In vitro testing of thick film and co-fired implantable ceramics with high-density printed and laser patterned Pt/Au electrodes: cytocompatibility, electrical performance, and protein adsorption

Abstract

Introduction: High density metal feedthroughs for implantable electronics are possible using high temperature co-fired ceramic (HTCC)[1],[2]. The biocompatibility of custom HTCC materials has been demonstrated in vitro[3]. For low costs or large scales, commercial materials may be preferred. This technology will make miniature hermetic packages possible for implantable electronics[4]. We hypothesise that commercial conductor pastes do not significantly reduce cell viability in vitro. Materials and Methods: Commercial Pt/Au paste (5837G, ESL) and Pt paste (5571, ESL) were used to manufacture thick film (fig. 1A) and HTCC (fig. 1B) electrodes respectively. ISO 10993-5 extract tests[5] were carried out on L929 fibroblasts (n=5 , 85011425, Sigma) followed by live/dead , Alamar Blue and MTT assays. We used samples at intermediate and final manufacturing stages, and phenol negative, and HDPE positive controls. Contact assays used PC-12 cells (n=5, 88022401, ECACC), 4.1/F7 Schwann cells (n=3, 93031204, ECACC), and primary rat dorsal root glial cells (n=3). The effect of stimulation during cell culture was assessed (1 kHz, 220 µA biphasic pulses). Cell response was assessed using IF microscopy. The effect of pre-treatment with fetal calf serum on cell response and protein adsorption (n = 3) were investigated. Results: L929 cell viability did not reduce with thick film extract compared with positive controls (p ≥ 0.054), and viability was significantly increased with Pt HTCC extract compared with positive controls (p ≤ 0.02). PC-12 cell response did not vary with electrode pre-treatment or stimulation on thick film (fig. 2A, p ≥ 0.199). On HTCC there were a significantly different cell responses with different electrode layouts and stimulation (fig 2B, p ≤ 0.001). 4.1/F7 cell response on HTCC samples was similar to or greater than on glass controls; except for at the electrode site with stimulation, where cell density was significantly lower (fig 2C, p ≤ 0.001). Primary glial cell number was lower on HTCC samples than on glass controls in all cases (fig 2D). Serum proteins adsorbed at rough electrode sites on thick film samples (fig 2E) and across the sample surface on HTCC samples (fig 2F). SDS-PAGE demonstrated that a ~70 kDa protein adsorbed to the samples, which was attributed to albumin. Additional bands were present at ~55 kDa and > 100 kDa. Discussion: Survival and growth of nervous system and neuronal-like cells was demonstrated on thick film and HTCC electrodes. Stimulation can reduce cell density, indicating possible damage in response to electrical current, in particular inhibition zones were observed around some active electrodes. Protein adhesion at stimulation sites may alter cell response[6], this would be of particular importance where stimulation occurs soon after implantation. Conclusion: The cytocompatibility of commercial conductive metal pastes in thick-film and HTCC processing was demonstrated to a superficial level; however, stimulation does affect cell response. The application of these technologies to implantable electronics will allow the development of miniature hermetic feedthroughs direct from electronics-to-body. We would like to thank Rebecca Porter and Dara McCreary for technical assistance.; HL would like to thank the EPSRC (UK) for funding through the M3S Doctoral Training Centre (EP/G036675/1) and the UCL Doctoral Prize Fellowship schemes.; AV would like to thank the Aspire Charity (UK), University College London and the Royal National Orthopaedic Hospital Trust (UK) for funding through Aspire CREATe.