Apparatus and method for sample preparation and direct spotting eluants onto a MALDI-TOF target

Abstract

A single- or multi-well sample preparation apparatus and method for desalting, concentrating and depositing samples prior to further analysis such as by MALDI TOF mass spectrometry. The apparatus in accordance with an embodiment of the present invention includes a plurality of wells each in fluid communication with a respective outlet or drainage opening, optionally containing a three dimensional membrane structure preferably comprising a plurality of sorptive particles entrapped in a porous polymer matrix so as to form a device capable of carrying out solid phase extraction. The apparatus is designed to allow for direct spotting onto a MALDI target, thereby eliminating a transfer step. Also disclosed is a method of sample preparation, deposition and analysis using the apparatus of the present invention.

Description

BACKGROUND OF THE INVENTION

Matrix-assisted laser desorption/ionization (MALDI) analysis is a useful tool for solving structural problems in biochemistry, immunology, genetics and biology. Samples are ionized and a time of flight (TOF) analyzer is used to measure ion masses. TOF analysis begins when ions are formed and are accelerated to a constant kinetic energy as they enter a drift region. They arrive at a detector following flight times that are proportional to the square root of their masses. A mass spectrum is created because ions of different mass arrive at the detector at different times.

Mass spectrometry can be a particularly powerful tool in the fields of drug discovery and development, genotyping, and proteome research. Current trends in research are to analyze larger and larger numbers of samples using automated handling equipment or robotics. Quantities of individual samples are from the nano-mole levels to atto-mole levels. As a result, instrumentation is becoming more sensitive and a need exists for sample handling formats to be miniaturized, high density and disposable.

Sample preparation prior to analysis (such as by MALDI TOF MS) often involves desalting and concentration of samples (e.g., peptides) down to a 1-2 microliter volume. These volumes are likely to decrease to nanoliter volumes in time. Simultaneous preparation and analysis of multiple samples is often desirable. Multiwell plates have been developed for simultaneous assay, typically consisting of 96, 384 or 1536 reaction vessels or wells per plate.

Certain sample preparation devices, such as the ZipTip® device commercially available from Millipore Corporation, are excellent tools for sample preparation prior to MALDI analysis. They are a single sample processor that can be used to spot sample onto the MALDI target manually or by automated equipment. More specifically, U.S. Pat. Nos. 6,048,457 and 6,200,474 teach the formation of cast membrane structures for sample preparation that are formed by phase inversion of a particle loaded polymer system at the housing orifice. The polymer is precipitated when the housing (containing the soluble polymer/particle lacquer) is immersed in a precipitation bath (typically water). The insertion creates a slight liquid pressure across the lacquer such that the water intrudes upon the polymer creating an open sponge-like structure upon precipitation. However, at the polymer-water interface on the structure there is a semipermiable membrane film that creates a high resistance to flow. When this barrier is either abraded of cut off, the resulting structure is highly permeable. The resulting device is designed to allow flow under the low differential pressures generated by a common 10 microliter hand-held pipettor (e.g. Gilson, Pipetman).

For high throughput sample processing, it would be desirable to use multiwell plates for sample handling and preparation, such as the removal of undesired salts and biochemical substances to improve the resolution and sensitivity of the mass spectrum. However, evaporation of elution solvent can be problematic. The protein and peptides need to be in as small a volume as possible to obtain an adequate MALDI TOF mass spectrum. Collection of the elution volume by centrifugation is possible but difficult, since the volume in each well may vary due to rapid evaporation during transfer of the multiwell plate to the centrifuge and especially during centrifugation. Also, eluants conventionally collected by vacuum methods tend to evaporate rapidly under negative pressure, thus requiring resuspension. Moreover, every time the sample is transferred, such as from pipette to collection plate, or is resuspended, sample is lost to due adherence to the interfaces of these devices. Since sample amounts are typically in the femotmole range, sample losses are unacceptable. Furthermore, centrifugation is also not amenable to automation, as the plate must be manually placed and removed into and from the centrifuge.

A key to achieving high sensitivity and strong MALDI signals is by eluting the bound peptides in as high a concentration as possible. This can be accomplished by using minimum elution volume and reducing handling steps.

It would be highly desirable to use the microtiter plate format for parallel sample preparation that is readily adaptable to automation.

It is therefore an object of the present invention to provide a device and method for spotting of a small volume of eluant directly from the sample preparation device onto a MALDI target.

It is a further object of the present invention to provide a device and method for direct eluant spotting of relatively high concentrations of sample onto a MALDI target.

These and other objects will become apparent by the following description.

SUMMARY OF THE INVENTION

The problems of the prior art have been overcome by the present invention, which provides a single- or multi-well sample preparation apparatus and method for desalting, concentrating and depositing samples prior to further analysis such as by MALDI TOF mass spectrometry. More specifically, the apparatus in accordance with an embodiment of the present invention includes a plurality of wells each in fluid communication with a respective outlet or drainage opening, optionally containing a three dimensional membrane structure preferably comprising a plurality of sorptive particles entrapped in a porous polymer matrix so as to form a device capable of carrying out solid phase extraction. The apparatus is designed to allow for direct spotting onto a MALDI target, thereby eliminating a transfer step.

The present invention is also directed towards a method of sample preparation, deposition and analysis using the apparatus of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a perspective view of a single well of a multi-well device, with the composite structure shown in expansion in Detail A;

FIGS. 2A, 2B, 2C and 2D show four steps of the elution and transfer process in accordance with the present invention;

FIG. 3 is a graph comparing % sequence coverage in various elution techniques;

FIG. 4A is a spectrum of a β-galactosidase sample eluted into a microtiter plate using vacuum in accordance with the prior art;

FIG. 4B is a spectrum of a β-galactosidase sample eluted using centrifugation in accordance with the prior art;

FIG. 4C is a spectrum of a β-galactosidase sample eluted directly onto a MALDI target in accordance with the present invention; and

FIG. 5 is a cross-sectional view of a well positioned over a MALDI-TOF target substrate and coupled to a vacuum manifold in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

Suitable substrate materials for the sample preparation device of the present invention are not particularly limited, and include plastics (such as polyethylene and polypropylene), glass and metal, such as stainless steel. The substrate materials should not interfere with the operation of the device or the chemicals to be used in the procedure. Polyolefins, and particularly polypropylene, are preferred materials.

The term “membrane” as used herein includes permeable and semi-permeable three dimensional structures with or without particles, having a porosity suitable for the desired application. The term “composite structure” as used herein includes filled membranes.

Turning now to FIG. 1, there is shown a single well 12 of a sample preparation device that has a plurality of wells, such as 384. A well 12 is defined by a vertically extending fluid impervious side wall 14 and a sloping bottom portion 13. The middle and upper portions of the well 12 preferably have a uniform diameter and are substantially cylindrical in cross-section, although other configurations are contemplated and within the scope of the present invention. The lower portion of the well 12 preferably tapers downwardly, in the direction of fluid flow, towards a bottom portion 13, which slopes inwardly towards a center, thereby having a frusto-conical configuration. Bottom portion 13 has a spout 15 that is preferably centrally located in the well. As best seen in Detail A of FIG. 1, the spout 15 is a bore, preferably cylindrical and axially aligned with the central longitudinal axis of the well 12. The dimensions of the bore determine the dimensions of the membrane structure contained therein, which determine, in part, the characteristics of the droplets that form upon flow through the membrane structure. Suitable bore diameters include from about 0.2 to about 2 mm, more preferably from about 0.4 to about 0.8 mm, with a diameter of 0.5 mm being preferred. Suitable bore heights include from about 0.2 to about 2 mm, with 1 mm being preferred. At least a portion of the spout 15 preferably includes an adsorptive membrane structure 25. Suitable adsorptive membrane structures are cast-in-place polymer bound, particle laden adsorptive membrane structures, such as those comprised of chromatographic beads which have been adhered together with a binder and disclosed in U.S. Pat. Nos. 6,048,457 and 6,200,474, the disclosures of which are hereby incorporated by reference. One such preferred structure is a three-dimensional structure comprising a plurality of sorptive particles entrapped in a porous polymer matrix and having an aspect ratio (average diameter to average thickness) of less than about 10, preferably less than about 5. The structure 25 may be coterminous with the bottom of the spout 15, and extends into the body of the spout 15, preferably extending through substantially the entire depth of the spout 15.

Devices in accordance with the present invention may incorporate a plurality of membrane structures having resin materials with different functional groups to fractionate analytes that vary by charge, size, affinity and/or hydrophobicity; alternately, a plurality of devices containing different individual functional membranes may be used in combination to achieve a similar result. Similarly, one or more membranes can be cast in a suitable housing and functionality can be added before or after casting.

Preferably the membrane structure 25 is located at the distal end of the drain 15, and has a volume of about 300 nanoliters. The drain preferably has a small internal diameter, such as about 0.5 millimeters, so that the membrane structure is relatively small and therefore requires less elution volume. In the preferred embodiment where the structure 25 is coterminous with the bottom open end of the spout 15, sample dilution is minimized due to the reduction or absence of deadspace.

In order to minimize elution volume and deposit the sample (i.e., “spot”) efficiently on the target substrate, the spout 15 must be kept in close proximity to the target. A molded stand-off collar or skirt 30 partially or completely circumscribing each spout 15 is formed in the device to support the device on the substrate and maintain a suitable gap or distance “x” between the outlet of the spout 15 and the target surface 50. Preferably this gap is smaller than the diameter of the liquid drop 51 that forms as the eluant transfers from the membrane structure 25 to the target surface 50. Thus, as the drop forms from the spout 15 as shown in FIG. 2B, it touches the target surface 50 (FIG. 2C) before it releases from the spout 15. When this occurs, there is increased surface tension adhesion on the target surface (due to, for example, the surface area difference and relative hydrophobicity of the target surface) that “pulls” the drop off the spout. The maximum dimensional offset between the spout outlet and the target surface 50 depends in part on the elution volume and the nature of the elution solution. A suitable gap “x” for a 1 microliter elution is about 0.15 to about 0.020 inches (about 0.5 mm), with a maximum gap of about 0.035 inches. A suitable gap for a 2 microliter elution is from about 0.020 to about 0.030 inches, with a maximum of about 0.05 inches. Gaps exceeding the maximum do not allow for the effective transfer of eluant in a minimum (or a reasonable amount) of volume. If the gap is too small, the transfer will occur, but the spots tend to be large and irregular because the drop does not fully transfer; it fills the gap and can bubble if air flows through the structure. The minimum gap is such that an elution droplet formed contacts the target surface 50 and releases from the spout outlet leaving a gap, so that the elution droplet is not disrupted by air flow through the spout. Once an effective transfer is made, the spots 52 on the target surface dry quickly under negative pressure, as depicted in FIG. 2D.

In the embodiment shown, the collar 30 is annular and conveniently defines a volume bounded by the bottom of the well and the target surface 50 that allows vacuum to reach the spout 15. One or more vents 54 are formed in the collar 30 for this purpose. Those skilled in the art will appreciate that posts or other structures could be used to create the gap and ensure that the spout can receive negative pressure.

Suitable substrates or targets are those conventionally used in MALDI TOF mass spectrometry. They are substantially planar, conductive, and are dimensioned to fit in ionization chambers of the MALDI instrument. Metallics such as stainless steel are typical.

In its method aspects, the present invention utilizes the evaporation problem discussed above as a processing solution, and eliminates a transfer step previously necessary when using conventional methods. To that end, sample is introduced into one or more wells of the multi-well plate by any suitable means, such as by pipetting. The molecules of interest are captured by the membrane structure 25 present in each well. A wash step is optionally carried out. As illustrated in FIGS. 2A-2C and 5, the plate is positioned on a vacuum manifold 60 (sealed with sealing gasket 61) and over a planar MALDI target substrate, for example, appropriately positioned below the spout outlet. To elute the molecules of interest from the membrane structures, vacuum (preferably about 5 inches Hg) is applied to each well, preferably to create a pressure differential of about 2-6 psi, and an elution solvent (about 1-2 microliters) is introduced into each well. Too high a vacuum tends to cause bubbles or spraying of the elution liquid, yielding poor spotting on the target surface. A suitable elution solvent such as an acetonitrile/matrix mixture, preferably a 50% acetonitrile/0.1% TFA mixture can be used, and vacuum is applied. The elution volume exits the spout 15 and contacts the target positioned beneath the spout and rapidly evaporates on the target, leaving the sample crystals ready for analysis in a convenient array (corresponding to the array of wells 12) such as by MALDI. Since a transfer step is eliminated, losses are minimized and sample processing time is reduced. Crystal formation is excellent, and MALDI signal sensitivity is enhanced.

EXAMPLE 1

One method of identifying an unknown protein is to digest it with ca. bovine trypsin generating a unique set of peptides. The collective masses of these peptides as determined by mass spectrometry (e.g. MALDI TOF MS) represent a fingerprint that can be searched against a database. The quality of the database match can be assessed by several complex-scoring systems. However, one simple means of scoring is the amount of protein sequence that can be identified by the mass spectrum. This parameter is typically referred to in the field as % sequence coverage or % coverage. In most cases, with a high performance MALDI TOF MS system that is accurate to 50 ppm of a mass unit, it is possible to identify a protein with as little as ca. 12% of its sequence.

FIG. 3 shows the sequence coverage obtained from β-galactosidase (E. coli) samples (50, 100 and 200 fmol) that were digested with bovine trypsin, transferred to a MALDI TOF MS target by 3 different means and analyzed. For the “vacuum indirect” method, the sample was desorbed from the plate in 15 microliters of volume (50% acetonitrile containing MALDI matrix, e.g. α-cyano-4-hydroxy-cinnamic acid) using a vacuum manifold (at 5 inches Hg) into a 96-well “V” bottom polypropylene microtiter plate. 15 microliters are required to form a sufficiently large drop that can fall off the spout by gravitational force. (Volumes less than this amount typically are held on the spout by surface tension.) From the collected volume (typically 10 microliters or less), two microliters were transferred by pipette to a MALDI TOF MS target and allowed to dry. This method provides acceptable sequence coverage with 200 fmol of sample. However, % coverage is virtually non-existent at lower peptide levels. Improved sensitivity for “indirect” transfer can be achieved by using less eluant volume (ca. 2 microliters). Centrifugal force (ca. 1500 Xg for 15 seconds) must be used to efficiently drive the small volume through the membrane and then the entirety of the collected volume is spotted onto the target. Acceptable % coverage can be obtained on as little as 50 fmol of protein. Although the performance of the method is good, due to the need for centrifugation, it is not suited for automation and would be not be entirely useful for high-throughput analyses.

Direct transfer (or spotting) from the sample preparation device to the MALDI TOF MS target, using vacuum, is preferred because: 1) it eliminates a handling step, 2) requires a minimum volume and 3) is fully automate-able. As can be seen in FIG. 3, this method (Vacuum Direct) provides % coverage results comparable to the “centrifuge indirect” method.

EXAMPLE 2

Comparative MALDI TOF MS Spectra of β-Galactosidase Tryptic Peptides

Three 50 fmol samples of β-galactosidase (E. coli) were digested with trypsin, bound to the membrane within the spout and eluted by different methods. FIG. 4A is the spectra obtained when the membrane was eluted into a microtiter plate well with 15 microliters of 50% acetonitrile containing MALDI Matrix using vacuum (5 inches Hg) and then spotted (2 microliters) onto a MALDI TOF MS target. FIG. 4B was obtained by eluting the membrane with 2 microliters of 50% acetonitrile containing MALDI Matrix using centrifugation (15 seconds@1500 Xg) and then spotted (2 microliters). FIG. 4C is a spectrum of a well that was eluted/spotted (2 microliters) by vacuum (5 inches Hg) directly onto the MALDI TOF MS target in accordance with the present invention. FIG. 4C shows coverage of 23%, compared to 20% using centrifugation and virtually no coverage with indirect vacuum.

Claims

1. A sample preparation and presentation assembly for preparing and presenting a sample having molecules of interest for analysis, comprising: a sample preparation device comprising a plurality of wells, each of said wells containing a membrane structure for binding said molecules of interest, each of said wells comprising a spout having an outlet and a collar at least partially circumscribing said spout; a vacuum manifold for eluting said bound molecules of interest from said membrane structure; and a conductive sample presentation substrate having a substantially planar surface positioned to support each said collar and directly receive said eluted molecules of interest from said sample preparation device, and adapted to present said eluted molecules of interest for MALDI TOF mass spectrometric analysis.

2. The assembly of claim 1, wherein said plurality of wells are arranged in an array.

3. The assembly of claim 1, wherein said substrate is positioned beneath said plurality of wells at a predetermined distance to form a gap, devoid of structure, between said spout outlet and said substrate.

4. The assembly of claim 3, wherein said elution causes drops to form on said spout outlet, each said drop having a diameter, and wherein said predetermined distance is less than said diameter.

5. The assembly of claim 3, wherein said predetermined distance is from about 0.015 to about 0.030 inches.

6. The assembly of claim 1, wherein said collar comprises one or more vents.

7. The assembly of claim 1, wherein said membrane structure comprises a plurality of sorptive particles entrapped in a porous matrix.