ARVO Annual Meeting in Vancouver
Decellularized optic nerve head model to study glaucoma biomechanics
Purpose : While elevated IOP is often associated with glaucoma, many patients develop glaucoma with normal IOPs or experience ocular hypertension without developing glaucoma, indicating that other factors contribute to the disease. Evaluating these factors in vivo can be difficult and costly, though, so we have developed a model using decellularized posterior poles to aid in studying glaucoma biomechanics.
Methods : Posterior poles were isolated from fresh porcine eyes, and the retina and choroid were removed. The dissected tissue was then subjected to a mild, detergent-based decellularization process. Histology was performed to show that decellularization removed cells while maintaining the architecture and basement membrane of the tissue. To evaluate the ability of the model to mimic dissected tissue, we subjected the tissue to 5 and 15 mmHg of pressure and imaged the lamina cribrosa at each pressure using two-photon microscopy both before and after decellularization. To ensure that the decellularized tissue was still able to hold pressure, which is needed to image pressure-induced structural changes, the entire tissue was infiltrated with a bioinert PEG-based gel or the optic nerve was coated with super glue prior to pressurization. The gel-infiltrated decellularized tissue could then be seeded with human cells.
Results : Histology and two-photon imaging confirmed that decellularization was successful in removing cells and maintaining the structural architecture of the posterior poles. Gel-infiltrated and glue-coated tissues produced displacements 34 ± 4% and 69 ± 2% different than dissected tissues, respectively. While both types of coatings produced strains significantly different than those of the dissected tissues (p < 0.0001), all gel and glue coated optic nerves were able to hold both 5 and 15 mmHg of pressure, indicating their potential to be used to model pressure-induced structural changes similar to those experienced by glaucomatous optic nerve heads.
Conclusions : Our decellularized optic nerve head model is capable of maintaining the structural qualities of dissected optic nerve heads, and gel infiltration is able to help partially restore the biomechanical properties of the tissue prior to decellularization. Therefore, this biomimetic model offers an alternative in vitro platform to study cellular factors that could contribute to glaucoma development and progression.
Manipulating Gene Expression of Human Lamina Cribrosa Cells and Astrocytes
Purpose : Primary open angle glaucoma is a neurodegenerative disease that affects over 60 million people worldwide. It is more common in certain populations with prevalence rates of 2.4%, 3.6%, and 5.4% in those of European Descent, Hispanic Ethnicity, and African Descent, respectively. Previously, we have found that the microstructure and mechanical strain in lamina cribrosa (LC) is significantly different and that gremlin gene expression may be differentially regulated in these three populations. In this study, we aim to manipulate the gene expression of gremlin by delivering siRNA and shRNA via adeno-associated virus (AAV).
Methods : Human LC cells and astrocytes were isolated and cultured using an approved IACUC protocol. Immunostaining of gremlin was performed to examine protein expression in human LC cells and astrocytes. To evaluate the effect of transforming growth factor beta 2 (TGFβ-2) on human astrocytes, TGFβ-2 was added to the culture for 72 hours. To test AAV transfection efficiency, human LC cells and astrocytes were transfected with four different serotypes of AAV-CMV-GFP virus (AAV2, AAV5, AAV8, and AAV9) and the transfection efficiency was quantified by GFP signal.
Results : Immunostaining results show that gremlin protein is present in both human LC cells and astrocytes (Figure 1A). Addition of TGFβ2 reduces gremlin protein expression in human astrocytes (Figure 1B). We found that both human LC cells and astrocytes are transfectable using an AAV virus, confirmed by GFP immunofluorescence (Figure 1C).
Conclusions : Prior work in the literature has shown that Gremlin is upregulated in the LC region of glaucomatous tissues. Our data suggest that TGFβ2 activity and extracellular matrix remodeling may contribute to the altered regulation of gremlin gene expression in astrocytes from donors of African Descent. We successfully delivered AAV into human LC cells and astrocytes and plan to examine the effect of gremlin knockdown on TGFβ2 activity and extracellular matrix remodeling in future work.
Biomechanical Response of the Lamina Cribrosa in Glaucomatous and Non glaucomatous samples
Purpose : Glaucoma is the second leading cause of blindness in the world. Some of the risk factors for this disease include age, race, gender and intraocular pressure (IOP). More recent studies showed that the biomechanics of the optic nerve head region (ONH) is a critical component for the development of glaucoma. The purpose of this study is to compare the strain response of glaucomatous and non-glaucomatous human eyes in the lamina cribrosa (LC), which a collagenous structure in the ONH.
Methods : We used six glaucomatous and five non glaucomatous (NG) eyes from the European descent that were above 50 years old. We conducted a pressure inflation experiment while simultaneously collecting images at 5,15,30 and 45mmHg using multiphoton microscope and calculated the displacement field using Digital Volume Correlation (DVC). DVC results were used to calculate the Green strain components. LC regions were subdivided into four quadrants (inferior, superior, nasal, temporal) and two rings (central, peripheral) to investigate strain regional variations. Statistical analysis was done using Linear mixed model with disease and LC regions as fixed effects and human and eyeballs as random effects.
Results : Our results show regional variation in shear and frontal (in-plane) strain between the glaucomatous and NG samples. A figure showing these differences in the second pressure set (15-30mmHg) is attached below. The xy strain in NG was significantly lower than that of the glaucomatous in all quadrants and rings (Fig 1 A, D and E). On the other hand, the strain in yz of the central rings and nasal quadrant of NG was significantly bigger than that of G (Fig1 B, C and F).
Conclusions : The shear strain in the sagittal plane (Eyz) of glaucomatous samples was lower than NG samples. Most quadrants and rings showed higher shear strain in the frontal plane (Exy) in the glaucomatous group. Our recent work for LC strain differences in racioethnic groups showed similar strain results in the at-risked groups (African and Hispanic groups). This may be due to LC biomechanical similarities present in these groups and glaucoma patients. In future work, we plan on getting axon counts for each sample to get additional information of the ONH region. We will also determine LC material properties of both glaucomatous and non-glaucomatous samples.
Purpose : While elevated IOP is often associated with glaucoma, many patients develop glaucoma with normal IOPs or experience ocular hypertension without developing glaucoma, indicating that other factors contribute to the disease. Evaluating these factors in vivo can be difficult and costly, though, so we have developed a model using decellularized posterior poles to aid in studying glaucoma biomechanics.
Methods : Posterior poles were isolated from fresh porcine eyes, and the retina and choroid were removed. The dissected tissue was then subjected to a mild, detergent-based decellularization process. Histology was performed to show that decellularization removed cells while maintaining the architecture and basement membrane of the tissue. To evaluate the ability of the model to mimic dissected tissue, we subjected the tissue to 5 and 15 mmHg of pressure and imaged the lamina cribrosa at each pressure using two-photon microscopy both before and after decellularization. To ensure that the decellularized tissue was still able to hold pressure, which is needed to image pressure-induced structural changes, the entire tissue was infiltrated with a bioinert PEG-based gel or the optic nerve was coated with super glue prior to pressurization. The gel-infiltrated decellularized tissue could then be seeded with human cells.
Results : Histology and two-photon imaging confirmed that decellularization was successful in removing cells and maintaining the structural architecture of the posterior poles. Gel-infiltrated and glue-coated tissues produced displacements 34 ± 4% and 69 ± 2% different than dissected tissues, respectively. While both types of coatings produced strains significantly different than those of the dissected tissues (p < 0.0001), all gel and glue coated optic nerves were able to hold both 5 and 15 mmHg of pressure, indicating their potential to be used to model pressure-induced structural changes similar to those experienced by glaucomatous optic nerve heads.
Conclusions : Our decellularized optic nerve head model is capable of maintaining the structural qualities of dissected optic nerve heads, and gel infiltration is able to help partially restore the biomechanical properties of the tissue prior to decellularization. Therefore, this biomimetic model offers an alternative in vitro platform to study cellular factors that could contribute to glaucoma development and progression.
Manipulating Gene Expression of Human Lamina Cribrosa Cells and Astrocytes
Purpose : Primary open angle glaucoma is a neurodegenerative disease that affects over 60 million people worldwide. It is more common in certain populations with prevalence rates of 2.4%, 3.6%, and 5.4% in those of European Descent, Hispanic Ethnicity, and African Descent, respectively. Previously, we have found that the microstructure and mechanical strain in lamina cribrosa (LC) is significantly different and that gremlin gene expression may be differentially regulated in these three populations. In this study, we aim to manipulate the gene expression of gremlin by delivering siRNA and shRNA via adeno-associated virus (AAV).
Methods : Human LC cells and astrocytes were isolated and cultured using an approved IACUC protocol. Immunostaining of gremlin was performed to examine protein expression in human LC cells and astrocytes. To evaluate the effect of transforming growth factor beta 2 (TGFβ-2) on human astrocytes, TGFβ-2 was added to the culture for 72 hours. To test AAV transfection efficiency, human LC cells and astrocytes were transfected with four different serotypes of AAV-CMV-GFP virus (AAV2, AAV5, AAV8, and AAV9) and the transfection efficiency was quantified by GFP signal.
Results : Immunostaining results show that gremlin protein is present in both human LC cells and astrocytes (Figure 1A). Addition of TGFβ2 reduces gremlin protein expression in human astrocytes (Figure 1B). We found that both human LC cells and astrocytes are transfectable using an AAV virus, confirmed by GFP immunofluorescence (Figure 1C).
Conclusions : Prior work in the literature has shown that Gremlin is upregulated in the LC region of glaucomatous tissues. Our data suggest that TGFβ2 activity and extracellular matrix remodeling may contribute to the altered regulation of gremlin gene expression in astrocytes from donors of African Descent. We successfully delivered AAV into human LC cells and astrocytes and plan to examine the effect of gremlin knockdown on TGFβ2 activity and extracellular matrix remodeling in future work.
Biomechanical Response of the Lamina Cribrosa in Glaucomatous and Non glaucomatous samples
Purpose : Glaucoma is the second leading cause of blindness in the world. Some of the risk factors for this disease include age, race, gender and intraocular pressure (IOP). More recent studies showed that the biomechanics of the optic nerve head region (ONH) is a critical component for the development of glaucoma. The purpose of this study is to compare the strain response of glaucomatous and non-glaucomatous human eyes in the lamina cribrosa (LC), which a collagenous structure in the ONH.
Methods : We used six glaucomatous and five non glaucomatous (NG) eyes from the European descent that were above 50 years old. We conducted a pressure inflation experiment while simultaneously collecting images at 5,15,30 and 45mmHg using multiphoton microscope and calculated the displacement field using Digital Volume Correlation (DVC). DVC results were used to calculate the Green strain components. LC regions were subdivided into four quadrants (inferior, superior, nasal, temporal) and two rings (central, peripheral) to investigate strain regional variations. Statistical analysis was done using Linear mixed model with disease and LC regions as fixed effects and human and eyeballs as random effects.
Results : Our results show regional variation in shear and frontal (in-plane) strain between the glaucomatous and NG samples. A figure showing these differences in the second pressure set (15-30mmHg) is attached below. The xy strain in NG was significantly lower than that of the glaucomatous in all quadrants and rings (Fig 1 A, D and E). On the other hand, the strain in yz of the central rings and nasal quadrant of NG was significantly bigger than that of G (Fig1 B, C and F).
Conclusions : The shear strain in the sagittal plane (Eyz) of glaucomatous samples was lower than NG samples. Most quadrants and rings showed higher shear strain in the frontal plane (Exy) in the glaucomatous group. Our recent work for LC strain differences in racioethnic groups showed similar strain results in the at-risked groups (African and Hispanic groups). This may be due to LC biomechanical similarities present in these groups and glaucoma patients. In future work, we plan on getting axon counts for each sample to get additional information of the ONH region. We will also determine LC material properties of both glaucomatous and non-glaucomatous samples.