Short biological polymers on solid supports
The present invention relates to a device and method for controlling and optimising or otherwise manipulating the characteristics of interaction between short biological polymers on solid supports to applied ligands. In particular, it relates to devices and methods of producing said devices comprising specific oligonucleotides on a solid support.
There are many methods described for coupling biological polymers to solid supports, including the use of biological and chemical spacers to confer functionality. It is known that the more the immobilised molecule is spatially removed from the solid support, the closer it is to the solution state and the more likely it is to react freely with a dissolved ligand. It is also known that beyond an optimal length, the functionality is lost which is possibly due to electro- static effects of the spacer. The optimum spacer length for hybridisation to elements comprising 12 bases of DNA has been determined to be in the region of 10 to 12 units, where one unit equals 6 carbon atom equivalents. However, spacers
longer than 2 units (12 carbon atom equivalents) have proven very difficult to make any type of quantity that lends itself to large scale utility, and because of this one unit (6 carbon atom equivalents) spacers are typically used in the fabrication and utility of solid support tools, such as gene chips (microarrays) to determine gene expression by hybridisation. This one unit spacer has been shown to be adequate for this type of requirement .
However, there is now a requirement for genetic microarrays to comprise every possible combination of nucleotide sequence of a given length or lengths of oligonucleotide. Such arrays have the utility for, but not limited to, antisense design, siRNA design, probe design, as well as many other possibilities for utilisation. In order for this type of array to be both cost effective and have routine applicability, there is a practical limit to the number of elements that is acceptable to be found on the array. Such a limitation realistically restricts the interaction defining element of for example an immobilised oligonucleotide (i.e. RNA, DNA or any native or synthetic analogue with or without a modified or non-nucleic acid backbone known in the art) to 4 to 8 units which gives a microarray with between 254 and 65,536 specific data generating elements. All of the prior art thus far has taught that hybridisation to such elements is both inconsistent and irreproducible. For example, in the publication "Nucleic Acid Research (1994), 22(24), 5456 - 5465" it is stated that the 12 nucleotide sequence (in a comparison of hybridisation as a function of length including 15 and 20 nucleotide sequences) was difficult to use due to its low melting
temperature. An ASO hybridisation sequence of 15 nucleotides was therefore chosen for all subsequent work.
It can be seen that it would be useful to have a method for spacing and attaching short oligonucleotides to solid supports in such a way as to optimise their ability to interact with any applied ligands. Here the term short should be taken to refer to oligonucleotides of less than 15 units and most preferably oligonucleotides of less than 9 units.
It can also be seen that it would be beneficial to provide devices and methods that have applicability for the determination of compositional and/or structural parameters pertaining to applied ligands.
According to a first aspect of the present invention, there is provided a method of attaching a polymer to a solid support, comprising the steps of attaching the polymer by means of a spacer, comprising 6 to 12 carbon atom equivalents followed by between 0 and 210 atoms of a nucleotide homopolymer (0 to 35 bases) followed by 54 atoms (9 bases) of a mixed nucleotide heteropolymer .
Preferably the spacer comprises 6 to 12 carbon atom equivalents of ethylene glycol.
Preferably the length and composition of the spacer portion is optimised to vary the distance of the defining sequence portion of the oligonucleotide from the solid support .
Preferably the oligonucleotides are printed on to the solid support .
Most preferably the oligonucleotides should be printed under conditions of high humidity.
Preferably the amount of time allowed for the oligonucleotide to conjugate to the surface of the solid support is optimised.
Most preferably the lowest required time for conjugation is used in order to prevent loss of specific signal on increase and background signal .
Alternatively the oligonucleotides may be printed onto the solid support under conditions wherein the oligonucleotides do not bind immediately to the surface but rather dry down and remain inert.
According to a second aspect of the present invention, there is provided an array of oligonucleotides conjugated to a solid support, comprising a solid support to which is attached a spacer molecule, comprising 6 to 12 carbon atom equivalents followed by between 0 and 210 atoms of a nucleotide homopolymer (0 to 35 bases) , followed by 54 atoms (9 bases) of a mixed nucleotide heteropolymer.
Preferably the space comprises 6 to 12 carbon atom equivalents of ethylene glycol.
Preferably the biological polymer is attached to the spacer molecule.
Preferably the array of oligonucelotides is stored under wet conditions.
Optionally the storage conditions are under water or under other aqueous or non-aqueous solution or solvent.
Alternatively the arrays are stored dry for subsequent reactivation prior to use.
A further option is that the storage conditions are under a salt or sugar solution that may be allowed to dry, thus coating the array in a semi-liquid film.
An additional option for storage conditions is that it is stored under or within an aqueous matrix, such as a gel.
Optionally the gel is agarose.
Another option is that the gel is Polyethylene glycol - (PEG) .
Alternatively the array is stored frozen in liquid.
Preferably the liquid will contain a cryo-protectant .
Optionally the cryo-protectant is chosen from the list glycerol, DMSO, sugar, ethylene glycol or PEG.
In order to provide a better understanding of the present invention, embodiments will now be described by way of
example only and with reference to the following Figures, in which:
Figure 1 is a bar chart which shows the effect of varying the length and composition of a spacer, indicating the change in signal in various cases with an increase in polynucleotide spacer lengths; and
Figure 2 is a line graph which shows the effect of varying printing and conjugation times once short oligonucleotides are printed on a solid support.
In the present invention, oligonucleotides are attached to a solid support by means of a spacer comprising a 6 and a 12 carbon atom equivalent of ethylene glycol, followed by between 0 and 210 atoms of nucleotide homopolymer (0 to 35 bases) , followed by 54 atoms (9 bases) of a mixed nucleotide heteropolymer. Attaching the oligonucleotide to the solid support in this matter allows modulation of the extent of binding and optimises the binding.
The use of a mixed heteropolymer as part of the spacer molecule serves at least two different functions. Firstly, it acts to distant the short interaction defining element away from the solid support. Secondly, it confers additional binding within a proportion of the individual molecules that comprise the population of oligonucleotides immobilised within the element. For example, the element may be made of DNA as described and ending in NNNNNNNXXXXXX 3', where N is any of C, A, G or T, and XXXXXX is a specific 6 base sequence which defines the element. Within such an element population, all
individuals will interact as 6 base oligonucleotides, and additionally one in four will act as a specific 7 base oligonucleotide with one and 16 acting as a specific 8 base oligonucleotide and so on. Such additional capacity to interact boosts any specific signal that has been generated, and therefore increases the quality of any data generated by the element .
The prior art teaches that for hybridisation to oligonucleotides on a solid support, increasing the spacing of the defining sequence from the surface by means of a homopolymeric nucleotide spacer will increase signal strength with increasing length of spacer (for example Guo et al (1994), Nucl . Acids Res. 22(24), 5456- 5465, (specifically figure 3d therein) . In this paper Guo et al test homopolymeric dT spacers up to 15 nucleotides in length and see increasing signal with increasing length. Further more they state "...the data suggests further gains in hybridisation signal strength may be attainable in this fashion".
Contrary to the teachings of the prior art and to the obvious inferences from those teachings, the present invention shows that there is an optimum nucleic acid spacer length for detection of ligand binding to short oligonucleotides immobilised on solid supports and that at other than optimum spacer length and composition, signal strength is lost.
The present invention teaches that for short immobilised oligonucleotides (or other polymers) it is necessary to optimise the spacer length in ways that are not envisaged in the prior art. Optimisation is achieved by varying
the length and composition of the spacer portion of the oligonucleotide to vary the distance of the defining sequence portion of the oligonucleotide from the solid support . It has been found in the present invention that the spacer should ideally comprise 6 to 12 carbon atom equivalents of ethylene glycol, followed by between 0 and 210 atoms of a nucleotide homopolymer (0 to 35 bases) followed by 54 atoms (9 bases) of a mixed nucleotide heteropolymer .
The characteristics of the surface must be taken into consideration during optimisation and the present invention teaches that the optimal spacer must be defined either experimentally or by the application of rules for each new surface derivation.
An example of varying the length and composition of spacer is shown in Figure 1. Note increasing and then decreasing signal strength with increasing poly- nucleotide spacer length. The example data also shows the effects of 12-carbon and 6-carbon linkers between the amino group (or other conjugation group designed to react with the surface) and the beginning of the poly- nucleotide spacer.
Figure 1 shows the effect of varying the length of the nucleic acid part of the spacer between the surface and the specific 6-base sequence at the 3' -end of the oligonucleotide on the array. The experiments were carried out on oligonucleotides separated from the surface by either a 12-carbon (C12+) or 6-carbon (C6+) linker in addition to the polynucleotide spacer. So the generic form of all oligos is:
Amino linker—carbon spacer (6 or 12 atom equivalent) - variable poly A spacer (n=10-35) -N9-X6
Where N is any nucleotide and X is a specific defined sequence that varies from oligo to oligo.
The oligos indicated on figure 1 are numbered 27 to 31 and represent a peak of accessibility on a model gene transcript, Ki-67.
The data shows that for any of the oligonucleotides increasing the carbon spacer from 6-atoms to 12-atoms increases the signal from that oligonucleotide.
The data also shows that for either C-12 or C-6, increasing the length of the poly-A part of the spacer initially increases signal and then beyond AlO the signal strength decreases . This is contrary to the teachings ±n the prior art and illustrates the importance of optimisation of this type of short-oligo array element.
Printing conditions have also been found to be important for functionality of short oligo arrays and are contrary to teachings in the prior art. The present invention teaches that short oligo arrays may have to be printed under conditions of high humidity and that they should not be dried out. Printing humidity should be tested and optimised up to 100 % and arrays may have to be stored under wet conditions. Storage conditions include but are not limited to e.g. under water or other aqueous or non- aqueous solution or solvent. Under a salt or sugar solution that may be allowed to dry thus coating the
array in a semi-liquid film. Under or within an aqueous matrix such as a gel made from e.g. but not limited to agarose or similar polymer or PEG or a similar polymer" . Arrays may also be stored frozen in liquid including a cryo-protectant such as but not limited to glycerol, DMSO, sugar, ethylene glycol, PEG or similar.
When printing short oligo arrays it is necessary to optimise the time allowed for the element to conjugate to the surface as this is also contrary to teachings in the prior art. For example, and shown in figure 2, printing of amino-modified short oligos onto derivatised plastic slides is optimum at less than 8 hours (in this example, in other specific cases times may vary) . Prolonged conjugation times lead to a loss of specific signal and an increase in background.
Figure 2 shows the effect of varying printing and conjugation times once short oligos are printed onto a solid support. This data is provided as an illustration of the negative effects of prolonged conjugation times when making a short oligo array.
There is a minimum conjugation time, before which oligos tend to just wash off the solid support because they are not properly attached. In the example given this is between 30 and 60 minutes of total printing and conjugation time. This time may vary with different slide types and should be optimised. If conjugation times are prolonged then high background results and specific correct signal is lost. This is illustrated 2by comparing the 2 hour map with the 48 hour map. The two hour map shows specific interactions on the array whereas
the 48 hour map is becoming random. To further illustrate the effect of conjugation time on different types of peak: those that give real specific signal (1-7, 15-21, 21-27, and 27-33) tend to give good signal with printing/conjugation times of between 2 and 8 hours and then signal gets increasingly noisy and unpredictable. Peaks that correspond to background regions on the access map and are not expected to give high signal are (5-15, 33-40 and 40-48) in all of these cases the signal remains low and as expected for printing/conjugation times of up to 8 hours and then becomes unpredictable. The combination of decreasing specific signal and unpredictable background is what cause the poor access map at the 48 hours time point.
This example illustrates that printing/conjugation times can be critical when making short oligo arrays and must be optimised. Precise optimisation may vary from system to system and this is anticipated in this disclosure. The way the signal and background varies with the way the short oligo arrays are printed and conjugated is not anticipated in the prior art.
An alternative is that the oligonucleotides may be printed onto the solid support under conditions wherein the oligonucleotides do not bind immediately to the surface but rather dry down and remain inert. For example oligonucleotides may be dissolved in a neutral salt solution (the spotting solution) and printed onto supports that normally require a high pH spotting solution such as for example carbonate buffer at >pH9. The neutral spotting buffer will not promote conjugation of the oligonucleotide to the support and the array can
therefore be printed and stored under standard low- humidity conditions . To activate the array it is exposed by dipping in high pH activation buffer for a time and at a concentration that should be determined by empirical or other optimisation. Any method of printing that requires subsequent activation of the array Joy chemical, electric, photo, or other means is anticipated here as an extension of this concept .
Further modifications and improvements may be incorporated without departing from the scope of the invention herein intended.