the initial clone. Restriction maps (pages 327–329) are made of the DNA of this
second set of clones, and, again, outward fragments are used for a new round of
selection of overlapping clones from the library. Hence the walking process
moves outward in two directions from the start site. Each clone can be sequenced
or otherwise tested, depending on the intent of the exploration.
Sometimes a large insert that is known to contain the linked marker will also luckily contain the sought gene, and subcloning and transformation will narrow down the appropriate region of the cosmid. The availability of a large number of neutral DNA markers (restriction fragment length polymorphisms) dispersed throughout most genomes has provided many useful start points. Positional cloning has been particularly useful for cloning human genes, many of which have no known biochemical function and cannot be easily selected by functional complementation. The human gene for cystic fibrosis, mentioned at the beginning of the chapter, was cloned by chromosome walking, and we shall examine its cloning in more detail in Chapter 12. For any case of chromosome walking, there must be some type of criterion to assess each step of the walk for the gene of interest, and these criteria depend on the individual gene concerned.
Using DNA insertion as a tag for marking and recovering a gene from the genome. The tag DNA can be transforming DNA or an endogenous transposon (movable element).
. One type of tag is transforming DNA. When exogenous DNA is
added by transformation or by other methods such as injection, it can integrate
into the genome and become part of the chromosome. Ectopic integration is random
throughout the genome, and apparently no segment of chromosomal DNA is immune to
integration. When integration takes place within or near a gene, the integrating
fragment acts as a mutagen, disrupting the function of the gene. This property
can be used to good advantage. Suppose that we use a specific cloned gene
x+ and transform x− cells of the
donor organism into x+. Many of the x+ transformants will
be mutant for the genes into which the transforming DNA has inserted
ectopically. A subset of such x+ cells will be mutant for the target
gene a+, the gene of interest, and will be of
phenotype a−. Hence among the x+ transformants,
a− phenotypes are identified. The next step is to cross the
transformants to determine if the a− phenotype
segregates with x+. If it does, the mutation is likely to have been
caused by the integration of the fragment containing
x+. The DNA of this mutant line is used to
construct a library, and gene x+ can be used as a
probe to recover the clone of the disrupted a gene. To recover
the intact wild-type a gene, a fragment of the disrupted
a gene sequence is used in another round of probing, this
time with a wild-type library.
.
Mutating a gene by the insertion of transforming DNA or a transposon allows the gene to be tagged as a prelude to its isolation.
