Home

Compact LED Array for High-throughput Optogenetic Manipulation

MAE 156B Senior Design Project, Spring 2014

University of California, San Diego

Sponsored by Dr. Brenda Bloodgood

Background:

        Optogenetics is a method used by neuroscientists to artificially manipulate the activity of neurons in living tissue by activating light-responsive ion channels in the cell membranes of neurons. The neurons are genetically sensitized to light and respond to the local delivery of specific wavelengths of light. Moreover, the frequency of the light delivery can be used to modulate the pace at which the manipulated neurons will fire. Optogenetics is commonly used in animal models to study the behavioral changes that result from manipulation of a particular set of neurons. Optogenetics experiments are conducted on living animals, ex vivo slices, and dissociated neuron cultures, and have provided effective results in studying the circuitry of the brain. Current commercially available devices that enable optogenetic experiments are too expensive and can only perform one experiment at a time.

Illustrative explanation of how optogenetics works[1]

Objective:

       This project, sponsored by Dr. Brenda Bloodgood’s Neuroscience Lab at UC San Diego, focuses on creating a compact, high-power LED array in order to provide neuroscientists with a means to conduct high-throughput optogenetic manipulation. Optogenetics is the use of light to artificially manipulate neuron firing rates in living brain slices, while high-throughput refers to the ability to perform large scale iterations of experiments with different conditions. The LED array is designed to use optogenetics in a high-throughput manner so that Dr. Bloodgood’s lab can run multi-condition experiments simultaneously. The main objective of this project is to build an array of high-powered LEDs that can be driven at specific blink frequencies by a microprocessor, while addressing the need for temperature control, compact size, and ease of use. This LED array project will allow Professor Bloodgood’s lab to research the cellular and molecular changes in activated neurons at a much faster pace than currently possible and has long-term applications in memory and Alzheimer’s research.

Graphical representation of the four types of blink frequency parameters the LEDs require

Primary Requirements

• Power source for the system must run for 30 minutes

• LED array must be compatible with a standard 24-well culture plate

• Each LED must be cyan, with a corresponding peak wavelength of 505 nm

• Blink frequency must range from 0.1-100 Hz and burst frequency must range from 0.5 to 10 Hz

• Lowest LED blink duration of 10 ms and each burst must contain at least 10 blinks 

• Each LED must have optical power of irradiance on the scale of 1-100 mW/mm2

• The brain tissue slices inside the well plates must be maintained at physiological temperature (33-39°C)

• The device needs to be properly heat sinked in order to avoid overheating of the LEDs, PCB components, and brain tissue

• System must be compact

Secondary Requirements

• Implementation of a GUI to input desired experimental parameters

• Wireless connection for user to execute commands in Arduino program

• Have all components in one package, making the entire system portable

Final Design:

• 4 x 6 array of 3W LEDs on LED printed circuit board (PCB)

• Control Board PCB that connects the 24 LEDs into 8 sets of 3 and keeps the current constant at 1A

• Arduino Mega microprocessor drives each set of LEDs at specific blink frequencies

• Acrylic LED cover to prevent the light of each LED from leaking into other wells

• Two (2) 12V 7Ah rechargeable sealed lead-acid batteries to power the system

• Three (3) fans and two (2) heat sinks to dissipate heat from LEDs and PCB components

• Deltaphase Isothermal Pad to keep the brain tissue samples at physiological temperature

• Acrylic casing to make system portable

• Code that controls LEDs by inputting parameters of the experiment, such as signal and burst frequency

• Two (2) XBee antennae, XBee dongle, and XBee shield to send code wirelessly from the Graphical User Interface (GUI) to the Arduino

3D CAD drawing (front view) of final design of LED system

3D CAD drawing (back view) of final design of LED system

Circuit diagram for one set of 3 LEDs

Performance Results:

Battery Test

The system was tested with the LEDs turned fully on for 30 minutes. The voltage of the batteries and the voltage across the resistors were measured every 5 minutes. At the start of testing, however, two of the resistors from two sets of LEDs burned out because the resistors they were connected in parallel with were of the wrong wattage. Thus, data could only be taken from 6 sets of resistors instead of 8. The plots below show the data from 4 sets of 3 LEDs (one battery) because the battery that ran with only 2 sets is inconsistent with the experimental conditions. Battery 1 corresponds to Resistors 1-4 and Battery 2 corresponds to Resistors 5-8.

        The Battery Voltage vs. Time Plot shows how the voltage of the battery dropped from 12.4 ± 0.1 V to 11.9 ± 0.1 V. That gives a drop in voltage of 0.5 ± 0.1 V. The plot in and the Resistor Voltage vs. Time Plot shows how the voltage across the resistors changed over time. The voltage across the resistors fluctuated with time, ultimately dropping 0.017 V at the most. The initial voltage across the resistor minus the final voltage across the resistor for Resistors 5, 6, 7, and 8 respectively are: 0.017 ± 0.1 V, 0.008 ± 0.1 V, 0.007 ± 0.1 V, and 0.007 ± 0.1 V.

Deltaphase Isothermal Pad Test:

        The Deltaphase Isothermal Pad was heated until all of the material inside changed from solid to liquid. The heating pad temperature was measured at 41 °C while a well of the cell plate was measured at 36 °C. This experiment was performed in order to test how well the foam insulates heat and how well the aluminum plate conducts heat. While some heat is lost between the heating pad and the cell plate, enough heat is being transferred to keep the cell plate at the desired temperature of 33-39 °C. The heating pad can reach temperatures of up to 50 °C, so the insulation and conduction methods successfully fulfill the system requirements. 

        Aside from the two resistors that burned out (which will be fixed by the poster presentation) the system works successfully. The power source lasts more than 30 minutes, the brain tissue slices can be maintained at physiological temperature, and the heat sinks keep the circuit board components cool to the touch.

Video 1: Working final product with the whole system running

[1]From http://neurobyn.blogspot.se/2011/01/controlling-brain-with-lasers.html