Van Eyk, Dunn - Proteomic and Genomic Analysis of Cardiovascular Disease - 2003 (522919), страница 67
Текст из файла (страница 67)
In the case of monoclonal antibodies, some means to attach themolecules via their Fc portions, without affecting the conformation of other regions of the molecule, is likely to be required. If allowed to attach randomly tothe support, a percentage of the antibodies to be immobilised may fortuitouslyalign themselves correctly, and in some cases this has been reported to have littleeffect on overall binding capacity [167]. Sensitivity of target protein detection islikely to be an issue with such antibody arraying conditions indicating that orientated binding may be required. Orientation of immunoglobulins is often achievedthrough the use of Protein A/G that has the required specificity for Fc regions,but such an approach to generating an antibody array does not seem very elegant.24124214 Protein Chip TechnologyHowever, several modifications of this standard immobilisation procedure arebeing investigated.
For instance, Protein A has been engineered to express 5 copies of the immunoglobulin G binding domain plus a cysteine residue at the C-terminus, the latter to allow strong binding to gold immobilisation surfaces [168].Others have exploited the carbohydrate moieties on the Fc portion of the antibodyto orientate them on an Affi-gel matrix [167].Many immobilisation strategies rely on non-covalent interactions between theprotein to be immobilised and the array surface. These may be based on the affinity of the base layer for the PRM, as in the case of Protein A, or on hydrophobic,ionic, or van der Waal’s interactions between the PRM and the absorptive surface[161, 169]. In order to covalently bind the PRMs, metal or silica surfaces arephotolithographically etched and proteins are attached via a chemical linker suchas aminosilane capable of forming covalent bonds with both array surface andPRM [161]. Alternatively, substances such as glutaraldehyde and N-succinimidyl-4maleimidobutyrate may be used as cross-linkers [161, 170, 171].Notwithstanding these considerations, attempts to array proteinaceous PRMshave been initiated successfully [161].
In one such, acrylamide gel has been photopolymerised into a grid pattern using photolithography, followed by transfer ofthe PRM (IgG or BSA) to the gel surface by a multi-pin device in much the sameway as the cDNA microarrays were prepared [172]. In later systems, the acrylamide was replaced by an absorptive protein layer such as streptavidin. The streptavidin was micropatterned using ink-jet printing techniques then exposed to biotinfollowed by the protein to be adsorbed [173]. Alternatively, a streptavidin layer wasoverlaid with photobiotin and the grid generated by exposure to UV light [173].Both of these techniques required multiple layers to be deposited sequentially andwere prone to poor specificity because of cross-contamination of the squares ofthe grid [173].
However, the fact that commercially available ink-jet printers maybe modified to generate arrays on cellulose paper makes them accessible to anyinterested party [174].One of the most recent microfabrication methods to have been proposed forpatterning biomolecules is termed ‘soft-lithography’ [104, 105, 161, 175]. Thiscomprises an elastomeric material or ‘hydrogel’ that can be moulded to formstamps or channels for the transfer of proteins to appropriate surfaces.
For instance, Gaber and colleagues have used drawn-out capillary tubes filled withfreeze-dried disaccharide acrylate polymers that, when exposed to aqueous proteinsolution, re-swell and form a protein-saturated nib [173]. The method is said to besuitable for arraying small, well-defined areas of delicate biomolecules with relative ease [104, 105, 175]. Electrospray deposition may also be effective for arrayingdelicate biomolecules since it only requires a support that is slightly conductive,such as damp membrane, to achieve immobilisation [176]. Whether the arrayingmethods described above will be as versatile and practicable as those currentlyused to generate DNA arrays remains to be seen.14.4 Protein Arrays14.4.2.2 Arraying of Nucleic Acid PRMsMethodologies already exist to array nucleic acids but these are constantly beingupgraded in an attempt to increase the number of targets per unit area, therebyincreasing the data generated per array or chip. The simplest, and therefore probably the most accessible arrays, comprise cDNA inserts or PCR products that havebeen ‘dot-blotted’ onto a nylon membrane of approximately the size of a 96-wellplate.
The DNA binds strongly to the membrane and in a manner that enables itshybridisation to single stranded probes [177]. Preparation of these arrays has beenboth automated and miniaturised so that thousands of cDNAs may be spottedonto a surface the size of a postage stamp [178, 179].The arraying protocol has been modified through the use of glass microscopeslides as supports for the DNA instead of membranes. The glass is silanated andthe cDNA is spotted onto the surface in a grid pattern using a computer-controlled tri-directional robot to which capillary-tipped pens are attached [180].
TheDNA is then immobilised by exposure to ultraviolet light. Glass microarrays suchas these are able to represent between 5000 and 10,000 genes per square centimetre [6, 178, 179].The most sophisticated method developed to date for arraying nucleic acids isthe in situ synthesis of oligonucleotides such that up to 64,000 ‘features’, each containing several million oligonucleotides, may be represented on a small glasschip. The oligonucleotides are synthesized from modified photo-labile deoxynucleosides that polymerise on exposure to ultraviolet light. The sequence of theDNA is directed by masking from irradiation areas of the chip where the availablenucleoside is not required [6, 179].
A recent modification of this technology involves virtual masking in which a computer describes the areas of the chip to beirradiated [181].Each of these technologies could be adapted as a platform for an array for protein expression profiling in which the PRMs were nucleic acid aptamers. The selected nucleic acid sequences could be spotted onto a membrane or synthesized insitu. Membranes are accessible and manageable, but suffer from potential background problems and relatively large surface areas. Glass chips are less susceptible to high backgrounds, would require considerably less target protein mixture,and are more amenable to automation. Thus, the future prospects for immobilising aptamers for protein arrays mirrors the current position for gene expressionarrays.14.4.2.3 Arraying of Synthetic PRMsThe development of MIPs is still at an early stage compared to organic PRMs, yetthey would have many advantages over proteins and nucleic acids in an array format.
There would be no difficulty with array longevity nor would the ability tocapture target be compromised by assay conditions. Combinatorial libraries ofMIPs are already under construction, and the process has been semi-automatedby the use of liquid handling robots [159], and miniaturised to produce microcolumns of 1mm internal diameter [182].24324414 Protein Chip Technology14.5Detection MethodsDetection of the interaction of proteins with recognition molecules on a ‘proteinchip’ in a rapid, highly parallel and sensitive manner is likely to prove very challenging. Many of the methods currently in development are very effective for lesscomplex devices but are probably not applicable to ‘protein chips’ at least as envisaged here.
For example, methods that require a second fluorescently labelled recognition molecule to detect the captured protein are unlikely to prove applicable.Despite the potential drawbacks, development of a fluorescence-based fibre-opticbiosensor in which the capture antibodies are coated directly onto a fibre-opticprobe is interesting [183]. In this approach detection is achieved by application ofa detection antibody labelled with cyanine dye and laser excitation of the fibre optic probe to enable quantification of the bound analyte by fluorimetry. The systemdeveloped by Tempelman and colleagues was able to analyse four samples simultaneously for the presence of a single analyte and had an optimum detectionrange of 5–200 ng/ml for Staphylococcal enterotoxins [183].
Another interestingfluorescence-based detection method has also been described in which an antibody-coated matrix is saturated with fluorescently labelled target antigen. The fluorescent target antigen is displaced by unlabelled target present in the test sampleand the released fluorescent material measured. The method has been used bythe US Federal Drugs Administration to quantify cocaine metabolites in urine[184] and by the Office of Naval Research to detect explosives [185]. Both of thesestudies reported detection limits in the femtomole-picomole range.
Although, inpresent form this would not be translatable to a ‘protein chip,’ one could foreseea system in which the protein recognition molecules arrayed on the surface aredesigned such that they all bind a generic fluorescent moiety that would be released on binding of the individual proteins. The fluorescence remaining at an individual site would then be inversely related to the amount of relevant protein inthe sample mixture.Biosensors, including surface plasmon resonance (SPR)/resonant mirror biosensors, allow real time detection of protein interactions with other proteins or ligands (see [166, 186]).
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
