Central corneal thickness does not correlate with TonoLab-measured IOP in several mouse strains with single transgenic mutations of matricellular proteins

Abstract

Accurate and reliable measurement of intraocular pressure (IOP) is crucial in the study of glaucoma using the mouse model. The purpose of this study was to determine the relationship between TonoLab-measured IOP and central corneal thickness (CCT) in mouse strains with single gene mutations of matricellular proteins. Wild-type (WT) and transgenic mouse strains with single gene mutations (KO) of thrombospondin-1 (TSP-1), thrombospondin-2 (TSP-2), osteopontin (OPN), hevin, and secreted protein acidic rich in cysteine (SPARC) were imaged at six weeks using optical coherence tomography (Stratus, Zeiss) to determine CCT. IOP was measured between 11am and 3pm using TonoLab, one week later. For all measurements, mice were anesthetized using intraperitoneal injection ketamine:xylazine. CCT and IOP were measured in 583 mice (TSP-1 n = 71 and 41, TSP-2 n = 60 and 32, OPN n = 81 and 50, hevin n = 59 and 76, SPARC n = 54 and 59, WT and KO, respectively). Mean CCT was 5-6% lower in three KO strains-TSP-1, OPN, and SPARC-compared to their corresponding WT (p = 1.55 × 10(-7), 1.63 × 10(-11), and 1.91 × 10(-7), respectively). The mean IOP was 8.3%, 6.6%, and 15.1% lower in three KO strains-TSP-1, TSP-2, and SPARC-compared to corresponding WT (p = 2.11 × 10(-5), 2.93 × 10(-3), and 3.76 × 10(-9), respectively. Linear regression of IOP versus CCT yielded no statistically significant within-strain correlations for TSP-1 (p = 0.12 and 0.073), TSP-2 (p = 0.473 and 0.92), OPN (p = 0.212 and 0.916), Hevin (p = 0.746 and 0.257), and SPARC (p = 0.080 and 0.056), reported as p-values considering a null hypothesis of zero slope (WT and KO, respectively). Neither C57-derived strains (TSP-1 and OPN) nor 129-derived strains (TSP-2, hevin, SPARC) demonstrated a correlation between mean IOP and mean CCT across different strains (p = 0.75 and p = 0.53, respectively). Taken together, these results indicate that CCT is not required to interpret TonoLab IOP readings in the mice when CCT varies 10% about the mean. This does not exclude the possibility of an IOP-CCT correlation for CCT values outside this range or for inter-strain comparisons where the mean CCT differs more than 10%.

Keywords: TonoLab; central corneal thickness; glaucoma; intraocular pressure; rebound tonometry; trabecular meshwork; transgenic mice.

Figure 1

Mean (A) CCT and (B) TonoLab-obtained IOP in sedated mice of five matricellular KO strains and WT counterparts. Error bars represent standard deviation (SD). Asterisks indicate a statistically significant difference (p < 0.05) by student t-test.

Figure 2

Correlation of TonoLab-obtained IOP values with CCT within five strains of WT mice and their matricellular KO counterparts, respectively. (A) p = 0.12 and 0.073 [n = 71 and 41], (B) p = 0.47 and 0.92 [n = 60 and 32], (C) p = 0.21 and 0.92 [n = 81 and 50], (D) p = 0.75 and 0.26 [n = 59 and 76], and (E) p = 0.080 and 0.056 [n = 54 and 59]. In each case, the mutated mouse gene is displayed at the top of the panel.

Figure 3

Correlation of mean TonoLab-obtained IOP with mean CCT across C57- and 129-derived mouse strains. (p = 0.75 and 0.53, respectively).