Molecularly Imprinted Polymers: synthetic receptors for diagnostic medical devices

Abstract

Irritable Bowel Syndrome (IBS) is a functional bowel disorder which is characterized by abdominal pain and changes in bowel habit. It has a profound negative impact on the quality of a patient’s life and is associated with more than €20 billion of indirect and direct medial costs per year. Most IBS therapies are based on relief of the symptoms, since the organic cause of the disease remains largely unknown. Recently, the involvement of biogenic amines such as serotonin and histamine has been postulated. The established technologies to determine the concentration of these target molecules have several disadvantages. The methods are costly, lack in speed, and require a labenvironment and sophisticated equipment. Therefore, the aim of this thesis is to develop a polymer-type sensor for the detection of serotonin and histamine for intestinal –and blood applications. The sensor platform should meet the following criteria: specific recognition in biological samples, low-cost (~10 euro per materials of the chip), fast response time (30 – 60 min), and offering the possibility of measuring in vivo. Molecularly Imprinted Polymers (MIPs) were used as polymer-type receptors since they are robust, can be produced at low-cost, and have a high affinity for their template molecules. They were synthesized by bulk polymerization and subsequently ground to obtain a powder. Next, aluminum electrodes were functionalized with MIPs by thermal treatment. The particles were then integrated into a sensor platform which can specifically detect small molecules by two read-out technologies. The first technique is based on electrochemical impedance spectroscopy, the second on heat-transfer resistance. First, a MIP for the specific detection of serotonin was developed. Various MIPs were synthesized and the MIP with the highest affinity for serotonin was selected by optical batch-rebinding experiments. The particles were then integrated in an open impedimetric sensor setup and a dose-response curve was determined in buffer solutions. For biological samples, a refined sensor cell was developed. This flow through cell was closed, ensuring safe administration of patient’s samples, and featured an integrated temperature unit which improves stabilization and control over the system. With this setup, native serotonin concentrations in human blood plasma were determined. The obtained results were in agreement with High Performance Liquid Chromatography (HPLC) reference tests. Furthermore, it was demonstrated that the impedimetric response upon binding of serotonin can be attributed to a capacitive effect at the interface between the MIP particles and fluid layer. Histamine is the other biogenic amine of interest in this thesis. In previous work, histamine concentrations in buffer solutions ranging from pH 7-9 were determined with impedimetric read-out. For diagnostic applications, this target molecule should be detectable in a wider pH range as is it mostly occurs in mildly acidic environments. To understand the pH-dependent response of the MIP sensor, we proposed a statistical binding analysis model. With this model, the theoretical performance of a MIP based on the monomer acrylic acid in the required pH regime was predicted. The results were verified experimentally by UV-vis spectroscopy, microgravimetry and impedance spectroscopy. Histamine could be detected with impedimetric read-out in the physiologically relevant nanomolar concentration range in neutral and in mildly acidic buffer solutions. As last validation step, this platform was used to analyze histamine concentration of mildly acidic bowel fluid samples of several test persons. It was shown that this sensor provides reliable data in the relevant concentration regime, which was validated independently by Enzyme-Linked Immunosorbent Assay (ELISA) tests. The electronic method requires minimal instrumentation, can perform fast measurements and is specific enough to determine concentrations in the physiologically relevant environment. An alternative read-out technique is also presented, the heat-transfer method (HTM), which is even more straightforward as it requires only two thermometers, an adjustable heat source and a Proportional-Integral-Derivative (PID) controller. The principle works as follows; upon rebinding of a target molecule, the heat flux through the nanocavity in the MIP is strongly reduced due to the presence of the template. As a result, the total heat transfer resistance (Rth) will be increased with the effect size being dependent on the amount of target molecule that is bound. For proof-ofprinciple purposes, aluminum electrodes were functionalized with MIP particles and L-nicotine measurements were performed in buffer solutions. The main focus of this thesis lies on serotonin and histamine; therefore dose-response curves in buffer solutions were also constructed for these target molecules. The achieved detection limits in buffer solutions are comparable as with impedance spectroscopy. This technique was applied simultaneously with the Rth and provided a direct validation of the results. As a proof-of-application, measurements were performed on saliva samples spiked with L-nicotine. A dose-response curve could be constructed, showing the applicability even in complex matrices. Next, it is evaluated to what degree the developed techniques fulfill the proposed criteria in terms of specificity, costs, speed, and possibility of in vivo measurements. The LOD of both techniques, compared to the physiologically relevant concentrations shown in Table 1, is low enough to specifically detect histamine and serotonin in buffer solutions. However, only the impedimetric read-out has been studied extensively with biological samples. Therefore, further clinical trials will be performed with this method. It has to be considered that HTM is not an established technology yet and there is still room for improvement, especially on reducing the noise on the signal. In terms of expenses, the main advantage of MIPs is that they are low cost due to their straightforward synthesis. From this we can conclude that the chemicals do not play an important role for the material costs of the chip. The impedance analyzer and sensor cell are all home made, with a total of around 1000 euro for the equipment. For HTM only two thermometers and an adjustable heat source are required, ensuring this is even less expensive. The material value of 10 euro per chip seems therefore reasonable. The other criterium is the measurement time, which should be within 60 minutes. For our experiments, the average value is taken of six points with an interval of three minutes. This means that at least the signal should be monitored for 18 minutes, with aditionally a stabilization time between 15 to 30 minutes. Therefore, it should be possible to perform a measurement in the required time frame. Both techniques are specific, low cost and fast, thereby fulfilling the specified criteria to a great extent. The last criterium, the in vivo measuring for intestinal applications, imposes some challenges. To overcome these problems, the sample preparation needs to be altered possibly by directly polymerizing the MIPs onto the surface of the electrodes. Furthermore, the biocompatibility of the MIPs and the sensor platform should be evaluated. The final step is to enter the industrial market of diagnostics. For this purpose, some future MIP platform designs were suggested. There are two more aspects we will have to consider, before thinking of commercialization of the product: i) A clinical trial should be performed to study a larger amount of biological samples and, from a diversity of individuals (healthy controls and patients). The latter is indispensable to determine if the technique is sensitive enough to detect occurring aberrations. ii) The sensor platform should be transformed to an array format which enables the simultaneous detection of a variety of targets. Due to the speed, low-cost, and high specificity of the developed techniques, they can be considered as methods with a high clinical and commercial potential. These methods could also mean a first step into the direction of incorporating MIPs into sensing devices for in vivo purposes. Measurements performed directly in the gastro-intestinal tract would provide an important insight into the pathogenesis of several gastrointestinal system-related disorders. This may lead to the development of novel, effective treatments for these disorders.