Chemistry: molecular biology and microbiology – Measuring or testing process involving enzymes or... – Involving nucleic acid
Reexamination Certificate
1999-07-23
2001-06-19
Yucel, Remy (Department: 1635)
Chemistry: molecular biology and microbiology
Measuring or testing process involving enzymes or...
Involving nucleic acid
C536S023100, C536S024300
Reexamination Certificate
active
06248530
ABSTRACT:
The present invention relates to a method intended in particular for eliminating specific sequences from DNA libraries.
The invention relates more particularly to a method for eliminating at least one specific, in particular abundant, sequence from a cDNA library.
The random sequencing of cDNA clones has been used for many years as a means for obtaining differential expression profiles and for discovering new sequences. Typically, it involves extracting the messenger RNAs (mRNA) of particular cells or tissues, cloning the corresponding cDNAs and then randomly selecting clones which will be sequenced. If the objective is the search for new sequences, the operation of exhaustive sequencing of a cDNA library is made expensive because of the repetitiveness of the mRNA population. Indeed, while, according to the tissues, 15,000 to 50,000 molecular species of mRNA are expressed, they represent a total of 200 to 500,000 molecules in cells (Davidson and Britten, 1979).
In the cells, the messengers are expressed at different levels according to the molecular species. Thus, 10 to 20 molecular species may represent 10 to 20% of molecules of mRNA. Many authors have studied the repetitiveness of libraries and have shown that, in some cases, only one molecular species could count for more than 20% of the clones. In order to clarify these data, the inventors studied the repetitiveness of mRNA sequences in several human tissues. For that, they classified by similarity groups the sequences of clones randomly selected from 5′-end libraries.
For 2 to 3000 clones sequenced, the fractions of clones belonging to similarity groups containing 10 sequences or more are, respectively, 39%, 6.8%, 22%, 54%, 34%, 46%, 27%, 27%, 7% and 39%, for tissues or cells such as the placenta, brain, spleen, pancreas, colon, lymphocytes, liver, kidney, ovary and heart (Table I).
TABLE I
%
%
%
TISSUES
TOP15
TOP20
TOP30
Pancreas
46
48
52
Kidney [sic]
15
17
20
Colon
18
21
26
Lymphocyte
25
28
33
Kidney
23
26
30.5
Dystrophic muscle
26
27.5
31.5
Liver
27
29
33
Ovary
7.5
8
10
Brain
8
9
11
Table I: Abundance profile for the clones most widely represented in tagged cDNA libraries. The libraries are constructed from messenger RNAs from the: pancreas, spleen, colon, lymphocyte, kidney, dystrophic muscle, liver, ovary and brain. The table shows the distribution of the 15, 20 or 30 clones most widely represented in each of our libraries.
Thus, the cost of a new sequence may be very high; indeed, between a quarter and half of the sequences have already been produced very rapidly during the sequencing. For each sequence, there is therefore a between 50 and 75% chance that this sequence has already been made. To limit the cost of new sequences, several solutions have been proposed.
The first consists in normalizing the cDNA libraries or the starting mRNA population. This consists in reducing the frequency of the abundant sequences so that it is near the frequency of the rare sequences. Several normalization methods have been proposed. One of them consists in using the genomic DNA as normalization template (Weisman, 1987). In this case, the distribution of the cDNA clones obtained should be correlated with the number of exons in the gene.
The normalization method most widely used is based on the kinetics of hybridization. When a population of nucleic acids is denatured, the rate of renaturation (or of reannealing of the molecules) is proportional to the number of molecular species present. In other words, the higher the number of molecular species present in the cDNA population, the greater the rate of renaturation of the corresponding cDNA molecules (Britten et al., 1974; Young & Anderson, 1985). The principal technical difficulty of this approach is to correctly separate the populations of annealed molecules from the populations of nonannealed molecules.
Most of the methods involve a chromatography on hydroxyapatite resin which releases the single-strand molecules and the double-strand molecules at different concentrations of phosphate ions (Patanjali et al., 1991). Strategies involving semisolid systems for eliminating DNA—DNA duplexes or DNA-RNA heteroduplexes have been recently proposed (Sasaki et al., 1994). In this case, cDNAs are synthesized with the aid of primers to which latex beads have been attached at the 5′ end. The cDNA populations obtained are used as hybridization template with an mRNA population and the heteroduplexes are collected by centrifugation.
All the techniques for normalization by hybridization have the limitation of not allowing discrimination between weakly divergent sequences. Recently, Bento Soares (Soares et al., 1994; Soares & Efstratiadis, 1995) has proposed a method which makes it possible to subtract the abundant sequences from the cDNA libraries. This method is based on the mass production of single-strand plasmids from a cDNA library produced by priming anchored on the poly-A tail of the mRNAs. The defined end thus obtained is used to prime the synthesis of short segments of DNA on the single-strand plasmid template. This makes it possible to obtain molecules which are locally double-stranded. The small size of the neosynthesized fragment (200±20-nt) is an advantage in the subtraction strategy because the 3′-untranslated sequences are much more divergent than the coding sequences. Furthermore, since all the fragments have roughly the same size, the kinetics of hybridization for the entire plasmid population is simpler. Subsequently, the normalization of the library is carried out by denaturing the duplexes and by exploiting their different rehybridization kinetics. The separation of the single-strand circular plasmids and of the heteroduplexes is carried out by means of hydroxyapatite columns. The single-strand plasmids are then converted to a double-strand by primer extension and transfected into bacteria. This method has the advantage of placing the hybridization partners under equimolar condition; on the other hand, it has two disadvantages:
1) the purification of the nonrecombinant clones,
2) for a good hybridization, the size of the fragment neosynthesized from the single-strand circular plasmids should be approximately 200 nt.
In the case of the 5′-end libraries produced here, the average size of the inserts is 200 to 250 nt. Thus, the Soares normalization strategy is hardly applicable.
The second solution proposed for limiting the cost of new sequences consists in ordering the cDNA library on filters and in hybridizing the filters with probes corresponding to the most abundant messengers and then in ordering, in a microtiter plate, the clones which are not identified by the probe. The choice of the most abundant groups of sequences may be made by analyzing a few thousand clones (Lanfranchi et al., 1985), or alternatively by using probes consisting of the total cDNA, the mitochondrial genome and/or a few specific clones chosen for their great abundance (Adams et al., 1995). This approach may be attractive because, unlike normalization, it allows the location and then the elimination, by virtue of the selection of nonrecognized clones, of the clones corresponding to the most abundant sequences.
However, the manipulations for ordering the clones on a filter and for hybridization are cumbersome and the selection of the clones not recognized by the probes used is often delicate (hybridization background noise, precise identification of the clones, possible errors during the reordering and the subcloning of the library).
In the present invention, the elimination is carried out by making unclonable the plasmids having inserts corresponding to the targeted specific sequences which are among the most abundant. This is carried out by selectively directing the activity of restriction enzymes to the sequences to be eliminated. The clones thus linearized are no longer capable of transforming bacteria.
Accordingly, the present invention relates to a method for specifically eliminating at least one type of double-strand DNA plasmid from a set of plasmids in parti
Dumas Milne Edwards Jean-Baptiste
Nahas Nasri
Genset S.A.
Knobbe Martens Olson & Bear LLP
Yucel Remy
Zara Jane
LandOfFree
Method for eliminating specific sequences when constructing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for eliminating specific sequences when constructing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for eliminating specific sequences when constructing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2519352