GENOMIC DNA CLONING (LEC 15) 

---

1)  Reduce the amount of individuals to cover the entire genome.
2)   Help obtain DNA fragments containing the entire gene of interest including promoter at
        the 5 'end, transcribed region including all exons and introns, and the 3' ends.  The pieces
        of DNA inserted in the larger vectors can even contain more than one gene.
3) Sometimes used for walking along chromosome to obtained closely linked regions of DNA.

LAMBDA GENOMIC LIBRARIES

First step in genomic library preparation is the isolation of large fragments of chromosomal DNA from
        plants.  Can be difficult because with plant cells because the cell wall forces you to use harsh
        extraction condiditons to break open the cells to release the DNA.  Extraction conditions will
        often shear DNA into small pieces.  Also must be able to stop nucleases in vacuole from
        digesting DNA and must remove polysaccharides that commonly contaminant DNA preparations
        and subsequently interfere with various enzymatic steps used  to manipulate the DNA later on.

Many techniques to isolate large MW DNA pieces.  Usually begin by grinding plant cells into powder
        while frozen to 77K.  Can break cells open without shearing DNA.   If even larger pieces are
        required for use in YACs and BACs,  usually isolate nuclei from gently broken cells.  DNA in the
        nuclear preparations is digested with restriction enzymes without extraction, and the large fragments
        are then purified by a special gel electrophoresis method for big DNA fragments called PULSE-
        FIELD GEL ELECTROPHORESIS (PFGE) or CLAMPED HOMOGENEOUS
        ELECTROPHORESIS FIELDS (CHEF).  Able to separate DNA fragments as large as
        chromosomes!  The large DNA fragments are extracted from the gels by dissolving the agar used
        directly in library construction.

Remember: all DNA preparations from plants will contain DNA from the CHROMOSOMAL,
        CHLOROPLAST  and MITOCHONDRIAL genomes.  Since chloroplast and mitochondrial genomes
        are much smaller than chromosomes, they are easily separated away, especially for AC libraries.

Once large fragments of DNA are isolated, they need to be cut into the right size to be inserted into the
        vector.   Lambda genomic vectors needs to be 15-20 kbp.  Cosmids need 35-40 kbp, BACs like
        ~ 100 kbp (YACs have an upper limit of around 1000 kpb).   Not only need the right sized
        fragments but also need to cut DNA randomly so that the library will contain a collection of
        OVERLAPPING FRAGMENTS.  With such a collection, you insure that ALL regions of the
        chromosome will be present.  Can be done by random shearing of DNA but it is impossible to
        control the size of the pieces.

 Best way is to PARTIALLY DIGEST the DNA with restriction endonucleases.  Will not only cut DNA
        but also generate fragments with STICKY ENDS.  Can't use 6 basepair cutters, even though they
        would seen to be the best (4096 bp average).  Because it is always possible that:
                1)  A region of DNA will not have the restriction site within 20 Kbp.  This digestion would
                        generate fragments  too large for packaging in lambda.
                2)  You could have a lot of the site such that digestion would generate fragments too small for
                        packaging in lambda.
                 3)  You would fail to generate overlapping fragments need to piece sections of the
                        chromosome together.

        Actually use a 4 basepair frequent cutter but only partially digest DNA.  In this way, many
                overlapping clones  will be obtained.  Most commonly used is Sau3A (or sometimes MboI).
                Sau3A and MboI cut DNA at same site (^GATC) but Sau3A is methylation insensitive.
                Under controlled conditions, you digest the DNA with Sau3A for sufficient time such that
                most DNA fragments are in 15-20 Kbp range (determine by gel electrophoresis).  Check for
                correct digestion conditions by analysis of the size of the DNA fragments versus the time of
                digestion.

DNA obtained is the correct size for lambda, comprising a collection of overlapping fragments, and
        have sticky ends for insertion in the lambda vectors.

 PREPARATION OF LAMBDA VECTOR

For preparation of the genomic library, you need large quantities of lambda left and right arms with
        sticky ends compatible with those in the genomic DNA.  Simple procedure using EMBL-based
        lambda vectors.  Have the STUFFER FRAGMENT bracketed by a MULTIPLE CLONING
        SITE (MCS) containing two restriction sites, BamH1 and EcoR1.  BamH1 recognizes
        G^GATCC  and generate the same sticky ends as Sau3A.
                By double digestion, you will generate Lambda with left and right arms with BamH1 sticky
        ends, a stuffer fragment with EcoR1 sticky ends, and a short fragment from the MCS with both
        ends.  By simple precipitation with isopropanol, the short fragment is removed and thus
        preventing the stuffer fragment from ever religating with arms.

PREPARATION OF GENOMIC LIBRARY

To generate the genomic library, excess lambda arms are added to Sau3A-digested genomic DNA.
        Allow sticky ends of  the DNAs to base pair and form concatemers that are ligated together with
        DNA ligase.  Package DNA into lambda capsids just like cDNA libraries.

Product is thousands of recombinant lambda phage each containing unique fragments of  chromosomal
        DNA;  some of them overlap with each other.  Only those lambda with the correct size inserts will
        replicate and produce viable progeny.  As with cDNA libraries, best not to try to amplify.
        Possible to lose fragments of DNA because they replicate poorly in E. coli or encode DNA that is
        toxic to the bacteria.
 

COSMID LIBRARIES

ARTIFICIAL CHROMOSOME LIBRARIES

Most common are BACTERIAL ARTIFICIAL CHROMOSOMES (BACs) and YEAST
        ARTIFICIAL CHROMOSOMES (YACs) containing DNA inserts from 150 kb to 1000 kb.
        The advantages of BACs over YACs include lower levels of chimerism (multiple DNA
        fragments connected together in the same AC) and ease of library generation and insert manipulation.

BAC vectors permit the cloning of DNA of ~150 kbp with an upper limit of ~ 350 kbp.  YACs can easily
        handle 100 to 1000 kbp.  Greatest limitation is usually the size of genomic DNA that can be
        prepared without shearing..

To construct library, the BAC vector is digested with a restriction enzyme (usually a six base cutter) that
        cuts within the LACZ or SACB genes and then dephosphorylated at the 5' ends with alkaline
        phosphatase to prevent self ligation.  High molecular weight DNA is partially digested with the
        same restriction enzyme and DNA >150 kb is size-selected on a CHEF gel.  Vector DNA is then
        mixed with genomic fragments and ligated together with DNA ligase.  DNA is then electroporated
        into E. coli.  bacterial cells with vector plus insert is selected by growth on chloramphenicol and by
        blue/white selection (LACZ) or growth on saccharose (SACB).

BAC libraries  are currently available for Arabidopsis, corn, rice, sorghum, wheat, soybean, tomato,
        and cotton (and maybe potato).

YACs are constructed similar to BACs, except that the YACs are introduced into yeast via
        electroporation and selected for by a double selection system (URA and antibiotic resistance)
        to help maintain the centromere along with the inserted sequence.

Because all of the DNA sequences cloned into ACs are not crtical to the survivial of the bacterial or
        yeast host, they can be easily lost without effect.  (The only sequences needed are those for
        replication and selection).  As a result, the cloned regions are somewhat unstable.  Segments of
        DNA in cosmids, BACs or YACs can be easily lost over time.
 

LIBRARY COMPEXITY

N = ln (1 - P)/ln (1 - I/GS)

        N = number of clones
        P = probability
        I = insert size
        GS = genome size
 

For example:  If you wanted 99% PROBABILITY that a specific DNA sequence is in a library
        and the the size of the inserted genomic DNA fragments is 15,000 bp (lambda),
        40 kbp (Cosmid), 150 kbp (BAC), or 500 kbp (YAC), you would need:

Arabidopsis (GS = 145 Mbp):
            44,000  lambda,     17,000  Cosmids,      4,500  BACs,         1,300  YACs

(For 80% PROBABILITY
            15,600 lambda,       5,800 Cosmids          1,500 BACs             467 YACs

Corn  (GS = 2504 Mbp):
            770,000 lambda:     290,000  Cosmids:      77,000   BACs:    23,000  YACs

Wheat (GS = 16,000 Mbp):
           5,900,000 lambda:    1,800,00 Cosmids:   490,000  BACs:    150,000 YACs

While most researchers want a 99% probability that their gene is in a particular library, a more
        practical probability is usually 80 - 90 %.  Even at the theoretical 99% probability you
        can never be sure that your sequence is there because of problems in cloning the specific
        fragment or its toxicity to the viral, bacterial, or yeast host.
Obviously, the SMALLER the genome, the EASIER it is to prepare genomic libraries and identify
        specific DNA fragments.
 

FURTHER READINGS:

Sambrock, J. et al. (1989) Molecular Cloning: A Laboratory Manual. 2d Edition. Cold Spring
        Harbor Press, Cold Spring Harbor, NY.
Nelson, D.L. and Brownstein, B.H. (1994)  YAC Libraries : A User's Guide. W.H. Freeman and Co.,
        NY
Davis, L.G. et al. (1994)  Basic Methods in Molecular Biology - 2nd Ed. Appleton & Lange,
        Norwalk, CT.
Gibson, S.I. and C. Somerville. 1992. Chromsome walking in Arabidopsis thaliana using yeast artificial
        chromosomes. In: Methods in Arabidopsis Research. (eds. Koncz, C. N.-H. Chua, J. Schell)
        World Scientific, Singapore. pp 119-143.
Clemson Univ. BAC Web Page: http://hubcap.clemson.edu/~schoi/BAC.html
Woo, S.-S., J. Jiang, B.S. Gill, A.H. Paterson, R.A. Wing. 1994. Construction and characterization of a
        bacterial artificial chromosome library of Sorghum bicolor. Nucleic Acid Res. 22:4922-4931.
Zhang, H.-B., Z. Zhao, X. Ding, A. H. Paterson and R.A. Wing. 1995. Preparation of megabase-size
        DNA from plant nuclei. Plant J. 7:175-184