Patent Application
20090061525
The present invention relates to analyzing a sample comprising different sample compounds in a microfluidic chip for electrophoretic separation and detection.
In microstructure technology applications as in the Agilent 2100 Bioanalyzer, by the applicant Agilent Technologies, fluid may be conveyed through miniaturized channels (which may be filled with gel material) formed in a substrate. For a capillary electrophoresis device as an example for such a microstructure technology application, an electric field is generated in the fluid channels in order to allow for a transport of components of the fluid through the channels using electric forces. Such an electric force or field may be generated by dipping contact pins of the capillary electrophoresis device into the fluid which may be filled in a well defined by a carrier element coupled to a microfluidic chip, and by applying an electrical voltage to such contact pins.
WO 00/78454 A1, DE 19928412 A1, and U.S. Pat. No. 6,814,846 by the same applicant Agilent Technologies show different microfluidic chips and applications. Other microfluidic devices and applications are disclosed e.g. in WO 98/49548, U.S. Pat. No. 6,280,589, or WO 96/04547.
The article “Thousandfold signal increase using field-amplified sample stacking for on-chip electrophoresis” by B. Jung, R. Bharadwaj and J. G. Santiago, Electrophoresis 2003, 24, pp. 3476-3483 describes a novel field-amplified sample stacking (FASS) capillary electrophoresis (CE) chip design that uses a photoinitiated porous polymer structure to facilitate sample injection and flow control for high-gradient FASS.
In the article “Optimization and Validation of a Quantitative Capillary Electrophoresis Sodium Dodecyl Sulfate Method for Quality Control and Stability Monitoring of Monoclonal Antibodies” by O. Salas-Solana, B. Tomlinson, S. Du, M. Parker, A. Strahan and S. Ma, Anal. Chem. 2006, 78, pp. 6583-6594, it is described that CE-SDS with precolumn labelling and laser-induced fluorescence detection is a robust methodology for the quantitative analysis of therapeutic rMAbs as it relates to size heterogeneity.
It is an object of the invention to improve detection sensitivity when detecting sample compounds that have been electrophoretically separated on a microfluidic chip. The object is solved by the independent claim(s). Further embodiments are shown by the dependent claim(s).
According to embodiments of the present invention, a method for analyzing a sample comprising different sample compounds is provided. The method comprises staining the sample compounds by adding a dye species to a solution of the sample, the dye species having reactive groups adapted for forming covalent bonds with specific groups of the sample compounds, and providing the modified sample compounds to a microfluidic chip, the microfluidic chip being adapted to provide an electrophoretic separation. The method further comprises electrophoretically separating the modified sample compounds, and detecting separated compounds.
In prior art on-chip electrophoresis, labelling of the sample species has been carried out by adding sodium dodecyl sulfate (also referred to as SDS) and fluorescence dye to the sample solution. Thus, complexes of sample species, fluorescence dye and SDS have been obtained. However, no chemical bonds have been formed between the sample species and the dye molecules.
In contrast to this labelling method of the prior art, it is proposed to perform a covalent labelling of the sample species before starting electrophoretic separation. According to embodiments of the present invention, chemical bonds are established between the reactive groups of the dye species and the sample compounds. The labelled sample is supplied to a microfluidic chip, and the microfluidic chip provides an electrophoretic separation.
Covalent dye labelling has never been employed for on-chip separation before. Especially in the context of microfluidic chips, covalent dye labelling seems to offer a wide range of advantages.
First of all, compared to conventional staining procedures, the level of background fluorescence is reduced. This leads to an increased signal-to-noise ratio. Hence, detection sensitivity is improved.
In prior art solutions, in order to reduce the background fluorescence, it has been necessary to dilute the sample solution after it has passed through the sieving matrix. This procedure has been referred to as “destaining”. According to embodiments of the present invention, the step of destaining may be omitted, because the amount of background fluorescence is so small that it is not necessary to dilute the sample solution before detecting separated compounds. Accordingly, on-chip analysis of the labelled sample species is simplified a lot. The extra well containing the destaining solvent may be omitted, too.
Destaining has always been a burdensome procedure, because it required a careful control of the electric currents flowing on the electrophoresis chip. Using covalent labelling of the sample species, the step of destaining is no longer necessary, and hence, the restrictions imposed with regard to the electric currents are removed as well. As a consequence, a high conductivity background buffer may be employed in the system. Using a high conductivity background buffer is especially advantageous with regard to an effect called “stacking”. Due to this effect, by using a background buffer of increased ionic strength, an increase of the respective concentrations of sample compounds is observed. As a result, the signal-to-noise ratio of an acquired peak pattern is considerably improved.
According to a preferred embodiment, the microfluidic chip comprises an electrophoretic separation channel, and the method comprises passing the modified sample compounds through the electrophoretic separation channel, thereby electrophoretically separating the modified sample compounds.
According to a further preferred embodiment, the microfluidic chip comprises a detection flow path that is fluidically coupled to the separation column's outlet, wherein separated compounds are detected in the detection flow path.
According to a further preferred embodiment, the sample compounds are covalently labelled with a fluorescent dye species. After the sample compounds have been electrophoretically separated, they may be detected by a fluorescence detection unit.
According to a preferred embodiment, the sample is a protein sample comprising a plurality of different protein species.
In a preferred embodiment, sodium or lithium dodecyl sulfate, also referred to as SDS or LiDS, is added to the protein sample. A solution of SDS or LiDS comprises CH3—(CH2)10—CH2—O—SO3− and Na+ or Li+. Adding SDS or LiDS to a protein sample and heating, the protein species are denatured, and negatively charged dodecyl sulfate-protein complexes are formed, whereby the mass-to-charge ratio is substantially constant. The dodecyl sulfate-protein complexes may be electrophoretically separated in a gel sieving matrix based on their respective size.
According to a preferred embodiment, the dye species is functionalized with reactive groups, said reactive groups being adapted for forming covalent bonds with amine groups of various different protein species. When adding the functionalized dye to the sample solution, the functionalized dye reacts with the protein species, and a dye-labelled protein species is obtained.
According to a preferred embodiment, a dye species that has been functionalized with N-hydroxy-succinimidyl-ester, also referred to as NHS, is used. N-hydroxy-succinimidyl-ester is a reactive group that forms covalent bonds with amine groups of the protein species.
According to a preferred embodiment, the conditions of the reaction between the dye species and the sample solution are controlled in a way that most molecules of the sample either react with only one dye molecule or with no dye molecule at all. Thus, a quantitative correlation between the fluorescence intensity detected by the detection unit and the amount of a respective sample compound is established. In particular, multiple dye labelling of a single sample molecule is avoided.
In a preferred embodiment, the dye species is added in stoichiometric deficiency to the sample in solution. According to another preferred embodiment, covalent labelling of sample species is performed under slowed-down reaction conditions. According to yet another preferred embodiment, the labelling reaction is carried out at low temperature, e.g. on ice.
According to a further preferred embodiment, for terminating the labelling reaction, a further species adapted for reacting with the remaining reactive groups of the dye molecules in solution is added. Thus, it is possible to control the reaction time of the reaction of sample species and reactive groups of the dye. By terminating the labelling reaction after a well-defined period of time, it is made sure that multiple dye labelling of a single sample molecule is avoided. According to a preferred embodiment, the further species is added in stoichiometric abundance. Further preferably, in case NHS-functionalized dye is used, lysine might e.g. be added for terminating the labelling reaction. Lysine comprises amine groups that react with the NHS-groups of those dye molecules that have not reacted with sample species yet.
According to a further preferred embodiment, the microfluidic chip comprises a detection flow path, and sample compounds passing through the detection flow path are detected by a detection unit that is external to the microfluidic chip.
In a further preferred embodiment, the microfluidic chip comprises a detection flow path, and a fluorescent dye solution is supplied to the detection flow path before analysing the separated sample compounds. The focus of the external detection unit is adjusted in dependence on the detected fluorescence image of the fluorescent dye solution flowing through the detection flow path. For example, the position of the external detection unit relative to the detection flow path may be adjusted until the image of the fluorescent dye flowing through the detection flow path is in focus.
According to a preferred embodiment, for adjusting the focus of the external detection unit, the fluorescent dye solution is electrokinetically moved from a dedicated well to the detection flow path. According to a further preferred embodiment, after autofocus, the fluorescent dye solution is moved back to the dedicated well, with the fluorescent dye being contained within its dedicated well for the entire chip run.
According to a preferred embodiment, fluid conduits of the microfluidic chip are filled with gel matrix and with a high conductivity background buffer. The sample is dissolved in a low conductivity sample buffer. The method comprises supplying a volume of low-conductivity sample buffer to a respective fluid conduit.
According to a further preferred embodiment, for electrokinetically moving the sample compounds, an electric field is applied to the respective fluid conduit. In the region of low conductivity sample buffer, the electric field is higher than in the region of high conductivity background buffer.
According to a further preferred embodiment, sample ions drift from the region of low conductivity sample buffer through a conductivity interface region and enter the region of high conductivity background buffer. When crossing the conductivity interface region, the concentration of the respective sample ions is increased. This effect is commonly referred to as “stacking”. The concentration increase of the sample compounds that is due to “stacking” leads to a corresponding increase in detection sensitivity.
When covalent labelling is performed before separating and detecting compounds of a sample, the background fluorescence is almost negligible. Therefore, it is no longer necessary to “destain” the sample solution before detecting the various sample compounds. As a consequence, restrictions imposed by the maximum allowable current become less burdensome, and a background buffer of increased ionic strength may be employed. The higher the ionic strength of the background buffer, the more pronounced the effects related to stacking will become. Hence, the increase of the ionic strength of the background buffer will cause an increase of the concentrations of sample compounds. This concentration increase gives rise to an improved signal-to-noise ratio when detecting the sample compounds. According to a preferred embodiment, the background buffer's conductivity is at least five times higher than the conductivity of the sample buffer with the sample dissolved therein.
According to embodiments of the present invention, a microfluidic chip adapted to provide an electrophoretic separation of sample compounds of a sample comprises a detection flow path, and a well filled with fluorescent dye, wherein the well is fluidically coupled with the inlet of the detection flow path. The microfluidic chip further comprises one or more electrodes adapted for electrokinetically moving the fluorescent dye from the well to the detection flow path.
For detecting sample species, a fluorescence detection unit has to be focussed onto the detection flow path of the microfluidic chip. According to a preferred embodiment, an extra well filled with fluorescent dye is provided. The fluorescent dye may be electrokinetically moved to the detection flow path. Thus, the fluorescence detection unit may recognize the fluorescent dye flowing through the detection flow path and focus on the detection flow path.
According to a preferred embodiment, after focusing, the dye is removed from the detection flow path and confined to its specific chip well during sample analysis. Therefore, the amount of background fluorescence during sample detection is substantially negligible.
A measurement apparatus according to embodiments of the present invention is adapted for analysing compounds of a sample. The measurement apparatus comprises a microfluidic chip as described above, a detection unit adapted for detecting separated sample compounds that pass through the detection flow path, and an adjustment unit adapted for adjusting the relative position of the detection unit relative to the detection flow path in dependence on the detected fluorescence of the fluorescent dye solution. The adjustment unit is adapted for varying the position of the detection unit until the detection unit is focussed onto the detection flow path.
Preferably, the detection unit is external to the microfluidic chip. For example, the detection unit might be a confocal microscopy unit.
Embodiments of the invention can be partly or entirely embodied or supported by one or more suitable software programs, which can be stored on or otherwise provided by any kind of data carrier, and which might be executed in or by any suitable data processing unit. Software programs or routines can be preferably applied for controlling operation of the microfluidic chip.
Other objects and many of the attendant advantages of embodiments of the present invention will be readily appreciated and become better understood by reference to the following more detailed description of embodiments in connection with the accompanied drawing(s). Features that are substantially or functionally equal or similar will be referred to by the same reference sign(s).
On-chip electrophoresis is a powerful technique for electrophoretically separating compounds of a given sample.
For size-based separation of protein samples comprising different protein species, sodium or lithium dodecyl sulphate, also referred to as SDS or LiDS, has traditionally been used for denaturing and solubilizing the protein species. A solution of SDS comprises CH3—(CH2)10—CH2—O—SO3− and Na+ or Li+. SDS or LiDS binds to polypeptide chains, whereby the sum of the negative charges of dodecyl sulphate anions is substantially proportional to the protein mass, resulting in similar charge densities and constant mass-to-charge-ratios. Hence, an electrophoretic separation of these dodecyl-sulphate-protein complexes based on size can be achieved in a sieving medium.
When using a fluorescence detection unit for analyzing separated protein species, the protein sample's compounds have to be labelled with a fluorescence dye before being subjected to electrophoretic separation. In prior art assays, a fluorescence dye has been added to a solution of protein sample and SDS. As a consequence, when protein-SDS-complexes are formed, fluorescence dye is embedded within these complexes, and thus, a fluorescence labelling of the denatured protein species is accomplished. However, one shortcoming of this prior art staining method is that the amount of background fluorescence is rather high, which leads to a decrease in detection sensitivity.
In order to reduce background fluorescence, prior art electrophoresis methods comprise a step of destaining, whereby a flow of non-fluorescent solvent is supplied to the separated sample compounds before they are detected. For example, the electrophoresis chip 1 shown in
In contrast to destaining methods of the prior art, embodiments of the present invention propose to perform dye labelling of the sample species by forming covalent bonds between the sample species and a dye species. Preferably, the dye species has reactive groups adapted for forming covalent bonds with specific functional groups of the sample species. After covalent dye labelling has been performed, the sample compounds are electrophoretically separated, and the separated sample compounds are detected in the detection flow path.
In the following, an example of protein labelling is given. In the example, a fluorescence dye shown in
First, a dye stock solution is prepared as follows:
As a result, 7.5 μl of 4 mM NHS-dye stock solution is obtained. At a temperature of −20° C., and protected from light, the dye stock solution may be stored for up to three months.
Starting from the above-described dye stock solution, a working dye stock solution is prepared as follows:
Thus, 5 μl of 0.4 mM working dye stock solution are obtained. The working dye stock solution may only be stored for three days at a temperature of −20° C., and protected from light.
Next, a protein labelling reaction is carried out as follows:
As a result, a modified protein species is obtained, the protein species being covalently labelled with fluorescence dye. The modified protein sample may be stored for up to three months at a temperature of −20° C., and protected from light, whereby freeze-thaw cycles should be avoided.
In
For the case of NHS being used as a reactive group of the dye species,
According to a preferred embodiment, the reaction conditions are controlled such that most polypeptides of the protein sample either react with only one dye molecule or with no dye molecule at all. This might e.g. be accomplished by adding the dye species in stoichiometric deficiency to the sample in solution. Furthermore, the reaction might be carried out under slowed-down reaction conditions, e.g. by keeping the sample solution on ice during the reaction time. By choosing adequate reaction conditions, it is made sure that most of the polypeptides either react with only one dye molecule or with no dye molecule at all. Hence, for most of the proteins in solution, multiple labelling is avoided.
When detecting the fluorescence of labelled polypeptides that pass the fluorescence detection unit, a detection peak corresponding to a certain level of fluorescence is recorded. Because each polypeptide comprises a single dye molecule at most, a correlation between the size and shape of the detected fluorescence peak and the amount of polypeptide in solution can be established. For example, the area below the detected peak may be translated into a corresponding concentration of the respective protein species in solution.
For terminating the reaction of the protein species with the functionalized dye molecules, 0.5 μl of 10 mM lysine is added in step 2.6 of the above-described protein labelling reaction. The chemical structure of lysine is shown in
The amount of lysine added to the sample solution does not impair further analysis of the sample solution. In contrast, the lysine-dye conjugate can be used as a lower marker for calibrating the peak pattern obtained during sample analysis. Lysine is an amino acid of low molecular weight and high mobility. Hence, the first peak of the acquired peak pattern can be attributed to lysine.
In addition to NHS-ester, there exist a variety of other reactive groups that may be used for forming chemical bonds between a dye species and a protein sample. For example, maleimide is another reactive group of particular interest for protein labelling.
Alternatively to covalently attaching a dye moiety to a protein by NHS chemistry one may also use maleimide chemistry, known to those skilled in the art. Maelimide chemistry would attach the dye moiety to the amino acid cysteine. Reaction control for minimal labelling by temperature and stoichiometry is possible here as well.
After the protein has been labelled, the protein sample is prepared for electrophoretic separation as follows:
In addition to the sample solution, a solution that is solely used for calibrating an obtained peak pattern is prepared. This solution comprises a plurality of protein species that correspond to a set of well-known peaks in a corresponding peak pattern. This solution is referred to as a “ladder”. The protein species of the “ladder” may e.g. be labelled with NHS-functionalized dye according to the above-described protein labelling reaction (cf. the above steps 2.1 to 2.7). Before being supplied to a dedicated well of the electrophoresis chip, the “ladder” is prepared for loading as follows:
When both the sample solution and the ladder are ready to load, “chip priming” is carried out, i.e. the electrophoresis chip is prepared for the chip run. First, gel is thawed up and supplied to one of the sample wells. Using a syringe capable of applying a pressure of several bars, the channels of the electrophoresis chip are entirely filled with gel. Furthermore, gel is supplied to some of the wells. Next, a well-defined volume of ladder is supplied to an appropriate well, and respective volumes of one or more sample solutions are supplied to respective sample wells. Furthermore, additional solvents might be supplied to appropriate wells. After chip priming has been carried out, the chip is placed into an analyzer unit, and a chip run is started.
By using covalent labelling for staining protein species, the amount of background fluorescence in a detection flow path of a microfluidic chip is significantly reduced. This leads to an improved signal-to-noise ratio and to an increase in detection sensitivity.
A further increase of detection sensitivity is related to an effect called “stacking”, which is illustrated in
The increase of sample concentration is strongly related to the ionic strength of the background buffer. By increasing the background buffer's conductivity, a more pronounced increase of sample concentration can be obtained. However, increasing the background buffer's ionic strength also increases the electric current flowing through the separation channel. The maximum electric current through the separation channel imposes an upper limit on the background buffer's ionic strength.
In prior art solutions, there has been a rather high concentration of fluorescence dye in the background buffer and accordingly, the amount of background fluorescence has been quite high. In order to reduce background fluorescence, it has been necessary to dilute the sample solution after it has passed the separation channel. For this purpose, a continuous flow of dilution solvent has been supplied to the sample solution before reaching the detection flow path. The step of diluting the sample solvent is generally referred to as “destaining”. For example, a continuous flow of destaining solvent might e.g. be supplied from the auxiliary well 8 shown in
According to embodiments of the present invention, by covalently labelling the protein species, the concentration of fluorescence dye in the background buffer is substantially negligible. Therefore, it is no longer necessary to dilute the sample solution before it reaches the detection flow path. In particular, the step of destaining can be omitted. As a consequence, the electrical current in the separation channel is no longer limited to 2.5 μA. In fact, the electrical current through the separation channel may now be raised up to about 25 μA, as it is no longer necessary to provide a current of ten times this magnitude in a destaining flow path.
Hence, covalent labelling allows for increasing the ionic strength of the background buffer. As a consequence, the effects due to stacking are enhanced.
In prior art solutions, a background buffer of 120 mM Tricine, 42 mM Tris-base, 0.2% SDS, has been employed at pH 7.7. The formula of Tris is (HO—CH2)3—C—NH2 and the formula of Tricine is (HO—CH2)3—C—NH—CH2—COOH. In contrast, according to embodiments of the present invention, a background buffer composed of 250 mM Tricine, 87.5 mM Tris-base, 1% SDS might be used. Compared to the formerly-used background buffer, the concentration of SDS is increased by a factor of 5, and the respective concentrations of Tricine and Tris-base are increased by a factor of 2. As a consequence, the conductivity of the new background buffer is increased by a factor of approximately 10 relative to the former background buffer's conductivity.
By enhancing the effects related to stacking, the respective concentrations of separated sample compounds are increased. The signal-to-noise ratio of the acquired peak pattern is improved, and the detection sensitivity is increased.
In prior art solutions, there has always been a certain amount of background fluorescence in the detection flow path 14. Hence, it has been possible to adjust the focus of the fluorescence detection unit 16 in dependence on an acquired fluorescence image of the detection flow path 14.
According to embodiments of the present invention, by covalently labelling the protein species, the concentration of fluorescence dye in the background buffer is almost negligible. According to an embodiment of the present invention, for performing the focusing, an auxiliary well 18 containing fluorescence dye is provided. By supplying suitable currents to electrodes 19, the fluorescence dye is electrokinetically moved from the auxiliary well 18 to the detection flow path 14. Then, the focus of the external fluorescence detection unit 16 is adjusted in dependence on the detected fluorescence image. After the optical unit is adjusted and focused to the microfluidic channel, the dye is moved back to its reservoir well and confined there during sample analysis. Therefore, fluorescent background during sample separation and detection is substantially negligible.
For example, in
| Number | Date | Country | Kind |
|---|---|---|---|
| 07115189.0 | Aug 2007 | EP | regional |
