Keratoconus is a corneal dystrophy that produces the most common need for transplant and is thus the focus of this research. However, transplants are limited by a global lack of donor tissue, where at least 55.3% of the global population lacks access to these resources [1]. While largely successful, transplant tissue efficacy deteriorates over time. Corneal graft survival occurred in 90% of patients at five years post-procedure and reduced to 82% at 10 years [1]. Studies measuring the quality of life for keratoconus patients report poor mental health as a consequence of the disorder [2]. This was attributed in part to the early onset progressive nature of the disease and anxiety around the possible need for keratoplasty. Specifically, clinical presentations of the disorder include sudden near-sightedness, astigmatism, blurred or distorted vision, sensitivity to bright light, and the need for recurrent prescriptions [3]. These outcomes reportedly leave patients with a loss of functional capacity, activity limitations, reduced participation in pleasurable activities, reduced social integration, and low self-efficacy [4]. Evidently, it would be significantly beneficial to develop a means of overcoming the accessibility and longevity limitations of current corneal restorative therapies.

The overarching goal of this research is to develop Mesenchymal Stem Cells (MSCs) as a regenerative medicine for corneal disorders. MSCs are hoped to provide an inexhaustible source of cells for eye regeneration, where they are continually replacing damaged cells and renewing the stem cell population. MSCs can be sourced from multiple sites, including the human umbilical cord. Relative to other sites, including bone marrow or adipose tissue, umbilical cord cells are less likely to develop cancerous teratomas and have higher proliferation and differentiation potential [5]. In combination with the immune-privileged nature of the cornea, stem cell transplant is a promising treatment for corneal disease.

My contribution to this research expanded on a previous Masters’ student’s work in the lab group, Vicky Wen, whose objective was to identify and quantify the distribution of MSCs along the length and within the 3-dimensional anatomy of the umbilical cord [6]. The purpose of her work was to optimise the recovery of MSCs by establishing where the greatest quantity of MSCs exist within the umbilical cord. She found evidence to suggest there is a gradient of MSCs between the foetal and placental ends of the umbilical cord. To date, no studies outside of this lab group have explored the MSC umbilical cord profile, especially their ability to differentiate into corneal keratocytes.

I received one umbilical cord (due to time and resource constraints) to perform gene expression analysis by droplet digital PCR. I dissected 1cm fragments along a 75cm umbilical cord, giving me 46 samples in total. For every 2cm of neighbouring tissue sections, 1cm was used in PCR, and 1cm was used for immuno-labelling, leaving 23 tissue samples available for each procedure.

Due to the Auckland flood and cyclone events, as well as the time waiting to retrieve the cord, the immuno-labelling portion of my project could not be completed with me present. This taught me the way research projects work sometimes, or don’t in this case, as highlighted by my supervisor.

Nevertheless, a significant portion of my project focused on learning PCR skills, where I extracted RNA from 23 umbilical cord samples to analyse for gene expression. Following DNase treatment to remove genomic DNA, I conducted RNA quality control processes, including the SPUD assay [7] to check for the presence of PCR inhibitors and the Tapestation instrument to check for the quantity and quality of RNA. After confirmation of sample quality and quantity, I synthesised cDNA from the umbilical cord RNA. The success of cDNA synthesis was checked by performing a PCR for the beta-actin gene and looking for the existence of the gene using gel electrophoresis. I continued on to learn ddPCR for gene expression analysis for my chosen genes of interest (GOI), in addition to a panel of six reference genes. The Normfinder algorithm was used to determine the most stable reference genes, and the geometric mean of the three most stable genes was used for normalisation. Normalised relative expression quantities were calculated and this allowed comparison between the 23 samples along the length of the cord [8].