Functionalized biochips for SPR-MS coupling

Abstract

The invention relates to a method for coupling in-line the analysis of molecular interactions by surface plasmon resonance (SPR) with a structural identification by mass spectrometry using the same functionalized support for both types of analysis.

Claims

1. A method for functionalizing the metal face of a glass support for analysis by surface plasmon resonance, said method comprising grafting a self-assembled monolayer of poly(ethylene oxide) directly onto the metal face of said support, in which the metal of the metal face is gold, and in which the poly(ethylene oxide) is a compound of formula (I)
S—(CH.sub.2).sub.n—(O—CH.sub.2—CH.sub.2).sub.x-D  (I) in which: n is equal to 1 or 2; x is an integer between 5 and 16; and D is a group able to bind biomolecules or is a group that may be transformed in a group able to bind biomolecules.

2. The method according to claim 1, in which n is equal to 2.

3. The method according to claim 1, in which x is equal to 8.

4. The method according to claim 1, in which the PEO is O-(2-mercaptoethyl)-O′-(2-carboxyethy)heptaethylene glycol of formula HS—CH.sub.2—CH.sub.2—(O—CH.sub.2—CH.sub.2).sub.8—COOH.

5. The method according to claim 1, which method comprises the series of following steps: 1) prior cleaning of the support; 2) grafting of the PEO onto the support; and 3) optionally, modification of group D of the PEO.

6. The method according to claim 5, in which the cleaning is carried out by UV-ozone treatment.

7. The method according to claim 5, in which group D represents a —COOH group which is modified in step 3) so as to give an N-hydroxysuccinimide group.

8. The method according to claim 5, in which the grafting is carried out by immersing the support in a vessel containing the poly(ethylene oxide) to be grafted, in solution.

9. A plasmon resonance, mass spectrometry support comprising a glass support, said glass support comprising a single, functionalized face, said face comprising a gold, surface plasmon resonance measuring surface, said measuring surface comprising a self-assembled monolayer of poly(ethylene oxide) directly grafted to the gold, the poly(ethylene oxide) being a compound of formula (I)
—S(CH.sub.2).sub.n—(O—CH.sub.2—CH.sub.2).sub.x-D  (I) in which: n is equal to 1 or 2; x is an integer between 5 and 16; and D is a group able to bind biomolecules or is a group that may be transformed in a group able to bind biomolecules.

10. The plasmon resonance, mass spectrometry support according to claim 9, further comprising biomolecules immobilized on said support via the D group.

11. A method of surface plasmon resonance comprising binding molecules to be detected by plasmon resonance to a functionalized support according to claim 9.

12. The method according to claim 11, said method comprising studying the molecular interactions of the molecules by surface plasmon resonance.

13. A method of mass spectrometry comprising structurally identifying molecules covalently linked to the support according to claim 9.

14. The method according to claim 13, in which the mass spectrometry is carried out with MALDI-type ionization.

15. The method consecutively performing surface plasmon resonance and mass spectrometry on a sample comprising immobilizing molecules on the support of claim 9; analyzing said molecules on the support by surface plasmon resonance and then analyzing said molecules on the support by mass spectrometry.

16. The method according to claim 15, in which analysis by surface plasmon resonance is an analysis by surface plasmon resonance i.

17. The method according to claim 15, in which the analysis by mass spectrometry is an analysis of MALDI.

18. A method for coupling an analysis by surface plasmon resonance (SPR) with an analysis by mass spectrometry comprising: 1) immobilization of one or more molecules on the support according to claim 9; then 2) placing of the support in an SPR analyzer and analyzing, by SPR, of the interactions between the one or more molecules immobilized and a sample of analytes; then 3) removing the support from the SPR analyzer and placing the support in a mass spectrometer, and performing structural analysis, by MS, of the analytes specifically retained by the one or more molecules during the SPR analysis.

19. A method for coupling an analysis by SPR with an analysis by MS comprising: 1) immobilizing one or more molecules on a support according to claim 9; then 2) placing of the support in an SPR analyzer and analyzing by SPR the interactions between the one or more molecules immobilized and a sample of analytes; 3) in situ localized enzyme digestion of the analytes retained on the one or more molecules immobilized on the support, then placing of the support in a mass spectrometer structurally analyzing, by MS the products of digestion of the analyte(s) present on the support.

20. A method for coupling an analysis by SPR with an analysis by MS comprising: 1) immobilization of one or more molecules on the support according to claim 9; then 2) placing of the support in an SPR analyzer and analyzing, by SPR, of the interactions between the one or more molecules immobilized and a sample of analytes; then 3) removal of the support from the SPR analyzer and placing of said support in a mass spectrometer, and structurally analyzing, by MS, the analytes specifically retained by the molecules during the SPR analysis; then 4) in situ localized enzyme digestion of the analytes retained on the molecules immobilized on the support, and placing of the support in a mass spectrometer, and structurally analyzing, by MALDI MS or MALDI MS/MS, the products of digestion of the molecules present on the support.

21. The plasmon resonance, mass spectrometry support of claim 9, wherein D is an N-hydroxysuccinimide group, a succinimidyl ester group, a sulfosuccinimidyl ester group, a maleimide functionalized group, an iodoacetyl functionalized group, or a carboxylic acid group.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1 is a diagram representing a support according to the invention, comprising a metal face made of gold and the steps for functionalizing this surface. The surface is functionalized with a PEO of formula HS—CH.sub.2—CH.sub.2—(O—CH.sub.2—CH.sub.2).sub.8—COOH, then this PEO is modified with an NHS group.

(2) FIG. 2A provides results of PEO-NHS biochip with a PEO-NHS-functionalized surface.

(3) FIG. 2B provides results of a PEO-NHS biochip with anti-β-lactoglobulin antibody spots.

(4) FIG. 2C provides results of a MUA-CDI biochip with MUA-CDI-functionalized surface.

(5) FIG. 2D provides results of a MUA CDI biochip with anti-β-lactoglobulin antibody spots, FIGS. 2A-2D represent the results of the comparison between PEO-NHS self-assembled monolayers and MUA-CDI self-assembled monolayers grafted to the gold surface of SPR-MS biochips.

(6) FIG. 3A shows successive injections of proteins, β-lactoglobulin then ovalbumin 1, 10, 100, 200 μg/ml, representing an SPRi experiment on a biochip having undergone a PEO-NHS surface treatment according to the invention.

(7) FIG. 3B shows successive injections of proteins, β-lactoglobulin then ovalbumin 1, 10, 100, 200 μg/ml and with FIG. 3A showing the specificity of interaction of the analytes for their receptors.

(8) FIG. 4A, with FIGS. 4B and 4C represents an example of SPRi-MALDI-TOF MS coupling on a biochip of the type self-assembled PEO-NHS monolayer on a gold surface.

(9) FIG. 4B shows results of PEO-functionalized gold, Anti-ovalbumin and Anti-β-lactoglobulin.

(10) FIG. 4C shows an example of SPRi-MALDI-TOF MS of the example.

(11) FIG. 5 represents an example of proteomic analysis after an SPRi experiment and then analysis by MALDI-TOF MS of the analytes digested in situ. The experiment is carried out on a biochip of the type PEO-NHS self-assembled monolayer on a gold surface.

(12) FIG. 6A with FIG. 6B represents an example of proteomic analysis after an SPRi experiment and then sequencing by MALDI-MS/MS of the peptides derived from the in situ digestion of the ovalbumin protein captured by the biochip. The experiment is carried out on a biochip of the PEO-NHS type self-assembled monolayer on a gold surface. The MALDI-MS/MS spectra of peptides derived from the digestion-on-biochip and the identification of the protein retained using the “Mascot” computer tool are represented on the figure.

(13) FIG. 6B shows sequence of chicken ovalbumin and peptides identified.

EXAMPLES

Example 1

Protocol for Preparing a PEO-NHS Self-Assembled Monolayer Slide

(14) Dimensions of the Detachable Biochip

(15) Glass slide 500 μm thick, with dimensions of 12 mm×28 mm.

(16) Thickness of chromium: 1 to 2 nm.

(17) Thickness of the gold layer deposited at the surface: 50 nm.

(18) Characteristics of the PEO

(19) The bifunctionalized PEO used is of commercial origin (Sigma-Aldrich; ref: 672688). It is O-(2-mercaptoethyl)-O′-(2-carboxyethy)heptaethylene glycol, composed of 8 EO [ethylene oxide] units and functionalized at its two ends:
HS—CH.sub.2—CH.sub.2—(O—CH.sub.2—CH.sub.2—).sub.8—COOH

(20) (M=458.56 g.Math.mol.sup.−1)

(21) The thiol (mercapto) function allows anchoring of the polymer onto the metal face of the glass slide. The carboxylic acid function at the other end allows the functionalization of the chain, necessary for the covalent bonding of the “receptor” biomolecules.

(22) Protocol for Preparing the Deposit of PEO-NHS in the Form of a Self-Assembled Monolayer at the Surface of the Biochip

(23) After a prior cleaning step, the PEO is first deposited so as to form a self-assembled monolayer, and then, in the next step, its carboxyl end is functionalized with the N-hydroxysuccinimide group (cf. diagram FIG. 1).

(24) 1) Cleaning of the Slides by UV-Ozone Treatment

(25) Gold-coated glass slides, the dimensions of which are given above, are placed on optical paper, gold face upward, and a UV-ozone treatment is carried out for 1 h.

(26) 2) Deposition of PEO: Protocol for a Slide

(27) In a 100 ml beaker, 3 mg (6.54×10.sup.−6 mol) of HS—CH.sub.2—CH.sub.2—(O—CH.sub.2—CH.sub.2—).sub.8—COOH are dissolved in 2.6 ml of absolute ethanol (final PEO concentration of 2.5 mM or 1.15 mg/ml). The solution is homogenized in the beaker, and then a slide is immersed in said beaker, directly after the UV-ozone treatment.

(28) The beaker is then covered with 3 layers of parafilm so as to prevent evaporation. The beaker is subsequently agitated for 6 hours (Rocking Platform speed=20). After this step, the PEO-treated slide is recovered and the excess solution is absorbed on blotting paper. The treated slide is then rinsed in two baths of absolute ethanol, dried, and then stored in a refrigerator while waiting to carry out the functionalization reaction with N-hydroxysuccinimide (NHS).

(29) 3) Functionalization of the PEO with N-hydroxysuccinimide

(30) A solution of 1.09 ml of DMSO containing 27.6 mg (2.4×10.sup.−4 mol) of N-hydroxysuccinimide, 53 mg (2.6×10.sup.−4 mol) of N,N′-dicyclohexylcarbodiimide and 3.5 mg (0.23×10.sup.−4 mol) of 4-pyrrolidinopyridine is prepared in a 100 ml beaker. The solution is then homogenized.

(31) The slide previously treated with the PEO is then placed in the freshly prepared solution of DMSO, and the beaker is covered with two layers of parafilm. The reaction is then carried out by leaving the solution to act for 24 hours with agitation (Rocking Platform speed=20). The excess solution is removed on blotting paper and the slide is then rinsed once in a bath of DMSO and then 5 times in baths of ultrapure water and once in a bath of absolute ethanol. The slide thus treated is then dried and then stored in a refrigerator, under dry conditions, before use.

Example 2

Presentation of an SPRi Experiment on a Biochip of the Type PEO-NHS Self-Assembled Monolayer at the Gold Surface

(32) This example relates to an SPRi experiment on a biochip having undergone a PEO-NHS surface treatment according to the invention and showing the specificity of interaction of the analytes for their receptors (cf. FIG. 3).

(33) This experiment relates to the capacity for adsorption of two protein analytes, ovalbumin and β-lactoglobulin, onto spots of anti-ovalbumin and anti-β-lactoglobulin antibodies (600 nM and 6 μM) immobilized on the functionalized surface, in comparison with measurements carried out on zones without antibodies (zones corresponding to the PEO-NHS chemistry without antibodies) and where no adsorption of analytes is recorded.

(34) The conditions of the experiment are the following:

(35) The interaction of ovalbumin and of β-lactoglobulin with the anti-ovalbumin and anti-β-lactoglobulin antibodies immobilized on the surface is monitored by SPRi. At the beginning of the experiment, lysine (100 M) is injected in order to neutralize the NHS groups of the PEO-functionalized surface which have not reacted with the antibodies. Next, the two proteins, ovalbumin and β-lactoglobulin (50 μg/ml), are successively injected at the flow rate of 50 μl/min. Solutions of increasing concentration of β-lactoglobulin (1, 10, 100 and 200 μg/ml) and then of increasing concentration of ovalbumin (1, 10, 100 and 200 μg/ml) are injected. The running buffer is 10 mM ammonium acetate, pH 7.5. The images shown illustrate the adsorption of the proteins onto their respective antibodies for concentrations of 200 μg/ml. No protein is adsorbed onto the lysine-inactivated, PEO-NHS-functionalized surface.

Example 3

Presentation of an SPRi-MALDI-TOF MS Coupling on a Biochip of the Type PEO-NHS Self-Assembled Monolayer on the Gold Surface

(36) This example relates to SPRi-MALDI-TOF MS coupling on a biochip having undergone a surface treatment according to the invention (cf. FIG. 4).

(37) This example relates to the SPR analysis of the interaction of ovalbumin and of β-lactoglobulin with their respective anti-ovalbumin and anti-β-lactoglobulin antibodies (600 nM and 6 μM) immobilized on the functionalized surface in the form of spots, and to the detection of the protein analytes by MALDI-TOF MS.

(38) The conditions of the experiment are the following. The interaction of the ovalbumin and of the β-lactoglobulin with the anti-ovalbumin and anti-β-lactoglobulin antibodies immobilized on the surface is monitored by SPRi. At the beginning of the experiment, lysine (100 μM) is injected in order to neutralize the NHS groups of the PEO-functionalized surface which have not reacted with the antibodies. Next, a mixture of the two proteins, ovalbumin and β-lactoglobulin (50 μg/ml), is injected at the flow rate of 50 μl/min. The beginning of the injection of the mixture of the two proteins corresponds to the time t=0. Next, the functionalized support is rinsed for 10 min (t=9 min to t=19 min) with the 10 mM ammonium acetate running buffer, pH 7.5. After the interaction data have been recorded by SPRi, the biochip is removed from the SPR instrument and then dried.

(39) The image shown in FIG. 4 illustrates the adsorption of the proteins onto their respective antibodies after 9 minutes of injection. The MALDI-TOF mass spectra are carried out in positive linear mode (100 repeated shots and 500 accumulations; acceleration voltage 25 kV; voltage applied to the grid 93%; extraction delay 450 ns; laser intensity 2800). The chosen MALDI matrix is HABA [2-(4-hydroxyphenylazo)benzoic acid], at 10.sup.−1M in 50/50 water/acetonitrile 0.1% TFA. The matrix is deposited on the spots of antibodies and also on a zone of gold functionalized with PEO-NHS-Lys but without antibodies.

(40) No interaction is detected on the support functionalized with PEO-NHS and neutralized with lysine. The anti-ovalbumin and anti-β-lactoglobulin antibodies present in the example specifically retained 1.8 fmol/mm.sup.2 of protein. The MALDI-TOF mass spectra obtained in this example show a peak corresponding to a mono-charged ion attributed, according to its mass, to the protein specifically retained on its antibody.

Example 4

Comparison of Types of PEO-NHS Vs MUA-CDI Self-Assembled Monolayers Grafted to the Gold Surface of SPR-MS Biochips

(41) This example relates to the comparison between the PEO self-assembled monolayer surface chemistry according to the invention and the surface chemistry described more commonly in the literature in SPR, i.e. functionalization with MUA.

(42) This comparison relates to the capacity for adsorption of β-lactoglobulin onto spots of anti-β-lactoglobulin antibodies placed on the two types of functionalized surface, and also to the level of nonspecific adsorption measured by SPRi on antibody-free zones.

(43) The conditions of the experiment are the following. The interaction of the β-lactoglobulin with the anti-β-lactoglobulin antibody (6 μM) immobilized on the two types of surface is monitored by SPR. At the beginning of the experiment, lysine (100 μM) is injected in order to neutralize the NHS and CDI (carbonyldiimidazole) groups of the surfaces functionalized with PEO and MUA, respectively, which have not reacted with the antibodies. Next, a solution of β-lactoglobulin (50 μg/ml) was injected at the flow rate of 50 μl/min. The beginning of the β-lactoglobulin injection corresponds to the time t=0. The functionalized supports are then rinsed for 5 min (t=9 min to t=14 min) with the 10 mM ammonium acetate running buffer, pH 7.5.

(44) The results are represented in FIG. 2.

(45) It is noted that the zone functionalized with MUA-CDI (FIG. 2C, no antibodies) retains proteins nonspecifically, despite rinsing for 5 minutes. This nonspecific adsorption should therefore be taken into account in the quantification of proteins specifically adsorbed onto the antibody (FIG. 2D). In addition, this nonspecific adsorption may be a source of difficulty during the identification, by mass spectrometry, of the proteins specifically retained on the probes.

(46) In the case of the PEO-based chemistry proposed in the present invention, this adsorption of proteins on the zone not treated with the antibody is completely negligible (FIG. 2A) and does not require any adjustment during the quantification of proteins specifically adsorbed onto the antibody (FIG. 2B).

(47) An MS analysis after an SPRi experiment according to the protocol described above on the PEO-NHS-functionalized and MUA-CDI-functionalized gold surfaces results in the detection of protein on the spots of antibody immobilized according to the two surface treatments. On the other hand, the MUA-CDI-functionalized surface devoid of antibody spots, which was inactivated by lysine treatment, results in the detection of protein by MALDI-MS. These results during the analysis by SPRi followed by a MALDI analysis illustrate the capacity of the MUA-CDI functionalization for creating nonspecific bonds which are detrimental to the analyses.

Example 5

In Situ Proteolytic Digestion During an SPR-MALDI-MS Analysis

(48) Presentation of a Proteomic Analysis after an SPRi Experiment and Analysis by MALDI-TOF MS of the Analytes Digested In Situ, Said Experiment being Carried Out on a Biochip of the Type PEO-NHS Self-Assembled Monolayer at the Gold Surface

(49) This example relates to a tryptic digestion analyzed in-line by MALDI-TOF MS following the capture of the analyte by the immobilized antibody during the SPRi analysis on the functionalized support according to the invention (cf. FIG. 5).

(50) This experiment relates to the ability to access proteomic data subsequent to an in-line SPRi-tryptic treatment—MALDI-MS coupling. This experiment concerns the interaction of ovalbumin and β-lactoglobulin on spots of anti-ovalbumin and anti-β-lactoglobulin antibodies deposited or immobilized on the functionalized surface.

(51) The conditions of the experiment are the same as in example 2. The interaction of the ovalbumin and of the β-lactoglobulin with the anti-ovalbumin and anti-β-lactoglobulin antibodies (6 μM) immobilized on the surface is monitored by SPRi. For this (as in example 2), at the beginning of the experiment, lysine (100 μM) is injected in order to neutralize the NHS groups of the PEO-functionalized surface which have not reacted with the antibodies. Next, 500 μl of a mixture of the two proteins, ovalbumin and β-lactoglobulin, (50 μg/ml), is injected at the flow rate of 50 μl/min. The beginning of the injection of the mixture of proteins corresponds to the time t=0. Next, the functionalized support is rinsed for 10 min (t=8 min to t=18 min) with the 10 mM ammonium acetate running buffer, pH 7.5. Once the SPRi data have been recorded, the biochip is removed from the SPR instrument and then dried. The amounts retained are 40 pg/mm.sup.2 of -lactoglobulin on the anti--lactoglobulin spots and 48 pg/mm.sup.2 of ovalbumin on the anti-ovalbumin spots. The proteolysis is carried out in situ by adding trypsin deposited on the antibody spots and incubating the chip for 1 h in a humid chamber at 37° C. The MALDI-TOF mass spectra are produced in reflectron mode (100 repeated shots and 500 accumulations; acceleration voltage 20 kV; voltage applied to the grid 68%; extraction delay 350 ns; laser intensity 2400). The chosen MALDI matrix is HCCA [-cyano-4-hydroxycinnamic acid], at 10.sup.−1 M in 50/50 water/acetonitrile 0.1% TFA. The matrix is deposited onto the antibody spots. The mass spectra obtained show various signals, among which it is possible to identify those corresponding to the analyte digestion products.

Example 6

Identification of a Protein Captured on the SPRi Biochip, by In Situ Proteolytic Digestion and Sequencing by MALDI-MS/MS Tandem Mass Spectrometry

(52) This example is a presentation of a proteomic analysis by MALDI-MS and MALDI-MS/MS carried out after a SPRi experiment. The sequencing of peptides formed by the in situ proteolysis of a specifically retained protein makes it possible to unambiguously determine the identity thereof, the experiment being carried out on a biochip of the type PEO-NHS self-assembled monolayer at the gold surface.

(53) This example therefore relates to the in-line sequencing of peptides by MALDI-MS/MS following the capture of a protein by the immobilized antibody during the SPRi analysis, and in situ tryptic digestion on the functionalized support according to the invention.

(54) This experiment illustrates the possibility of identifying a protein specifically retained following an in-line SPRi-tryptic treatment—MALDI-MS and MALDI-MS/MS coupling. The example serving to illustrate the principle concerns the interaction between ovalbumin and the anti-ovalbumin antibody immobilized on the functionalized surface (cf. FIG. 6).

(55) The immobilization of the anti-ovalbumin antibody on the surface (prepared according to example 1) and then the interaction of the ovalbumin with the anti-ovalbumin antibody is monitored by SPRi. The ovalbumin (200 μl, 100 μg/ml) is injected at the flow rate of 50 μl/min. Once the SPRi data have been recorded, the biochip is removed from the SPR instrument and then dried. The amount of ovalbumin retained is ˜29 pg/mm.sup.2 on the anti-ovalbumin antibody spots. The proteolysis is carried out in situ by depositing trypsin (0.5 μg/spot) and incubating the chip for 1 h at ambient temperature. The proteolysis reaction medium is maintained by adding, to the spots, twice 1 μl of 0.1 M ammonium acetate buffer, pH 8 (trypsin dilution buffer) every 20 minutes. MALDI-MS and MALDI-MS/MS mass spectra are then recorded from the surface of the same biochip (laser intensity 4900; acceleration voltage Source 1: 8 kV; collision cell 7 kV; Source 2: 15 kV). The chosen MALDI matrix is HCCA [-cyano-4-hydroxycinnamic acid], at 10.sup.−1 M in 50/50 water/acetonitrile 0.1% TFA. The MALDI-MS mass spectra obtained show various signals, among which 9 peptides correspond to the ovalbumin digestion products (m/z 1345.73; 1555.72; 1571.71; 1581.72; 1597.71; 1687.83; 1773.89; 1858.96; 2008.94). Among these ions, the most abundant three were selected for MS/MS experiments (m/z 1555.7; 1687.8 and 1773.9). The fragment ions obtained from each of these peptides made it possible to identify, using the MASCOT software well known to those skilled in the art (http://www.matrixscience.com), the following peptide sequences:

(56) TABLE-US-00001 m/z 1555.7: (SEQ ID NO: 2) AFKDEDTQAMPFR; m/z 1687.8: (SEQ ID NO: 3) GGLEPINFQTAADQAR; and m/z 1773.9: (SEQ ID NO: 4) ISQAVHAAHAEINEAGR.

(57) A data bank (SwissProt 56.2) search using the MASCOT software makes it possible to accurately identify, on the basis of these three peptide sequences, the protein specifically retained on the spots (cf. FIG. 6). The ovalbumin is identified unambiguously (Mowse score 145, sequence coverage 11%).

(58) According to an identical approach, this SPRi-MALDI-MS/MS analysis was repeated with another protein, β-lactoglobulin, specifically retained by anti-β-lactoglobulin antibodies grafted onto the biochip, and then digested directly on the biochip. Eight peptides derived from the digestion were identified by MALDI-MS. On the basis of two of them (m/z 2313.27 and m/z 2707.3), MS/MS experiments followed by a data bank (SwissProt 56.2) search made it possible to identify the corresponding two peptide sequences (VYVEELKPTPEGDLEILLQK (SEQ ID NO:5) and VAGTWYSLAMAASDISLLDAQSAPLR (SEQ ID NO:6), respectively) and, consequently, the protein retained with a Mowse score of 120 (sequence coverage 25%).