The development of an in vitro model of spinal cord injury using microfabrication

Ross, Lewis (2013) The development of an in vitro model of spinal cord injury using microfabrication. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2013rossphd.pdf] PDF
Download (73MB)
Printed Thesis Information:


This project aimed to develop an in vitro model of spinal cord injury. A device was developed that could, in conjunction with an existing myelinating culture system, aligned axons into parallel rows and contain the soma to a predefined area; thus creating a structure similar to the spinal cord. It was intended for the device to be simple to
produce and use, provide easy accesses to the axons for wounding and allow simple measurement of axonal regrowth. An additional aim of the project was to develop a method of cell wounding that was reproducible, precise and allowed observation of the wounding process.

A series of ridges were created on PDMS devices, which were used to align the axons. These ridges were 5 μm deep and 12.5, 25 or 50 μm wide. The 12.5 μm ridges aligned an average of 54% of the axons within 10° of the axis parallel to the ridges. The 25 and 50 μm only managed to align 37% and 28% in this direction. Flat control devices aligned 16% of the axons in an equivalent direction, which corresponded to a random alignment. From this it was concluded that 12.5 μm ridges should be used to align the axons.

Rows of 25 μm high, 20 μm diameter pillars were placed across the devices in an attempt to contain cells to one side of the device. The pillars had gaps between them of 6, 8 or 10 μm. MG63 cells were used to test the pillars’ containment ability. It was found that the cells could climb over the pillars and so multiple rows of pillars where created to attempt to trap the cells. The multiple rows did not full contain the cells but it was found that 5 rows with gaps sizes of 6 μm could slow the cells migration across the devices.

It was found that the pillar rows were not necessary to contain the neuronal soma, as they would not migrate far from their initial seeding area. However, the pillars had an
effect on the axons’ alignment. One, three and five rows of pillars reduced the alignment to 40%, 24% and 28% respectively. It was hypothesised that the axons were using the pillars as a “turning post”.

An automated method of wounding the cells was developed using a microscope and micromanipulator. This method was trialled on layers of astrocytes and MG63s grown on flat and grooved PDMS. Cutting astrocytes perpendicular to the grooves resulted in wounds with an area five times larger than those caused by cutting parallel to the grooves. Perpendicular wounds were also twice the area of those on flat PDMS. Similar effects were seen on MG63s. It was hypothesised that the difference in wounds sizes was due to the ridges causing the cells to make stronger mechanical connections longitudinally with the grooves.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Cell engineering, microfabrication, spinal cord injury, SCI, wounding, central nervous system, CNS, axonal guidance, contact guidance, in vitro model, photolithography
Subjects: Q Science > QH Natural history > QH301 Biology
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Molecular Biosciences
Supervisor's Name: Riehle, Dr. Mathis, Gadegaard, Dr. Nikolaj and Susan, Prof. Barnett
Date of Award: 2013
Depositing User: Lewis Ross
Unique ID: glathesis:2013-4365
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 13 Jun 2013 09:36
Last Modified: 13 Jun 2013 09:39

Actions (login required)

View Item View Item


Downloads per month over past year