The discovery of materials for a particular application can be accelerated by using high-throughput methods with which to screen a wide variety of potential candidates. This is useful when developing materials for application in environments where there is limited understanding of the interaction mechanism, e.g. cell adhesion to biomaterials. For this purpose, polymer microarrays containing hundreds of unique materials on a single glass slide are well suited as a means to present a combinatorial material chemistry library.[1-3] Subsequently, surface chemical analysis of the microarray can be performed and correlated with the biological assay data to identify materials for scale up manufacture (Fig. 1).[5, 6] The polymer microarrays produced in our lab are printed from a large monomer library of acrylates and methacrylates containing a wide variety of chemical functionalities. In order to create a chemical combinatorial library, monomers are typically mixed pair-wise, allowing a broad range of materials to be formed. ‘Hit’ compositions can then be selected with a desired biological performance for a second generation microarray where incremental changes in the ‘hit’ compositions are explored.[5-7] This strategy enables the optimal polymer composition to be identified for a given application.
There are several printing methods available for the production of a polymer microarray. Contact printing methods using solid and quilled pins have been previously employed in our lab and others.[2, 8, 9] Another method of printing polymer microarrays is ink-jet printing. This printing method exhibits some significant advantages over other approaches, for example there is no contact with the substrate, the drop volume can be tuned to control the amount of material transferred to the substrate, and each ink-jet nozzle can be easily flushed to avoid contamination when aspirating samples. An increased library of monomers for microarray formation was explored to significantly broaden the number of materials that can be screened. To date, the maximum number of monomers used for microarray formation is 24. However, the quality of polymer spots on the polymer microarray is critical to ensure additional variables such as polymer roughness do not convolute the interpretation of the biological response to the materials. Surface chemical analysis techniques such as time-of-flight secondary ion mass spectrometry (ToF-SIMS) can be used to determine the success of the printing process by analyzing characteristic ions of the substrate and the polymer spots. Specifically, ToF-SIMS imaging enables the rapid assessment of the distribution of chemical moieties across the array. Furthermore, ToF-SIMS determines the surface chemistry of only the top 1–2 nm which is the region of most importance for biological interactions. Previously, ToF-SIMS imaging has been used to determine the identity of particular features within a polymer microarray by composing images of specific characteristic ions.[4, 10] Demonstrated herein is a polymer microarray printed via ink-jet printing that is subsequently imaged with ToF-SIMS to demonstrate the chemical distribution over relatively large areas.
Polymer microarray synthesis
Printing was carried out under an argon atmosphere at O2 < 1300 ppm, 25 °C and 30% relative humidity. Contact printed polymer microarrays were formed using a XYZ3200 pin printing workstation (Biodot). Slotted metal pins (946MP6B, Arrayit) with a tip diameter of 220 µm were used to transfer approximately 2.4 nL of polymerization solution onto pHEMA-coated substrates before slides were irradiated with a long wave UV source for 1 min, resulting in an average polymer spot size of 435 µm. pHEMA-coated substrates were produced by dip-coating epoxy functionalized glass slides (Genetix) into a 4% (w/v) pHEMA (Sigma, cell culture tested) solution in ethanol (95%). Contact printed polymerization solution was composed of 75% (v/v) monomer, 24% (v/v) DMF and 1% (w/v) photoinitiator 2,2-dimethoxy-2-phenylacetophenone. Samples were subsequently dried at < 50 mTorr for 7 days. Monomers were used as purchased from Sigma. Ink-jet polymer microarrays were synthesized onto pHEMA-coated slides using an ink-jet printer (Scienion S11 SciFlex Arrayer) fitted with piezoelectric ink-jet capillaries. Piezo dosing was conducted at 83 – 119 V for 49 – 55 µs depending upon the monomer to obtain a drop volume of 300 pL at 500 Hz which is polymerized in-situ by UV photo-curing to obtain a polymer spot of approximately 300 µm. Ink-jet printed polymerization solution was composed of 10% (v/v) monomer, 89% (v/v) DMF and 1% (w/v) 2,2-dimethoxy-2-phenylacetophenone as photoinitiator.
ToF-SIMS analysis was performed using a ToF-SIMS IV (ION-ToF GmbH) instrument utilizing a 25 KeV Bi primary ion source. Bi3+ primary ions were used with a target current of <1 pA. In order to acquire the data for this relatively large area (6.5 × 13 mm2), a stage scanning function was employed with 8 shots per pixel and a resolution of 100 pixels/mm. Owing to the non-conductive nature of the samples, charge compensation, in the form of a low energy electron floodgun, was applied. Chemical images were reconstructed retrospectively using Surfacelab 6 software.
Results and discussion
To manufacture polymer microarrays that are suitable for a biological assay, the polymers spots must be of a suitable size; this is generally regarded to be approximately 300 µm. This is a balance between having sufficiently large polymer spots to provide a statically representative measurement of material performance and screening the largest number of materials on one slide. Control of spot size can be achieved by altering the individual drop volume and the number of drops ejected from the piezoelectric capillary nozzle to form each spot. When printing a monomer library of a wide range of viscosities and surface tensions, it is difficult to obtain reproducible droplets from the piezo dispenser at the high concentrations typically employed using contact printing. Ink-jet printing is inherently limited to printing low viscosity solutions to obtain reproducible droplets. One method to overcome this limitation is to dilute monomer solutions to lower the viscosity and enable reliable printing. To determine a suitable concentration to print the polymer microarray, two linear arrays of phenyl methacrylate, which is a monomer of low viscosity, were printed with a varying number of drops from 25% and 10% v/v DMF solutions; the optical microscopy images of these are shown in Figs. 2a and 2b, respectively.
The optical images show that from both 25% and 10% v/v DMF solutions, phenyl methacrylate can be printed and polymerized to give circular spots down to 250 µm in diameter. Figure 2a shows a black inner region to the polymer spots that enlarges with the increasing number of drops dispensed from the ink-jet nozzle. This can be attributed to the amount of monomer present as the polymer spots are opaque at higher concentrations. In both Figs. 2a and 2b, polymer spots display a raised outer rim which is due to the ‘coffee ring’ effect of evaporation of the monomer solution on the substrate prior to polymerization. It was determined from these results that the polymer microarray would be printed from 10% v/v monomer solutions in DMF to keep the viscosities as low as possible and print one drop per monomer to obtain polymer spots of approximately 250 µm in diameter. 86 monomers with varied chemistry were chosen for the ink-jet printed polymer microarray. Monomer solutions were aspirated from a source plate and printed on to p(HEMA)-coated glass slides that subsequently underwent polymerization via UV irradiation (monomer identities can be found in Supplementary Fig. S2).
Previously in our laboratory, polymer microarrays have been produced using contact printing under the conditions described in the experimental details. This produces circular polymer spots of approximately 300 µm in diameter. A contact printed polymer microarray consisting of 95 monomers was analyzed as a comparison to the ink-jet printing analogue (monomer identities can be found in Supplementary Fig. S1).
ToF-SIMS imaging was used to characterize both the contact printed polymer microarray and the ink-jet printed polymer microarray. The distribution of chemistries across the area within the polymer microarray was displayed using images of specific ions that are characteristic of particular monomers. The C2H5O+ ion image and the positive total ion image of the contact printed polymer microarray are presented in Figs. 3a and 3c, respectively. The C2H5O+ ion image was chosen over other ion species as it is characteristic of the p(HEMA) substrate and therefore provides good contrast against printed monomers and can be used to rapidly evaluate the presence of printing irregularities or chemical spreading. Both the C2H5O+ ion and positive total ion images highlight chemical spreading that was not apparent in the optical image of the contact printed polymer microarray (Fig. 3b). This particular polymer microarray would not be appropriate for a biological assay to screen for materials that support cell growth as the chemical spreading could support cell binding to regions where chemistries from different polymer spots have overlapped, thus leading to false ‘hit‘ materials. This emphasizes the importance of discrete polymer spots within a polymer microarray when screening for novel materials for biomedical applications.
The C2H5O+ ion image and the negative total ion image of the ink-jet printed polymer microarray are presented in Figs. 4a and 4c, respectively. Figure 4a shows that all polymer spots are discrete and no spreading of the chemistry had occurred prior to polymerization. Furthermore, the polymer spots are in the range of 250 – 400 µm in diameter, which is comparable to previous polymer microarrays produced using contact printing methods.[2, 7] Moreover, polymer spots that had spread in the contact printed array did not exhibit the same behavior using ink-jet printing which broadens the combinatorial space to include such monomers. The high velocity of the monomer drops fired from the piezoelectric mechanism leads to polymer spots which are flatter compared to polymer spots that have been contact printed. Therefore, the optical image of the ink-jet printed polymer microarray provided poorer contrast for optical microscopy (Fig. 4b). It is worth noting that not all monomers were successfully printed using ink-jet printing as shown by blank spaces in Fig. S2. This can also be seen in Figs. 4a and 4c and can be attributed to the presence of air bubbles within the ink-jet nozzle which can disrupt monomer droplet formation during printing.
Polymer microarrays have been synthesized using contact and ink-jet printing methods. These arrays were subsequently imaged using ToF-SIMS and optical microscopy. ToF-SIMS imaging showed that chemical spreading had occurred within the contact printed polymer microarray prior to polymerization. However, no such spreading was observed in the ink-jet printed polymer microarray which gave circular and discrete spots of 250 – 400 µm in diameter. This demonstrates a novel method of assessing polymer microarrays for their use in biological assays and highlights the utility of ToF-SIMS imaging as a tool for the analysis of large surface areas.