The first step in the production process of the DiaCHIP was to genetically fuse antigen coding sequences to a 10xHis Tag that was used for surface immobilization later. The whole expression cassette including promoter and terminator regions was amplified by PCR using an amino-labelled reverse primer. Via this amino group, the DNA was immobilized on an activated PDMS surface. The forward primer used for this PCR was labelled with Cy3. As it is shown in figure 1, spotting the DNA on the activated surface resulted in a distinct pattern visualized by Cy3 fluorescence.

Cell-free expression

To establish a cell-free expression system, a bacterial lysate was produced and supplemented with several energy sources as well as co-factors and ions. GFP expression in our system was compared to a commercially available kit. In figure 2, the relative fluorescence of the samples is shown over time. After two hours the fluorescence indicates a x fold change of the amount of GFP in the sample compared to the beginning of the reaction. The commercially available kit reaches an x fold increase in relative fluorescence.
Additionally, it was shown that the expressed GFP is not only functional in terms of fluorescence but it also exhibits the same binding affinity to a commercial anti-GFP antibody as conventionally purified GFP (figure 3). Thus, our cell-free expression system can be used to mediate the copying process from a DNA template to a protein microarray.

Figure 2: Comparison with self-made cell-free expression system and a commercially available kit. The relative fluorescence of cell-free expressed GFP is monitored over 2 hours in comparison with a commercial kit. For each system a no-DNA control was added enabling to calculate the fold change.

Invalid Link

Figure 3: Western Blot of conventionally and cell-free expressed GFP.

Specific surface

Using a cell-free expression system requires the establishment of a specific surface that prevents the binding of non-target proteins. A stable protocol for the production of Ni-NTA surfaces was developed. The 10xHis Tag complexes nickel ions on the surface resulting in a sufficiently strong binding of the protein. In figure 4, an unspecific PDITC surface and a Ni-NTA surface are compared. Flushing the slides with the same antibody solution resulted in a weaker signal for PDITC, which is explained by the binding of many non-target proteins blocking the surface for the His-tagged protein.
Additionally, we have shown that cell-free expressed and therefore non-purified protein can be efficiently immobilized on the surface in a sufficient amount to detect antibody binding by iRIf (figure 5). Figure 6 shows that several different antigen spots can be distinguished by flushing the slide with antibodies that are specifically binding to just one of them. Only the spot where the antigen corresponding to the used antibody is immobilized exhibits a signal.

Invalid Link

Figure 4: Quotion picture of an iRIf meassurement of non-purified, His-tagged GFP(?) on PDITC compared to Ni-NTA. Whole cell lysate was spotted either on an unspecific PDITC surface (A) or on a specific Ni-NTA surface (B) and flushed with anti-GFP(?). The quotion picture shows the change in thickness at distinct spots related to the rest of the surface.

Figure 5: Binding curve of anti-GFP to cell-free expressed GFP on a Ni-NTA surface. Cell-free expressed GFP was spotted on a specific Ni-NTA surface and flushed with anti-GFP. The binding curve indicated a binding event a certain spot.

Diagnostics

These results can be used to establish a diagnostic device based on the detection of specific antibodies in a patient’s blood sample. We were able to express a specific antigen for Salmonella Typhimurium and immobilize it on the surface. Flushing the slide with a self-purified single chain variable fragment specifically binding to the antigen resulted in strong signal at the Salmonella spot, whereas no signal was seen at the negative control spot (figure 7).
Another antigen that was expressed is a specific marker for Clostridium tetani. We obtained serum samples of a person before and after vaccination against C. tetani and showed that the DiaCHIP can be used to detect antibodies against the bacterium in the positive serum sample. In contrast, no binding could be observed in the negative sample (figure 8).

Figure 7: iRIf meassurement of spotted Salmonella Typhimurium antigen. The self-purified single chain variable fragment was used to detect an antigen specific for Salmonella Typhimurim spotted on a Ni-NTA surface. A: Binding curve. B: Quotion picture.

Figure 8: iRIf meassurement of spotted, cell-free expressed C. tetani antigen. Cell-free expressed tetanus antigen was spotted on a Ni-NTA surface and flushed with a human serum sample before (A and B) and after vaccination (C and D). A, C: binding curves. B, D: quotion pictures.

Conclusion

All in all, we established a method for the production of a protein microarray by copying a DNA template combined with label-free detection of antibody binding events. We showed that DNA can be immobilized on activated PDMS and serve as a template for cell-free expression. Cell-free expressed protein is immobilized on another surface via a specific tag system. Additionally, we were able to distinguish several immunologically relevant antigen spots by specific antibody binding out of a human serum sample.