Corneal enzymatic digestion resistance in the presence of oestradiol and oestradiol plus selective tissue oestrogenic activity regulators (STEAR).

The human cornea is an essential part of the optical system and is mainly composed of collagen types I, II and IV. 1 Its shape determines the refractive properties of the cornea. 2 The biomechanical stability of the cornea, and therefore its shape, is primarily governed by its extracellular matrix, which is composed mainly of collagens, proteoglycans and glycosaminoglycans. 3 This stability is facilitated by cross-links between the amino side chains of the collagen fibres and the proteoglycans in the extracellular stromal matrix. 1 4 5 In addition to providing biomechanical strength, these cross-links offer resistance against the increased activity of collagenolytic enzymes, often seen in inflammatory and infectious conditions such as infectious keratitis and progressive keratoconus. 6 Oestradiol can play several important roles in human physiology. These include a direct effect on collagen fibres throughout the body 7 and primarily involve the upregulation of matrix metalloproteinases (MMPs), 8 which can lead to the activation of collagenolytic enzymes responsible for the degradation of type I collagen, 9 leading to increased corneal distensibility and reduced stiffness. 3 8 10 11 Oestradiol can also enhance prostaglandin release, which subsequently activates collagenases, which can then disrupt the stability and shape of the cornea. 12 During the third trimester of pregnancy, increased levels of oestradiol in the blood can weaken the biomechanical stability of connective tissues. 13 This weakening impact has also been observed in the cornea following the discovery of oestradiol receptors in 2001 14 : when ex vivo porcine corneas were subjected to elevated levels of oestrogen, they displayed a distinct reduction in their biomechanical strength. 3 A previous study hypothesised that drug-induced disturbances to the oestradiol pathway might disrupt the integrity of stromal collagen in a manner similar to oestrogen. One well-characterised method of assessing this is through the use of ex vivo porcine corneal button enzymatic digestion assays. 615 22 Furthermore, a case report of a female patient with dormant keratoconus who experienced keratoconus progression after receiving 28 days of therapy with the selective tissue oestrogenic activity regulator (STEAR), tibolone, was published recently. 23 This study examines how oestradiol and tibolone influence the stromal capacity to resist enzymatic digestion, both alone and in combination with STEAR therapies.

This study included a total of 48 corneas that were evenly distributed into three groups (n=16 each). Digestion results are displayed in figure 2 . For group 1 control corneas, the mean digestion times were 31.38±2.03 hours. Corneas in group 2 (oestradiol group) showed a mean digestion time of 27.25±1.84 hours, displaying a significantly lower digestion resistance than group 1 control corneas (p<0.01). Finally, corneas in group 3 (oestradiol and tibolone) showed a mean digestion time of 22.38±2.47 hours, being significantly less resistant to enzymatic digestion than group 1 control corneas (p<0.01) but also showing significantly less resistance than group 2 (oestradiol) corneas (p<0.01).

The human cornea is mainly composed of a collagen scaffold, predominantly consisting of collagen subtypes I, III and IV. 24 Several factors can influence the optical properties of the cornea, but the most important is its shape, and the cornea needs to maintain a certain biomechanical stability to preserve its shape. 25 Several ocular pathologies are associated with reductions in corneal biomechanical strength, such as the family of corneal ectasias, or inflammatory and infectious processes. 26 29 A significant contributor to this reduction is the upregulated activity of collagenolytic enzymes, especially the MMP family, which digest stromal collagen by binding and cleaving the collagen molecule at specific binding sites. Any factor interfering with the regulation of MMP expression can directly affect the biomechanical stability of the cornea. Oestrogen plays a vital role in a multitude of physiological processes in humans. Oestrogen receptors are present in most connective tissues throughout the body, and in late pregnancy, elevated oestradiol levels prepare the body for the act of birth by loosening and softening the connective tissues. Pregnancy is associated with significant changes in the ocular anterior segment and corneal biomechanics, including transitory variations in corneal curvature that typically reverse postpartum, and these hormonal shifts can exacerbate conditions like iatrogenic keratectasia, particularly in patients with a history of refractive surgery, despite previous corneal cross-linking. 30 32 Both alpha and beta oestrogen receptors have been identified in the cornea of mice and humans. 14 33 34 Furthermore, β-oestradiol, the free form of oestrogen, has been shown to cause an upregulation of MMP genes in immortalised human corneal epithelial cells. 35 Several case reports and studies have shown that changes in hormone levels—particularly heightened serum oestradiol and hypothyroidism—may directly affect corneal biomechanics, exacerbating or triggering the onset of corneal ectasia in corneas previously affected by hypothyroidism. 3236 39 In 2007, Spoerl et al were the first to show that ex vivo porcine corneas exposed to β-oestradiol for several days showed markedly reduced biomechanical stability, as evidenced by stress-strain measurements. 3 Tibolone is a selective tissue oestrogen activity regulator with a chemical structure similar to oestrogen. 40 Tibolone is metabolised into 3α-hydroxytibolone and 3β-hydroxytibolone, both of which bind oestrogen receptors and display oestrogenic effects. 41 While the precise mechanism through which oestrogen impacts biomechanics remains unclear, in vivo studies have shown that oestrogen receptors possess the capacity to influence the biosynthesis of collagen and glycosaminoglycans. 42 43 These elements play a pivotal role in the composition of the corneal stroma and can therefore influence the biomechanics of the cornea. However, the reduced resistance to collagenase digestion observed in our in vitro experiments cannot be attributed to any active biological processes; instead, it is likely due to the direct structural effects of oestradiol and tibolone on the corneal stroma. Additionally, a direct inhibitory interaction between the hormonal agents and collagenase cannot be ruled out, as oestradiol and tibolone may alter enzyme activity or binding affinity, further affecting the rate of stromal digestion. Clinically, tibolone is widely used as a hormone replacement therapy to mitigate osteoporosis and menopausal symptoms in postmenopausal women, and it is also used to treat endometriosis. 40 44 A previous case report of tibolone therapy described reactivation and exacerbation of a previously dormant keratoconus in a female patient aged over 50 years, 23 and this, combined with its widespread clinical use, was the reason for choosing tibolone as the agent under investigation. This study has several limitations, including its ex vivo design and the lack of age and sex data for the porcine eyes, as this information was not available from slaughterhouse records. However, the clinical manifestation of tibolone-induced ectasia progression has already been documented in a previous case report. The aim of the present study was not to replicate this clinical observation, but rather to further characterise the stromal susceptibility to enzymatic digestion under controlled conditions following exposure to oestradiol and tibolone.

There is a clear interaction between heightened oestradiol levels and tibolone on corneal biomechanical strength, but the potential for this drug to alter corneal stromal resistance to collagenase enzymatic digestion was not fully understood until now. Our ex vivo porcine cornea digestion study demonstrates that oestradiol reduces this digestion resistance and that additional exposure to STEAR substantially weakens the ability of the stroma to withstand enzymatic digestion. These findings suggest that STEAR therapy may pose an additional risk for the progression of keratoconus and could worsen corneal melting in infectious keratitis. Consequently, assessing the hormonal status of keratoconus patients regarding STEAR therapy should be considered an essential part of the patient’s medical history evaluation.

Freshly enucleated porcine eyes (<6 hours postmortem) were obtained from a local slaughterhouse. Following epithelial removal using a hockey knife (FEATHER, pfm medical ag, Cologne, Germany), corneal buttons were created using Westcott scissors, leaving a 3 mm corneoscleral rim. The corneal buttons were then immersed in a 400 mOsmol/L phosphate-buffered saline solution for 10 min, followed by central trephination using an 8 mm biopsy punch (SMI, St. Vith, Belgium). After trephination, the buttons were placed in 24-well plates (Merck, Darmstadt, Germany), with each well containing 2.0 mL of 0.3% collagenase A solution (Roche, Basel, Switzerland). The plate was placed on a thermoshaker (37°C with 150 revolutions per minute; figure 1 ) and underwent tissue culture in a 5% CO 2 environment, as previously described. 3 The corneal buttons were visually inspected and photographed every hour. The time until complete enzymatic digestion for each cornea was recorded. Complete digestion was defined by the complete dissolution of the button and the formation of a dust-like layer. One individual recorded all images and was unblinded regarding the allocation of corneas to each group. We randomly assigned 48 porcine eyes to one of three study groups, each consisting of 16 corneas. Control corneas were subjected to digestion with 2 mL of 0.3% collagenase A solution diluted in Dulbecco’s Modified Eagle Medium (DMEM; ThermoFisher Scientific, Reinach, Switzerland). Corneas were incubated in a 2 mL total solution containing DMEM, collagenase A at a final concentration of 0.3% (w/v) and β-oestradiol at a final concentration of 20 µmol/L (E2257, Sigma-Aldrich Chemie, Buchs, Switzerland). Corneas were incubated in a total volume of 2 mL of a solution containing DMEM, collagenase A at a final concentration of 0.3% (w/v), β-oestradiol at 20 µmol/L and tibolone (Tibolon-Mepha Tablets 2.5 mg, Mepha, Aesch, Switzerland) at a final concentration of 2.5% (w/v). The main outcome measure was the time (in hours) to complete the digestion of each corneal button. Statistical analysis was performed by using R Studio (V.2022.12.0), Prism 9 (V.9.5.1), and Microsoft Excel (V.16.70). The descriptive statistics were presented as mean±SD. Normal distribution was tested and confirmed for all groups using the Shapiro-Wilk test. Analysis of variance was therefore used to analyse the differences between the groups, with a p-value <0.05 indicating statistical significance.