|
https://www.abbs.info ISSN 0582-9879 |
A
High Efficient Approach Used for BAC-contig Extension of Oryza sativa
with PCR Screening the BAC Clone Pools
HU
Hao￥,
LI Tao￥,
MU Jie, HAN Bin*, HONG Guo-Fan*
(
National Center for Gene Research, Shanghai Institutes for Biological Sciences,
the Chinese Academy of Sciences, Shanghai 200233, China )
Abstract
To extend 8 BAC contigs, which were previously
located in the 56.1―68 cM region of the chromosome 4 of the Oryza
sativa indica GuangLuAi4, 14 pairs of primers were designed
according to the terminal sequences of the existing seed BACs and were
deliberately divided into 3 groups. With the 3 groups of primer mixtures, 233
pools of BAC DNA that represent 22 368 BAC clones from O.sativa indica
GuangLuAi4 genomic library were screened. 65 positive clones corresponding
to the 8 contigs were isolated and 29 clones of them were confirmed to be
extended to the seed BACs by end-sequencing and fingerprinting. The protocol
greatly enhanced the efficiency of the contig extension and was also superior
for its specificity, sensitivity and reusability to the colony in situ
hybridization which is a conventional method employed in contig extension and
physical map construction.
Key words Oryza
sativa; BAC; PCR; contig
Clone by clone
strategy, which is adopted by our laboratory in sequencing chromosome 4 of Oryza
sativa indica GuangLuAi4, is based on an extensive physical map[1―5].
Three steps were taken to construct physical map in our laboratory[6―10].
Firstly, anchoring by
genetic markers: The 22 368 BAC
clones representing the whole library were blotted on nylon membranes and colony in situ
hybridization using genetic markers as the probes were carried out.Thus the
first batch of seed BACs were anchored to the specific loci in the chromosome.
Secondly, chromosome walking: colony in situ hybridization was further
carried out using short DNA fragments in the terminal sequences of seed BACs as
the probes, thus BAC contigs were extended. Thirdly, end-sequencing, fingerprinting
and Southern blotting were carried out to corroborate the overlapping
relationships. Extending sizes as well as the overlapping size were also
determined.
Colony in situ
hybridization is a conventional method employed in contig extension and
physical map construction. But it has two major disadvantages. Firstly, it was
very time-consuming. Secondly, the sensitivity of hybridization is not high
enough so that some weak signals, which probably come from the low-content
templates, may be ignored. Thirdly, since repeat sequences take up about 50
percent of the rice genome, the non-specific hybridization rate could be very
high.
To improve the process
of the contig extension, an alternative approach based on PCR screening[11―15]
was employed to extend BAC contigs in the region from 56.1 cM to 68 cM on the
long arm of chromosome 4 of Oryza sativa indica GuangLuAi4 in the
study. The results were corroborated by several methods such as fingerprinting,
end-sequencing and especially the finished sequencing results.
1
Materials
The HindIII BAC
library of the O.sativa indica GuangLuAi4 genome was constructed
by TAO Quan-Zhou and HONG Guo-Fan et al. in 1994; ECL hybridization kits
were from Amersham Pharmacia Biotech Inc.; rTaq enzyme kits and DL 2000
DNA marker were from TaKaRa Biotechnology(Dalian) Co., Ltd; primers were
synthesized by BioAsia Biotechnology Co., Ltd, Shanghai;restriction enzymes
were from Boehringer Mannheim; chemicals were from Sigma Chemical Co., Ltd.
2
Methods
Templates of PCR were
prepared as following: 22368 BAC clones in the HindIII library are
preserved on 233 wells of 96-well plates. Firstly, each BAC clone was
inoculated in deep 96-well plates filled with 1.4 ml LB media with 25 mg/L
chloromycetin and incubated at 37 ℃
overnight, vigorously agitated. Then the plasmids were extracted by alkaline
lysis method. The product in each well was dissolved in 30 ml
TE buffer. Meanwhile, 233 tubes of DNA mixtures called template pools were
obtained in the similar way, each of which was equivalent to the mixture of the
plasmids from a specific plate. Thus the 233 plates and the same number of
tubes represented the 22 368 BACs, namely the HindIII library. They are “single
templates”
and “ template pools”
of PCR, respectively. Furthermore, before use, the 233 tubes of plasmids
mixture were dispatched on three 96-well plates, one well corresponding to one
tube.
Primers were designed
according to the following six criteria with the aid of software
GAPv4.4:(1)Primers were based on the sequences near the ends of seed BACs anchored by genetic markers.(2)The
sequences of the primers should not be repetitive sequences or have highly
similar sequences in the rice genome, which was affirmed by BLAST.(3)The
lengths of PCR products were dispersed in a range from 200―500
bp thus length information obtained from agarose electrophoresis image could
help reduce the amount of labor, yet the range was not further wider for the
sake of PCR condition uniformity which directly affected the PCR efficiency.(4)As to the two primers
that are in a pair, there should not be continuous base pairing that is 4 bp or
longer both in the same primer and between the two primers.(5)The lengths of
primers range from 18―22
bp.(6)The annealing temperature is about 52 ℃.
Thus, altogether 14 pairs of primers were designed and synthesized.
The sequences of the
primers are listed in Table 1.
The 14 pairs of
primers were divided into 3 groups each including 3, 5, 6 pairs of specific
primers.The groups were divided following 2 criteria:(1)No base pairing that is
4 bp or longer exists in the same primer group lest the false result happen or
the PCR efficiency drops.(2)The lengths of PCR products in one group should be
different enough to be distinguished on the agarose electrophoresis image,
which help find the specific pair of primers corresponding to the PCR products.
2% agarose was used for best resolution.
The PCR procedure was
as following: Denature: ⑴
95 ℃, 20 s; First round: ⑵
95 ℃, 20 s; ⑶
54 ℃, 30 s; ⑷
72 ℃, 30 s; ⑸
go to ⑵
for 5 cycles; Second round: ⑹
95 ℃, 20 s; ⑺
52 ℃, 30 s; ⑻
72 ℃, 30 s; ⑼
go to ⑹
for 25 cycles; Store: ⑽
4 ℃ store. The annealing temperature of
first round was 2 ℃
higher than that of second round so that false positive rate could be
reduced. The PCR system was
premixed as indicated in Table 2.The most important in the Table 2 is the
concentration of the templates. Too high concentration would lead to complex
false positive results.
PCR reactions were
carried out with a 3-step strategy. PCR step 1: To determine the relation between
the primer groups and the template pools. In this step, primer group were
reacted with template pools. For example, the primer group 1, which contained 3
pairs of primers, reacted with the three 96-well plates representing the whole
233 template pools. Only with one 4-block thermal cycler and in 80 minutes,
information was obtained about which template pools had positive clones
corresponding to the certain primer group (Fig.1).
Fig.1 Gel electrophoresis of the pool-PCR
results using Group1 primers
There are three pairs of
primers(H815c01HP1F/R, H622f05SP6F/R, H123b08HP1F/R) in primer Group 1. The
product of H123b08HP1F/R was
shorter than 250 bp( 213 bp ) while the products of the other two pair were longer
than 250 bp( 296 bp and 331 bp ). Since one band of the DL2000 marker
represents 250 bp, each positive result was attributed to H123b08HP1F/R(when
<250 bp) or the other two primer pairs(when>250 bp). Such an attribution
reduced the amount of labor in step 2. The positive results belonging to
H123b08HP1F/R: H321, H506; The positive results belonging to either
H815c01HP1F/R or H622f05SP6F/R : H302, H505, H602, H801, H821, H823; The false
results: H312, H315, H321, H414, H424, H506, H523, H622, H701, H815, H802,
H820, H925.
Moreover, since the
length of products haddistinguishable differences, positive results could be attributed
to fewer, such as 2 specific pairs of primers within the primer group 1. PCR
step 2: To determine the relation between the specific pair of primers and the
template pool(s). In this step, every positive template pool was attributed to
a certain pair of primers in the primer group (Fig.2) by PCR.
Fig.2 Part of the results from PCR step 2
Each positive result in Fig.1 was picked
out and reacted with one of the pair(s) of primers to which it was attributed
in PCR step 1. Thus each positive result (representing 96 wells on a certain
plate) was attributed to a single pair of primers.
PCR step 3: To determine
the positive BAC(s) for the specific pair of primers. In this step, the
specific pair of primers reacted with the certain 96-well plate templates
corresponding to the positive template pool, to determine the positive BAC
template (Fig.3).
Fig.3 Part of the results from step 3
After PCR step 1 and step 2, the
relationship between single plate and single pair of primers was established.
In step 3 each one of the 96 BAC templates that were on a positive plate was
reacted with the pair of primers to which the positive plate was attributed. A
positive BAC clone H505f09 was picked out.
Positive results
obtained from PCR were subjected to mature techniques in our lab for
confirmation, such as end-sequencing, finger-printing, Southern blotting (see
Fig.4).
Fig.4 One of the results of fingerprinting
(A) and hybridization [(B), seed BAC as the probe]
1,
l/HindIII marker; 2, H502b11(the seed BAC); 3,
H123f08; 4, H204e02; 5, H615d10; 6, H712d03.
Overlapping size
between the seed BAC and the positive BAC was estimated based on fingerprinting
and Southern blotting results and accurately corroborated by end sequencing.
Clones that had overlapping size less than 25% but more than 3 kb were selected
for sequencing.
3
Results
65 BAC clones were
identified by using pool-PCR screening to be located in the region from 56.1―68
cM on the chromosome 4 of O.sativa indica GuangLuAi4. 29 BAC
clones of them were further identified to be located on the 8 contigs by
Southern hybridization and BAC end-sequencing analysis. The relationship
between the identified BACs and the seed BACs are shown in the Table 3.
In total, the 8
contigs have been extended 510 kb from both directions. One of the identified
BAC clones H321a05 was selected and sequenced and the sequencing results
supported the conclusion in this work( Fig.5).
Fig.5 The BAC contig map of the chromosome 4
of O.sativa indica GuangLuAi4 (56.1―68
cM)
4
Discussion
Pool-PCR has 4
advantages over traditional method:(1)High efficiencyPooled templates and
primer groups reduced the time of contig extension greatly. To finish the
elongation from 16 BAC ends, a proficient technician need only 2―3
weeks using pool-PCR while several months will be needed by traditional
methods. (2)High sensitivityFollowing the protocol in this article, all seed
BAC clones in the physical map that has been identified by traditional methods
have been picked out without exception. Furthermore, PCR consumed by far less
DNA templates than end-sequencing, fingerprinting and Southern blotting.
Through a series of tests we have found the least yet still efficient template
dosage as little as 1 ng. Too much template could cause false positive results.
(3)ReusabilityOnce the BAC template pools has been prepared, they are enough for at least 60 times of
chromosome walking from 14 BAC ends. In most situations, it is well enough for
the whole chromosome walking.(4)SpecificitySome information from the
electrophoresis images could help us to distinguish the false positive results
from the real ones thus the specificity and readability were greatly enhanced
compared to the colony in situ hybridization. Firstly, the lengths of
the false positive products were different from what be supposed in most
cases(see Fig.1). Secondly, false positive results due to cross-contamination
of templates or homologous sequences always gave out weaker signals. Thirdly,
since the cross-contamination often happened in the wells near the real
positive ones. If on the electrophoresis image a strong signal was followed by
a series of weaker signals especially in a declining pattern, conclusion could
be made with confidence that the strongestsignal represented the real positive
result.
To make the protocol
more efficient, we also tried to use the bacteria in the BAC library directly
as the templates for PCR. In 8 times of test in which all reaction conditions
were the same, only 2―3
times had we got positive results . It seemed that the sensitivity and
repeatability were greatly reduced compared with that DNA itself acting as the
templates. Further experiments will be carried out to optimize the reaction
conditions for bacterial PCR.
However, pool-PCR also
has its disadvantages. From a lot of experimental results, we can see PCR is a
very sensitive and complex reaction conforming to Chaos theory, somewhat like
explosion process. Small disturbance
can grow more and more violent until at last a negative clone could give out a
seemly positive result. Such
disturbance comes from 2 resources: (1)“Agglomeration”
among primersWe have reduced such disturbance to a tolerable extent through
computer-aided primer segregation (See Methods). (2)Cross-contamination of
templatesPCR is a very sensitive reaction that even a single copy of template
can be identified. We’ve developed a set of strict rules to minimize
cross-contamination, from plasmid extraction to preservation, and to PCR reaction.
Such a set of rules can reduce false positive results greatly. In most cases,
false positive results can be distinguished from real positive results for an
experienced researcher.
References
1 Hong GF. Rice Genome Project,
Shanghai: Shanghai Scientific and Technical Publishers, 1999
2 Shimamoto K. The molecular biology of
rce. Science, 1995, 270(5243): 1772―1773
3 Tao QZ, Zhao HY, Qiu LF, Hong GF.
Construction of a full bacterial artificial chromosome (BAC) library of Oryza
sativa genome. Cell Res, 1994, 4: 127―133
4 Tao QZ, Qian YM, Zhao HY, Yu SL, Qiu
LF, Wu BQ, Zhu J et al. Construction of Oryza sativa genome
contigs by fingerprint strategy. Cell Res, 1995, 5: 263―271
5 Hong GF, Qian YM, Yu SL, Hu X, Zhu J,
Tao WH, Li W et al. A 120 kilobase resolution contig map of the rice
genome. DNA Seq, 1997, 7(6): 319―335
6 McPherson JD, Marra M, Hillier L,
Waterston RH, Chinwalla A, Wallis J, Sekhon M et al. A physical map of
the human genome. Nature,
2001, 409(6822): 934―941
7 Zhang HB, Wing RA. Physical mapping of
the rice genome with BACs. Plant Mol Biol, 1997, 35(1-2): 115―127
8 Hong GF. A rapid and accurate strategy
for rice contig map construction by combination of fingerprinting and
hybridization. Plant Mol Biol, 1997, 35(1-2): 129―133
9 Qin S, Nowak NJ, Zhang J, Sait SN,
Mayers PG, Higgins MJ, Cheng Y et al. A high-resolution physical map of
human chromosome 11. Proc Natl Acad Sci USA, 1996, 93(7): 3149―3154
10 Chen ZH, Zhou B, Han B, Qian YM, Hong
GF. Structural analysis of a gene cluster encoding two cationic and three
anionic peroxidases from rice chromosome 4. Acta Biochim Biophys Sin,
2001, 33(2): 163―172
11 Aburatani H, Stanton VP Jr, Housman DE.
High-resolution physical mapping by combined Alu-hybridization / PCR screening:
Construction of a yeast artificial chromosome map covering 31 centimorgan
in 3p21-p14. Proc Natl Acad Sci USA, 1996, 93(9): 4474―4479
12 Munroe DJ, Loebbert R, Bric E, Whitton
T, Prawitt D, Vu D, Buckler A et al. Systematic screening of an arrayed
cDNA library by PCR. Proc Natl Acad Sci USA, 1995, 92(6): 2209―2213
13 D’Esposito M, Mazzarella R, Pengue G,
Jones C, D’Urso M, Schlessinger D.
PCR-based immortalization and screening of hierarchical pools of cDNAs. Nucleic
Acids Res, 1994, 22(22): 4806―4809
14 Asakawa S, Abe I, Kudoh Y, Kishi N,
Wang Y, Kubota R, Kudoh J et al. Human BAC library: Construction and rapid screening. Gene,
1997, 191(1): 69―79
15 Yoshida K, Strathmann MP, Mayeda CA,
Martin CH, Palazzolo MJ. A simple and efficient method for constructing high
resolution physical maps. Nucleic Acids Res, 1993, 21(15): 3553―3562
Received: November 20,
2001 Accepted: January
9, 2002
This work was involved
in the International Rice Genome Sequencing Project (IRGSP) and was supported
by the Chinese Academy of Sciences, The Ministry of Science and Technology and
Shanghai Municipal Commission of Science and Technology
￥Both
authors contributed equally to this work
*
Corresponding authors: HAN Bin: Tel, 86-21-64845260; Fax, 86-21-64825775;
e-mail, [email protected]; HONG Guo-Fan: Tel, 86-21-64516371; Fax,
86-21-64825775; e-mail, [email protected] Communication