Revision as of 23:48, 1 November 2017

Design

The studying of our model resulted in the discovery of 3 ion channels suited for our project. More information on these ion channels is mentioned below. The next step was measuring the rhythm of our transfected cells as well as the influence certain drugs have on this rhythm, With the results from the Patch Clamp showing indeed a change in frequency of the rhythm, the team started looking at the future. We started brainstorming about the theoretical design if our device. Further research brought the encapsulation method to light, the measurement of the rhythm in the human body would be done through a Multi-Electrode Array (MEA), as suggested by the nanotech company IMEC. More information about the in-vivo concept could be found below.

Alpha-1-G – A voltage-sensitive calcium channel

The fast voltage-sensitive calcium channel, α1G, is activated by depolarizations that reach a threshold of -65 mV. As a result, there is a rapid influx of calcium ions which will depolarize the membrane even further, until all the ion channels deactivate after a couple of milliseconds. The channels will reactivate when repolarizing under the activation threshold, preparing itself for the next cycle.

hERG - A voltage-sensitive potassium channel

The voltage-sensitive potassium channel, hERG, becomes activated around a slightly more positive threshold then α1G. It allows for an outflux of potassium ions, which causes the membrane potential to drop again. Its activation speed is also slower than α1G, which is why the membrane potential rises first due to α1G, followed by steep decline mediated by hERG just a fraction of a second later.

HCN2 - A slower voltage-sensitive sodium channel

HCN2 is peculiar, since it becomes activated at a potential below -40 mV. The more negative the membrane potential gets, the bigger the ion flow that passes through which causes a steady depolarization. [THESE TEXTS COME IN THREE COLUMNS WITH ] In theory, you could design a spiking cell with only channels such as α1G and hERG. We hypothesize however, that you need at least one ion channel of the HCN type to create a stable rhythm.

Ivabradine

Ivabradine is a drug commonly used for the symptomatic management of heart-related chest pain, also called stable angina pectoris, and heart failure not fully managed by beta blockers. It is the first specific heart-rate lowering agent. This reduces the cardiac load which prevents ischemia of the heart muscle. Ivabradine is a specific blocker of the HCN channel and thereby lowers the funny current in the sinoatrial node. This results in a slower depolarization of the node cells without affecting the duration of the action potential which leads to a slower heart rate and no reduction in heart contractility. Ivabradine blocks the HCN channel in a dose-dependent manner by entering the channel pore from the intracellular side.¹

Ethosuximide

Ethosuximide is a common anti-epileptic drug that reduces the amount of seizures in epilepsy patients. It is known that ethosuximide blocks the T-type calcium channels of thalamic neurons in a voltage-dependent way and thereby reduces the calcium current. This current is of great importance for several cellular processes such as muscle contraction, hormone or neurotransmitter release, gene expression, cell motility, cell division and cell death. In the specific case of epilepsy, seizures originate because of synchronized neuronal activity in the thalamic nerves. Blockage of the T-type calcium channels inhibits depolarization which will decrease the amount of synchronized neuronal activity and reduce the risk of initiating seizures.²

Cyclosporin

Cyclosporin is an immunosuppressant commonly used to prevent organ transplant rejection or to treat several autoimmune diseases. Cyclosporin mediates its effect by binding to an intracellular receptor cyclophilin, the formed complex inhibits calcineurin, a Ca2+ calmodulin-dependent phosphatase 2B. By inhibiting calcineurin cyclosporine prevents calcineurin-dependent interleukin-2 transcription and T-cell activation and thus causes immunosuppression. Next to this mode of action, it is also proven that cyclosporin affects several ion channels. First of all, an inhibition of voltage-gated Ca2+ channels in hippocampal neurons was reported. Secondly, cyclosporine also inhibits voltage-gated K+ channels in human lymphocytes. Not only voltage-gated K+ channels are inhibited but also hERG channels expressed in a HEK293 cell line are inhibited by cyclosporin. This does not directly indicate cardiotoxicity since HEK293 cells have a different phospholipid composition, the inhibitory effect observed can be different in cardiomyocytes. This means that blockade of hERG in HEK293 cells does not imply that cyclosporine causes QT elongation and possible Torsades de Pointes syndrome.³

Patch Clamp

Whole-cell patch clamp is an electrophysiological technique to study ionic currents by measuring membrane potential directly on single-cell level. More specific: it allows you to measure the electrical current or voltage differences across the membrane of a single cell. In our HEKcite project, we aimed to create stably oscillating HEK cells, by transfecting three key ion channels. Afterwards we wanted to influence this rhythm by adding drugs. Thus, it is crucial for our project to study our ion channels in the most precise manner.

MEA (referentie: paper Braeken)

A multi-microelectrode array is a device that measures electrical activity in cells, usually neurons. It is a transducer converting the voltage induced by ionic currents through the cell membrane to a current due to the movement of electrons. The MEA that was used for the experiments with the HEK cells can measure both extra- and intracellularly. The array is based on CMOS technology. A wide range of MEAs are commercially available, ranging from a few electrodes to several thousands. One of the most important advantages of MEAs is the possibility to stimulate and record each electrode separately. Sometimes amplification is already included in the chip, as is the case for the one from IMEC. The MEA chip is mounted on a printed circuit board and has a small glass chamber on top where the cells go. The intracellular measurement is made possible by the MEA’s option to create nanopores in the cell’s membrane by applying voltage pulses. These holes are so small, they hardly affect the cell’s survival chance and the cell membrane repairs itself quickly. For measurement with the MEA, the peaks of the membrane potential of the cells need to be sharp/steep enough. This has to be verified before measuring. The output of the MEA is an analog signal.

The implantable device we eventually aim to create consists of several parts. First and foremost, there is the multi-microelectrode array (MEA) for which we collaborated with IMEC, a leading research center in micro- and nanoelectronics. The output of this chip still needs amplification, filtering and conversion from analog to digital. This all will be included in a second chip. This second chip will also include Bluetooth to send the signal to for example a bracelet the patient is wearing, and components for power supply. The cells and the electronics are then encapsulated by two membranes. The purpose of these membranes is to keep the cells from escaping while letting nutrients and the medication to be measured in. All of these components will now be discussed in more detail.

Second chip

The signal processing of the MEA output includes amplification of the signal, filtering and DC to AC conversion. This can all be included in one integrated circuit. There are some companies on the market that produce these, Intan being one example. The output of this chip is digital.

Encapsulation

In order to use cells with an oscillating membrane potential as a means of continuous drug monitoring in the blood, a method needed to be established via which the cells could efficiently and safely be employed in a medical context. We first performed a preliminary investigation, during which various strategies were explored and evaluated for their potential clinical applicability.

To successfully measure the concentration of different drugs in the blood, the HEKcite cells have to be in perpetual contact with the bloodstream. As one might expect, simply injecting these cells was ruled out quite early in the evaluation process, as this would be both dangerous to the patient and would generate a situation in which it would be an absurdly difficult (if not impossible) task to measure the membrane potential of the circulating cells.

Further downstream the screening process, we stumbled upon the cell encapsulation technique. This method not only seemed promising for our project, but was already being investigated by multiple research groups for potential future applications. Furthermore, we were able to schedule appointments with two different experts willing to advance our understanding of in vitro encapsulation of living cells.

Before these meetings, we assessed the possible dangers and implications regarding capsules. This exploratory analysis revealed a list of primary risk factors associated with capsule implantation. This list includes (but is not limited to): Possible viral infections, risk of thrombosis, DNA leakage out of the capsule, and an immune response against the foreign material. Bernard Schneider, a senior scientist in the neurodegenerative studies laboratory in the university of Lausanne, was able to provide us with a list of answers to various questions like the ones listed above.

Type of cell

First of all, we made an inquiry about the type of cells we would use. HEK 293 cells, like the ones we created a sinus rhythm in, are known to have a high growth rate. This would result in the cells completely filling the capsule, which would cause them to starve and die. Furthermore, they contain adenoviral DNA, which makes them unacceptable for entering the human body. These two factors reduce the applicability of our HEK cells strictly to this proof of concept study. For this reason, we started looking for other cells.

As noted by professor Cosson, the key decision criteria for choosing new cells are: (1) the ease with which they are engineered, (2) the ability to store them in vitro in large batches, (3) the ease to be cultured in cheap media, and finally (4), the cells need to be in biosafety category 1 to comply with regulation. Two cell types that meet these requirements, human myoblasts and ARPE-19 cells, were investigated to be used in the finalized capsule.

Specifics of chip in capsule

Secondly, we asked professor Schneider about the possibility of a chip in a capsule, and whether drugs could enter it through the bloodstream. The answer was that, in principle, these should both be possible. The implant itself could be quite simple, as it could simply be an empty space that can be filled with a gel and cells. Electrodes, used for the continuous measurements of the changing membrane potential, could be introduced, with the cells being measured on an electrode array. As noted by Schneider however, one should keep in mind that the cells need oxygen and nutrients. Thus, when a chip is inserted, the exchange surface should be large enough to accommodate for the number of cells in the capsule.

Entry of drugs into the device

Regarding the entry of drugs into the device, the answer lies within neovascularization, according to Schneider. The formation of new blood vessels and their subsequent contact with the membrane should allow tight contact between blood and device. To establish whether a specific molecule can enter the capsule however, the structure of the semipermeable membrane and the size of its pores are determinant. Before commercial application of such an apparatus, these parameters should first be assessed.

While entry of drugs is an important parameter, the possible leakage of DNA out of the capsule is another. It has, however, been proved in multiple clinical trials, that the amount of DNA released from capsules is negligible. Professor Schneider (and we agree on this) thus believes the use of such capsules as ethically acceptable.

Risk of thrombosis

When implanting small capsules, one should always consider the risk of that device ending up in the bloodstream, and causing thrombosis. In our case, the macro-capsule would be relatively big (bigger than a 2€ coin in general). This makes it physically improbable that such a device could enter the blood stream and cause thrombosis.

Immune reaction

Last but not least, we inquired about possible dangers associated with immune responses against such an implanted device. When implanting a device (or something else that is considered ‘foreign’) into the human body, the typical reaction will be the foreign body response. To avoid going into great detail, the endpoint of this reaction is the formation of fibrous tissue that encapsulates the foreign body. Important to note, is that a low level of inflammation would help this kind of neovascularization (which is positive), but chronic inflammation caused by cellular hypoxia will cause an excess of fibrosis which will prevent oxygen and nutrients from entering the capsule.

An obvious design parameter for the capsules is the biocompatibility of the membranes to minimize host response. To help us decide on a suitable material we spoke to prof. Patterson of the department of Materials Engineering. Together we decided on the following specifications of the membrane: the material should be inert, non-degradable, cheap and of course biocompatible. Taking all this into account we landed on Polypropylene (PP). This polymer is a well-known biocompatible material which is commercially available. Because of its use in multiple applications it is well characterized. It is also relatively non-degradable and inert. The two membranes that encapsulate the chips and cells should be adhered to each other with welding techniques, since glues don’t tend to hold on PP. To manufacture the membranes in the desired shape, injection molding could be used. A hydrogel to grow the cells on, is not needed since the device only needs a sheet shaped 2D layer of cells. It could however be useful to place between the cells and the membranes to keep everything in place. A commonly used hydrogel that could be used is Polyethylene Glycol (PEG). Since PP is commercially available, the membranes could just be ordered and not custom made and designed, which reduces cost.

\[\frac{\partial C_{O_2}}{\partial t}=-\nabla V-V\nabla C_{O_2}\] If the liquid is incompressible, we can assume that \[\nabla V=0\] Blood is slightly compressible, but for simplicity make this approximation. We can do this because the Mach number is very small and the blood thus behaves as incompressible. \[Ma=\frac{V}{C}=\frac{1\rightarrow100 ^{cm}/_s}{1540 ^m/_s}\] Hence, we can state that \[\frac{\nabla C_{O_2}}{\nabla t}=-V\nabla C_{O_2}\]

Advection

The signal processing of the MEA output includes amplification of the signal, filtering and DC to AC conversion. This can all be included in one integrated circuit. There are some companies on the market that produce these, Intan being one example. The output of this chip is digital.

Navier-Stokes

Navier-Stokes can be used only for Newtonian fluids. Blood is a non-Newtonian fluid which means we again have to make an approximation.
\[\rho\displaystyle(\frac{\nabla V}{\nabla t}+v\nabla v)=\underbrace{-\nabla p}_\text{forces due to pressure differences}+\underbrace{\mu\nabla^2v}_\text{forces due to viscosity}+\underbrace{f}_\text{body force}\] The left hand side is equal to zero since the blood has a constant velocity on average. This means that the Navier-Stokes equation can be simplified to \[\nabla p=\mu\nabla^2v+f\]

Blood through membrane: Darcy

Since the membrane is a semi-permeable and thus prous, the flow of the blood through it can be describes by Darcy’s law: \[Av=-\kappa(\nabla p-fg)\] with A, the cross-sectional area of the pores, equal to 0,45 μm, g=9,81 N/kg and κ????

Diffusion: Fick

Besides active flow through the membrane, there’s also diffusion. This process can be described by Fick’s law of diffusion: \[\frac{\nabla C_{O_2}}{\nabla t}=\nabla(D_{O_2}\nabla C_{O_2})\] We can approximate that the oxygen is distributed homogeneously around the membrane and thus the equation becomes \[\frac{\nabla C_{O_2}}{\nabla t}=D_{O_2}\nabla^2 C_{O_2})\] We assume steady state \[\implies C_{O_2}(x)=C_0+(C_1-C_0)\frac{x}{d}\]in which C₀ is equal to the C_0₂ outside the membrane and C₁ is the C_0₂ inside the membrane.

When we combine all these equations, we get the following equations: \[\begin{cases} \frac{\partial C_{O_2}}{\partial t}=D_{O_2}\nabla^2C_{O_2}-v\nabla C_{O_2}-\underbrace{QC_{cell}}_\text{consumption}\\ \frac{\partial C_{cell}}{\partial t}=\underbrace{AC_{cell}(1-\alpha C_{cell}}_\text{growth}-\underbrace{dC_{cell}}_\text{death} \end{cases}\] To further describe Q, we can use the Michaelis-Menten equation: \[Q=\frac{C_{O_2}}{K_m+C_{O_2}}\] \[d=f(C_{O_2})=1-\frac{C_{O_2}}{K_d+C_{O_2}}\].

References:

1. Ivabradine – the first selective sinus node If channel inhibitor in the treatment of stable angina. S. SULFI, A. D. TIMMIS]

2. Clinical Significance of Animal Seizure Models and Mechanism of Action Studies of Potential Antiepileptic Drugs H. Steve White
Properties of Antiepileptic Drugs in the Treatment of Idiopathic Generalized Epilepsies
Block of Cloned Human T-Type Calcium Channels by Succinimide Antiepileptic Drugs]

3. A calcineurin inhibitor, FK506, blocks voltage-gated calcium channel-dependent LTP in the hippocampus. Onuma H, Lu YF, Tomizawa K, Moriwaki A, Tokuda M, Hatase O, Matsui H
Immunosuppressors Inhibit Voltage-Gated Potassium Channels in Human Peripheral Blood Lymphocytes
Inhibitory Actions of HERG Currents by the Immunosuppressant Drug Cyclosporin A]

Difference between revisions of "Team:KU Leuven/Design"