BACTERIAL ARTIFICIAL CHROMOSOME (BAC) LIBRARIES CONSTRUCTED FROM THE GENETIC STANDARD OF UPLAND COTTONS

JIANMIN DONG¹, Russell J. Kohel¹, Hongbin Zhang², John Yu¹

¹USDA-ARS, Crop Germplasm Research Unit, 2765 F&B Road, College Station, TX 77845, USA; and ²Texas A&M University, Department of Soil and Crop Sciences, 2474 TAMUS, College Station, TX 77843, USA


I. Introduction

Cotton is the leading textile fiber and the second most important oilseed in the world. In the United States, the combined raw-product value of the cotton fiber and cotton-seed oil and meal exceed $5.5 billion annually. Cotton-related annual revenue in the economy of the country exceeds $120 billion. Despite its economic importance, genomics research of cotton lags behind that of maize, soybean, and wheat. Among the problems that have led to this lag in cotton genomics research, the most significant one is the absence of infrastructure and tools. The purposes of this study are to develop essential resources for cotton genomics research, large-insert BAC and plant-transformation-competent binary BAC (BIBAC) libraries that are widely adaptable and publicly available for cotton genomics research.

Genome physical mapping, i.e., reconstruction of genomes from ordered DNA clone libraries, is the centerpiece of structural, functional and comparative genomics research. This is because a genome-wide integrated physical and genetic map will provide revolutionary tools and infrastructure for large-scale, high-throughput EST mapping, target marker development, effective positional cloning of genes and QTLs important to agriculture, determination of structure and organization of every gene, and genome sequencing. Therefore, we are developing a genome-wide, BAC/BIBAC-based, integrated physical and genetic map of the cotton (AD) genome from these large-insert cotton BAC and BIBAC libraries.

II. Materials and Methods

1. Why Did We Develop the Cotton BAC and BIBAC Libraries from Cotton Acc. TM-1?

Several BAC and BIBAC libraries have been previously developed to facilitate cotton genomics research, most of which (five) were developed by our group (see http://algodon.tamu.edu and http://hbz.tamu.edu). The average insert sizes of these BAC and BIBAC libraries range from 93 to 150 kb. We develop large-insert BAC and BIBAC libraries for Gossypium hirsutum acc. TM-1 because these TM-1 libraries are essential for development of a genome-wide BAC/BIBAC-based physical map of the cotton genome:

• The cultivated cotton acc. TM-1 contains A and D genomes. It is a must to develop a genome-wide physical map of cotton based on the AD genome because there is no strong evidence in supporting either A or D genome as predominant in gene distribution. Instead, the genes important to cotton fiber production and quality are likely to distribute randomly between A and D diploid genomes.

• Homogeneous and homozygous plant materials are essential to a successful assembly of BAC contigs by fingerprint analysis according to our previous experience in development of the genome-wide BAC/BAC-based physical maps of rice and Arabidopsis. The homogeneity and homozygosity of the source DNA are also required to truly reflect the genome structure of a species. Among cultivated cottons, no accession meets such requirements better than TM-1. TM-1 is maintained at 54^th selfing generation by single-seed descent to ensure its homogeneity and homozygosity at College Station, Texas.

• TM-1 is the genetic standard for cotton genetic studies. Extensive genetic mutants and cytogenetic stocks have been developed from this cotton line.

• TM-1 is one of the two parents used to derive a permanent population for integrative genetic and physical mapping of the cotton genome.

2. Library Development

The BAC and BIBAC libraries of cotton TM-1 were constructed according to the procedure that we previously developed. Leaves were collected from cotton seedlings. Nuclei were isolated and embedded in low-melting-point agarose gel plugs. Megabase nuclear DNA was purified, partially digested with Bam I, Hind III and Eco RI, respectively and double-size-selected on pulsed-field gels. The DNA fragments ranging from 100 – 400 kb were excised from the gels, electroeluted into solution by pulsed-field electrophoresis, and ligated into the Bam HI site of the plant-transformation-competent binary vector pCLD04541 and the Hind III and Eco RI sites of the conventional BAC vector pECBAC1. The ligated DNA was transformed into E. coli strain DH10B by electrophoration, and plated on LB agar medium containing appropriate antibiotics, IPTG and X-gal. The white colonies containing cotton DNA inserts were arrayed as individual clones in 384-well plates and maintained at –80^oC. The insert sizes of the clones randomly selected from the libraries were analyzed by Not I digestion followed by pulsed-field gel electrophoresis.

III. Results and Discussion

• We have developed two large-insert BAC libraries and one plant transformation-competent BIBAC library from cotton acc. TM-1, the standard line for cotton genetics and genomics studies (Table 1). A total 53,760 clones have been arrayed in 384-well plates for the Hind III BAC library and the clones of the Eco RI BAC library and the Bam HI BIBAC library are being arrayed. We plan to array about 50,000 clones for the Eco RI BAC library and 60,000 clones for the Bam HI BIBAC library. Together, these three TM-1 libraries contain 163,760 clones and cover 10.2x cotton (AD) haploid genomes. Fig. 1 shows the partial digestions of the TM-1 megabase DNA with Bam HI, Hind III and Eco RI, respectively; Figs. 2 - 4 show random BACs and BIBACs of these three TM-1 libraries. These three TM-1 BAC and BIBAC libraries plus the EST (the nearly-isogenic line of TM-1) BAC libraries and the Auburn 623 library that were previously developed by this group collectively covers 20.1x cotton (AD) haploid genomes.

• These three TM-1 BAC and BIBAC libraries have provided desirable resources for cotton (AD) genomics research, especially for genome-wide integrative physical and genetic mapping, chromosome walking and genome analysis. These are because:

1. TM-1 contains both AD genomes of important to cotton fiber production, is homogeneous and homozygous, and the standard line for cotton genetic studies. Therefore, it is the most desirable for cotton genomics research. The integrated map developed from its BAC and BIBAC libraries will be widely adaptable to different areas of cotton genomics research.

2. The two TM-1 BAC libraries represent the ones that have the largest insert sizes of cotton BAC libraries developed to date, which is essential for assembly of large contigs for genome-wide physical mapping. Although the insert size of the BIBAC library is 120 kb, smaller than those of the BAC libraries, it is the first large-insert library of cotton that is competent for genetic transformation via Agrobacterium. This BIBAC library will streamline map-based cloning, genetic engineering and molecular farming of the genes and QTLs important to cotton production.

3. These three TM-1 libraries are complementary to each other in genome coverage. This is because they were cloned in the three different cloning sites of the two different vector systems (pECBAC1 is F-factor plasmid-based and pCLD04541 is bacterial P1 plasmid-based). Therefore, the three libraries are likely to be equivalent to a shotgun randomly sheared library in genome coverage. This is significant for development of a genome-wide physical map of a high genome coverage by fingerprint analysis.

4. Additionally, these TM-1 libraries have only about 0.6% of their clones derived from chloroplast DNA (Fig. 5). The low contamination of chloroplast DNA-derived BACs will facilitate development of genome-wide physical mapping of the cotton genome from the libraries.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Please be aware of Disclaimer, Copyright and Trademark Statement

Dated:11 November 1999