Introduction 

Construction of Large-Insert Bacterial Clone Libraries

Fig. 1. A general procedure for construction of a large-insert, ordered BAC or BIBAC library (from Zhang et al. 1996a).

The ability to clone large (>100 kb) DNA fragments is crucial to modern structural, functional and comparative genomics of large, complex-genome organisms. The first large DNA fragment cloning system was reported by Burke et al. (1987). This system is based on yeast artificial chromosomes (YACs) consisting of two chromosome arms, telomeres, centromere, replication origin, cloning site and transformant and recombinant selection markers. It allowed cloning and maintenance of DNA fragments up to 1,000 kb in yeast, a quantum leap relative to the pre-existing cosmid system (40 - 50 kb). Because of its large DNA fragment cloning capacity, the YAC system was quickly adopted for genomics research of humans and other species. However, several difficulties of YACs, such as their high level of chimerism, insert instability, and difficulty of purifying cloned insert DNA, have limited the utility of YAC libraries.

            To minimize the above problems of the YAC system, three alternative systems using bacteria as the hosts were soon developed. Shizuya et al. (1992) reported large DNA fragment cloning in Escherichia coli using a bacterial artificial chromosome (BAC) system based on the E. coli fertility (F-factor) plasmid. Ioannou et al. (1994) reported similar facts using a P1-derived artificial chromosome (PAC) system, which combined the features of the bacteriophage P1 and the F-factor-based BAC cloning systems. Both BAC and PAC systems are based on single-copy, “artificial chromosome” cloning vectors. Tao and Zhang (1998) discovered that the conventional plasmid-based cloning (PBC) vectors (e.g., plasmids and cosmids) have the same capacity as the BAC and PAC vectors for large DNA fragment cloning in bacteria. They also showed that the stability of large-insert bacterial clones is not contingent on the unique copy of the clones in host cells.  These results have not only uncovered the molecular basis and established the new concept of large DNA fragment cloning in bacteria, but also indicated that many of the plasmid-based vectors previously developed for different biological research needs could be used as vectors for cloning of large DNA fragments in bacteria. BACs, PACs and PBCs are all referred to as large-insert bacterial clones in this manual. If they are amenable to direct plant transformation, they are referred to as BIBACs (see below), otherwise they are simply referred to as BACs. BAC, PAC and conventional plasmid or cosmid vectors all are capable of cloning and stably maintaining DNA fragments over 300 kb inE. coli. While the sizes of these large-insert bacterial clones are somewhat smaller than YACs, they confer several major advantages over YACs such as low levels of chimerism, facility and speed of insert DNA purification, and high stability in the host cells. Therefore, these large-insert bacterial cloning systems have been quickly assumed a central position in genome research.  

            Another essential element of large DNA fragment cloning is preparation of megabase-size DNA of high quality and high yield. The key to megabase-size DNA preparation is protection of the DNA from physical shearing during preparation and subsequent manipulation. To achieve this, the cells of organisms must be embedded into low-melting-point (LMP) agarose matrix in forms of plugs or microbeads. The cells are lysed, and the DNA are purified in the supporting matrix. For preparation of megabase-size DNA from animals, the cells of animals can be directly embedded in LMP agarose plugs or microbeads, and then the cells are lysed and the DNA are purified in the agarose plugs or microbeads. Preparation of high-quality megabase DNA from plants is far more challenging than from animals. This is because plants have cell walls, which can not be directly embedded into LMP agarose as animals cells.  Plant cells are rich in metabolic substances such as phenolic compounds, some of which may interact with   DNA and thus interfere the DNA with restriction enzyme digestibility. To facilitate preparation of megabase-size DNA from plants, Zhang et al. (1995) have developed a simple, efficient procedure for preparation of megabase-size DNA from plant nuclei. In this procedure, nuclei are first isolated from plant tissues by simply grinding the tissues in liquid nitrogen and subsequent centrifugation, and then embedding in LMP agarose matrix. Using this  procedure, large amounts of high-quality megabase-size nuclear DNA can be readily purified. This procedure has been successfully used in preparation of megabase nuclear DNA from a variety of plant species, and thus, has become the most widely used procedure for plant megabase DNA preparation. It is this procedure that has made it possible to rapidly develop large-insert, high-quality bacterial clone libraries from different plant species.

The large DNA cloning systems have contributed to research of large, complex genomes mainly in two aspects. First, they have made it practical to develop ordered DNA libraries from large-genome species. The large inserts of the clones have reduced the number of clones needed for a complete DNA library by a few-fold relative to the preexisting cosmid cloning system, and therefore, the individual clones can be arrayed in microtiter plates. This has, for the first time,  made “DNA clone libraries” into true “libraries”  in which all clones are individually arrayed and encoded. Therefore, ordered DNA libraries are much more convenient to use than unordered libraries for genomics research. Second, large-insert, ordered DNA libraries have made it feasible to carry out chromosome walking in large genomes for positional cloning and genome analysis, and to rebuild genomes from the clones of ordered libraries. The rebuilt genomes or genome physical maps are essential platforms for large-scale genome sequencing and functional analysis, and large-scale gene discovery, cloning, characterization and utilization. 

Construction of large-insert bacterial clone libraries has been significantly improved and advanced technically since the BAC, PAC and PBC systems were established. Figure 1 shows a general procedure for large-insert, ordered bacterial clone library construction. This procedure has been successfully used in the construction of many BAC, PAC and PBC libraries of plants, animals, insects and microorganisms. It consists of four parts: vector preparation, megabase DNA partial digestion and size selection, DNA transformation, and library assembly. For vector preparation, the vector DNA must be highly purified, vector DNA digestion must be appropriate (neither over-digested nor under-digested), and dephosphorylation of the restricted vector DNA must be complete. The techniques for DNA transformation and library assembly are usually consistent as long as competent cells having a transformation efficiency higher than 10¹⁰ cfu/mg are used. The most technically difficult of the procedure is the megabase DNA partial digestion and size selection. To date, all large-insert bacterial clone libraries reported were generated from the restriction enzyme partial digests of megabase DNA. In the early technology for large-insert bacterial clone library development, the partially digested DNA were selected on LMP agarose by pulsed-field gel electrophoresis (PFGE), and the selected DNA fragments were released from the agarose by digestion with agarase or gelase.  The agarose digestion required that the LMP agarose gel must be melted at 68^oC – 70^oC, which damaged the DNA, especially those sequences that are rich in A/T contents, for instance, the dicotyledonous species DNA. As a result, cloning results varied from time to time and from species to species, and the insert sizes of generated clones was far smaller than the sizes of selected fragments from the pulsed-field gel. To overcome such problems, our laboratories have modified the procedure for size selection of partially digested DNA (Tao et al. 2000). In the modified procedure, the partially digested DNA is selected on regular agarose gels, which has greatly increased size selection efficiency. The selected DNA is released from the agarose by electroelution with PFGE, which eliminates the damage to the DNA due to heating at  68^oC – 70^oC. Therefore, the modified procedure allows one to consistently develop bacterial clone libraries with an average insert size of about 150 kb.   

            Large-insert, ordered bacterial clone libraries are considered long-term resources for genomics research and thus, must be maintained permanently, especially when they are used for genome-wide physical mapping and sequencing. Therefore, prior to the library development it isnecessary  to consider the genotype of a species to be used as the DNA source, the types of large-insert, ordered bacterial clone library to be developed, and the strategies to be used to develop the library. It is desirable to use the reference genotype of a species for genomics research as the library DNA source because such a library will have much wider applications. In term of the types of library to be generated,  two options are available: general large-insert, ordered bacterial clone (BAC) libraries or large-insert, ordered binary bacterial clone (BIBAC) libraries. General BAC libraries can be only used as large-insert DNA libraries, whereas BIBAC libraries not only can be used as large-insert DNA libraries, but also can be directly transformed into plants by theAgarobacterium-mediated transformation method. The latter feature of BIBAC libraries greatly enhances the utility of  the libraries in genomics research. Studies have documented that in species with large genomes, such as most of the important crop species, many genes have large-range spans and many traits are controlled by clusters of genes (e.g., disease resistance genes). For cloning and engineering such genes, or gene clusters for plant genetic improvement, it is desirable to develop large-insert, ordered BIBAC libraries that are also suitable for direct plant transformation.

A few BIBAC vectors have been developed, such as BIBAC2 (Hamilton et al. 1996), and pCLD04541 and pSLJ1711 (Jones et al. 1992; Tao and Zhang 1998). BIBAC libraries have been developed for several plant species. Technologies for BIBAC transformation have been established in several plant species. Hamilton et al. (1996,1999) were able to transform an intact 150-kb human DNA fragment cloned in BIBAC2 into tobacco and tomato by the Agrobacterium-mediated transformation method.  The same 150-kb human DNA fragment cloned in BIBAC2 was also transformed into rice (Choi et al. Abstract #P609, Plant & Animal Genome VIIII, 2000). Wuet al. (2000) indicated that the large-insert clones of pCLD04541 are stable in Agrobacterium, which is critical to large-insert clone transformation in plants via Agrobacterium. Liu et al. (1999) transformed an 80-kb Arabidopsis BIBAC and Chang et al. (in preparation) transformed a 120-kb soybean DNA fragment cloned in pCLD04541 into Arabidopsis.  In contrast, the BACs cloned in general BAC vectors, such as pBeloBAC11 (Kim et al. 1996) and pECBAC1 (Frijters et al. 1997), must be subcloned into a plant transformation binary vector (e.g., BIBAC2, pCLD04541 or pSLJ1711) for plant transformation. Unlike the inserts of BACs from dicotyledonous plant species, it is difficult to subclone the entire insert of a monocotyledonous plant BAC into the BIBAC2 vector because there are often several Not I sites in it. The Not I is often used in subcloning of a BAC insert into the BIBAC2 vector. Therefore, development of BIBAC libraries is important for use of the libraries in gene cloning, engineering, molecular breeding and molecular farming. 

All large-insert BAC and BIBAC libraries developed to date were generated from partial digests of megabase DNA with restriction enzymes. However, the distribution of the sites of a restriction enzyme in a genome is uneven. Consequently, the genome regions with a high, or low density of the restriction sites are difficult to clone because too small (<40 kb) or too large (>350 kb) DNA fragments generated by partial digestion are removed by PFGE-based size selection during BAC cloning. Therefore, it is desirable to develop complementary large-insert BAC or BIBAC libraries with different individual restriction enzymes having different nucleotide contents in their restriction sites. Such large-insert BAC or BIBAC libraries are similar to shotgun libraries in genome coverage, whose cloned DNA fragments are randomly derived from the different regions of a genome, and thus, would have truly high genome coverage.  The number of clones from such complementary libraries equivalent to 6 – 8 x haploid genomes would be sufficient for different genome research purposes, including global genome physical mapping and sequencing.

Large-insert, ordered BAC and BIBAC libraries have been developed for many of the important plant and animal species (see http://hbz.tamu.edu). These libraries have been widely used in genomics research of plants and animals. Figure 2 summarize the applications and significance of large-insert BAC and BIBAC libraries for genomics research.

Fig. 2.  Applications and significance of large-insert, ordered BAC and BIBAC libraries for genomics research.

This manual introduces the state-of-the-art techniques and procedures for construction and manipulation of large-insert, ordered bacterial clone libraries. These include preparation of megabase DNA from plants, animals and insects, preparation of BAC and BIBAC vectors, and BAC and BIBAC library construction. Several of the most useful procedures for manipulation and applications of large-insert, ordered BAC and BIBAC libraries are also presented.

Experiment 1   

Preparation of High-quality Megabase Size DNA

Megabase or high molecular weight (HMW) DNA is essential for large DNA fragment cloning and complex genome analysis. Unlike conventional-size DNA preparation, the megabase DNA must be protected from physical shearing during preparation. One of the popularly used approaches for doing so is to isolate protoplasts (plants), cells (animals), or nuclei (plants and animals), and then embed them in low-melting-point (LMP) agarose in the form of plugs or microbeads, followed by cell lysis and DNA purification in the LMP agarose plugs or microbeads. Megabase DNA is manipulated in LMP agarose plugs or microbeads. Since the preparation of megabase DNA from nuclei is simple, results in low contamination with cytoplast DNA, and is economical and applicable for preparation of megabase DNA from a wide variety of plant species, the nuclei method is widely used for preparation of megabase DNA in plants. In this experiment, we will prepare megabase DNA from different organisms using different methods.

1.1  PREPARATION OF MEGABASE-SIZE NUCLEAR DNA FROM PLANTS

I.  Plant Materials

Plant leaves or whole plants of divergent species, including grasses, legumes, vegetables, and trees can be used as materials for preparation of megabase-size DNA by this method. The tissues can be either frozen in liquid nitrogen and stored in a -80oC freezer or kept fresh on ice before use.

II.  Reagents

10 x homogenization buffer (HB) stock: 0.1 M Trizma base, 0.8 M KCl, 0.1 M EDTA, 10 mM spermidine, 10 mM spermine, adjust pH to 9.4-9.5 with NaOH. The stock is stored at 4oC.

1 x HB:  A suitable amount of sucrose is mixed with a suitable volume of 10 x HB stock.  The final concentration of sucrose is 0.5 M and HB stock is 1 x. The resultant 1 x HB is stored at 4oC.  Before use, b-mercaptoethanol is added to 0.15%. 

20% Triton X-l00 1 x HB: Triton X-100 is mixed with 1 x HB without b-mercaptoethanol to 20%. The solution is stored at 4oC. 

Wash Buffer (1 x HB plus 0.5% Triton X-100):  The buffer is prepared by mixing 1 x HB without b-mercaptoethanol with 20% Triton X-l00 in 1 x HB and stored at 4oC. Before use, b-mercaptoethanol is added to 0.15%.

Lysis buffer: 0.5 M EDTA; pH 9.0-9.3, 1% sodium lauryl sarcosine, and 0.1 - 0.5 mg/ml proteinase K. The proteinase K powder is added just before use.

III. Preparation of Intact Nuclei

For homogenization of the plant tissues, two methods can be used.

1A. Homogenization of Frozen Tissues 

a.  Grind about 100 g of the frozen tissue into fine powder in liquid nitrogen with a mortar and pestle and immediately transfer the powder into an ice cold 1000 ml beaker containing 800 - 1000 ml ice-cold 1 x HB plus 0.15% b-mercaptoethanol and 0.5% Triton X-l00. 

b.  Gently swirl the contents with a magnetic stir bar for 10 minutes on ice and filter into six ice-cold 250 ml centrifuge bottles through two layers of cheesecloth and one layer of Miracloth by squeezing with gloved hands.

a.  Wash about 100 g of fresh tissues with tap water, and if necessary, cut into suitable pieces for homogenization with a kitchen blender (Osterizer 10 Speed Blender). 

b. Homogenize 20 g of the tissue each time in 200 ml ice cold 1 x HB plus b-mercaptoethanol in the kitchen blender at speed 4 or "puree" for 20 - 40 seconds. 

c.  Filter the homogenate into an ice cold 250 ml centrifuge bottles as above. 

d.  Repeat steps b and c to complete the remaining tissues. 

e.  Add 5 ml 1 x HB plus 20% Triton X-l00 (the final concentration of Triton X-l00 is 0.5%) to each bottle (200 ml 1 x HB buffer), gently mix the contents, and incubate on ice for 20 minutes.

2.   Pellet the homogenate prepared by either of the above two methods by centrifugation with fixed-angle rotor at 1,800 g at 4oC for 20 minutes.

4.  Gently resuspend the pellet with assistance of a small paintbrush soaked in ice cold wash buffer, combine the resuspended nuclei from all bottles into a 40 ml centrifuge tube and finally, fill the tube with ice cold wash buffer.

    If particulate matter remains in the suspension, filter the resuspended nuclei into the 40 ml centrifuge tube through two layers of miracloth by gravity. 

        Centrifuge the contents at 57 g, 4oC for 2 minutes to remove intact cells and tissue residues and transfer the supernatant fluid into a fresh centrifuge tube. 

 5.  Pellet the nuclei by centrifugation at 1,800 g, 4oC for 15 minutes in a swinging bucket centrifuge.  6. Wash the pellet 1 - 3 additional times by resuspension in wash buffer using a paintbrush followed by centrifugation at 1,800 g, 4oC for 15 minutes. Note that this step is necessary to minimize the contamination of cytoplast organelles in the nuclei.

6. Wash the pellet 1 - 3 additional times by resuspension in wash buffer using a paintbrush followed by centrifugation at 1,800 g, 4oC for 15 minutes. Note that this step is necessary to minimize the contamination of cytoplast organelles in the nuclei.

7. After the final wash, resuspend the pelleted nuclei in a small amount (about 1 ml) of 1 x HB without b-mercaptoethanol, count the nuclei, if possible, under the contrast phase of a microscope, bring to approximately 5 x 107 nuclei/ml with addition of the 1 x HB without b-mercaptoethanol, and store on ice.

The concentration of nuclei can be also estimated empirically, a concentration of nuclei that is just transparent under light is estimated to be 5 – 10 x 10⁷ nuclei/ml.  The concentration of nuclei varies, depending on the genome sizes of different species.  In general, 5 – 10 mg DNA / 100-ml plug or microbeads are suitable for large-insert BAC and BIBAC cloning.

IV.  Embedding the Nuclei in Agarose Microbeads and Plugs

1. Prepare 10 ml of 1% low-melting-point (LMP) agarose in 1 x HB without b-mercaptoethanol and Triton X-100, cool down to 45oC and maintain in a 45oC water bath before use.

2A. Embedding the nuclei in agarose microbeads 

a.  Prewarm 20 ml of light mineral oil in a 50 ml Falcon tube to 45oC in a water bath (about 15 minutes).

b.  Place a 500 ml flask in a 45oC water bath and prewarm for at least 10 minutes.

c.  Pour about 150 ml ice cold 1 x HB without b-mercaptoethanol and Triton X-100 into a 1000 ml beaker and place the beaker in an ice bath on the top of a magnetic stir plate. Vigorously swirl the solution using a magnetic stir bar.

d.  Prewarm the nuclei to 45oC in a water bath (about 5 minutes). 

e.  Mix the prewarmed nuclei suspension with an equal volume of 1% LMP agarose in 1 x HB without b-mercaptoethanol and Triton X-100 kept in a 45oC water bath, pour into the prewarmed 500 ml flask, and add 20 ml of the prewarmed light mineral oil at 45oC.

f.   Cool down the contents to about 37oC (body temperature), shake the contents in the flask vigorously for 2 - 3 seconds and then immediately pour into the ice-cold 1000 ml beaker containing 150 ml of ice cold 1 x HB without b-mercaptoethanol and Triton X-100 that is vigorously swirling with a magnetic stir bar. Continue to swirl the contents for 5 - 10 minutes on ice. This allows for the agarose microbeads to form uniformly in size. 

g.  Harvest the agarose microbeads by centrifugation at 1,200 g, 4oC for 15 - 20 minutes in a swinging bucket centrifuge. 

h.  Discard the supernatant fluid and resuspend the microbeads in 5 - 10 volumes of lysis buffer. 

 2A.  Embedding the nuclei in agarose plugs 

a.  Prewarm the nuclei to 45oC in a water bath (about 5 minutes). 

b.  Mix the prewarmed nuclei with an equal volume of the prewarmed 1% LMP agarose in 1 x HB without b-mercaptoethanol and Triton X-100 using a cut off pipette tip.

c.  Aliquot the mixture into ice cold plug molds on ice with the same Pipette tip, 100 ml per plug. When the agarose is completely solidified, transfer the plugs into 5 - 10 volumes of lysis buffer. 

 3.   Incubate the agarose plugs or the microbeads in the lysis buffer for 24 - 48 hours at 50oC with gentle shaking. 

 4.  Wash the plugs or the beads once in 0.5 M EDTA, pH 9.0-9.3 for one hour at 50oC, once  in 0.05 M EDTA, pH8.0 for one hour on ice, and store in 0.05 M EDTA, pH8.0, at 4oC. The DNA at this step can be stored at 4oC for one year without significant degradation.

Preparation of megabase DNA from animals is much easier and simpler than preparation of megabase DNA from plants. Animal cells can be directly embedded into LMP agarose plugs or microbeads after briefly washing them. Although the nuclei of animal cells can be used, the whole cells are frequently used for megabase DNA preparation. In this experiment, preparation of chicken megebase DNA plugs is used as an example.

I.  Materials

Whole chicken blood is used as the starting materials. The whole blood is collected in heparinized tubes. The cell density of the blood should be approximately 10⁸ cells / ml.

II. Procedure 

1.    Centrifuge the chicken blood at 1,000 g, room temperature for 5 minutes. Discard the supernatant and resuspend the cells in the phosphate-buffered saline (PBS) buffer.  

2.    Spin at 1,000 g, room temperature for 5 minutes. Discard the supernatant, resuspend the cells in the PBS buffer and bring the concentration of cells to 5 – 10 x 10⁷ cells / ml. 

3.     Pre-warm the cell suspensions at 45^oC for 5 minutes in a water bath, and mix with equal volume of 1% LMP agarose in PBS molten and maintained in a 45^oC water bath. Aliquot into ice-cold 100-ml plug molds with a pipette using a cut tip and keep the molds on ice for 15 minutes to allow the plugs to completely solidify.

4.    Transfer the plugs into the lysis buffer (0.5 M EDTA, pH 9.0, 1% lauryl sarcosine, and 0.5 – 1 mg/ml proteinase K) at approximately 1 ml lysis buffer per 100-ml plug.  

5.    Incubate at 50^oC with gently shaking for 24 – 48 hours. 

6.    Wash the plugs once in 0.5 M EDTA, pH 9.0-9.3 for one hour at 50oC, once in 0.05 M EDTA, pH8.0 for one hour on ice, and store in 0.05 M EDTA, pH8.0, at 4oC.  The DNA at this step can be stored at 4oC for one year without significant degradation.

1.3  Preparation of Megabase-size DNA from Insects 

Preparation of megabase DNA from insects is similar to those from plants and animals. Because of difficulty of obtaining a large amount of tissue from some species of insects, plugs are often made in preparation of megabase DNA. In this experiment, mosquito megabase DNA is prepared. 

I.  Materials 

The L1 larvae of mosquito are used as materials. Either fresh or liquid nitrogen-frozen L1 larvae are suitable for preparation of megabase DNA. 

II.  Procedure 

1.   Grind the fresh or frozen L1 larvae to fine powder in liquid nitrogen with mortar and pestle, and transfer the powder into the suspension solution (100 mM NaCl, 200 mM sucrose and 10 mM EDTA, pH 8.0), or add 10 – 20 ml of the suspension solution into the mortar and let it stay at room temperature until the frozen sample and solution just melts. 

2.   Filter the cell suspensions through 1 – 2 layers of miracloth into a centrifuge tube on ice to remove the large pieces of cell debris and spin at 5,000 rpm, 4^oC for 10 minutes to collect the cells.  

3.   Discard the supernatant and resuspend the cells in about 0.5 – 1.0 ml of the suspension solution, depending on the amount of the sample used. From 500 mg fresh weight of mosquito larvae, we resuspend the cell pellet in 0.8 ml of the suspension solution and make about 20 100-ml plugs.  

4.   Pre-warm the cell suspensions at 45^oC for 5 minutes in a water bath, and mix with an equal volume of molten 1% LMP agarose in the suspension buffer maintained in a 45^oC water bath. Aliquot into ice-cold 100-ml plug molds using a cut pipette tip and keep the molds on ice for 15 minutes to allow the plugs to completely solidify.  

5.      Transfer the plugs into the lysis buffer (0.5 M EDTA, pH 9.0, 1% lauryl sarcosine, and 0.5 – 1 mg/ml proteinase K) at approximately 1 ml lysis buffer per 100-ml plug.

6.    Incubate at 50^oC with gently shaking for 24 – 48 hours.

7.   Wash the plugs once in 0.5 M EDTA, pH 9.0-9.3 for one hour at 50oC, once in 0.05 M EDTA, pH8.0 for one hour on ice, and store in 0.05 M EDTA, pH8.0, at 4oC.  The DNA at this step can be stored at 4oC for one year without significant degradation.