ABSTRACT
Refractive errors are the most common ocular disorders and are a leading cause of visual impairment worldwide. Although ocular axial length is well established to be a major determinant of refractive errors, the molecular and cellular processes regulating ocular axial growth are poorly understood. Mutations in genes encoding the PRSS56 and MFRP are a major cause of nanophthalmos. Accordingly, mouse models with mutations in the genes encoding the retinal factor PRSS56 or MFRP, a gene predominantly localized in the retinal pigment epithelial (RPE) exhibit ocular axial length reduction and extreme hyperopia. However, the precise mechanisms underlying PRSS56- and MFRP-mediated ocular axial growth remain elusive. Here, we show that Adamts19 expression is significantly upregulated in retina of mice lacking either Prss56 or Mfrp. Using a combination of genetic approaches and mouse models, we show that while ADAMTS19 is not required for ocular growth during normal development, its inactivation exacerbates ocular axial length reduction in both Prss56 or Mfrp mutant mice. These results suggest that the upregulation of retinal Adamts19 expression is part of an adaptive molecular response to counteract impaired ocular growth. Using a complementary genetic approach. We further demonstrate that loss of PRSS56 or MFRP function prevents excessive ocular axial growth in a mouse model of developmental myopia caused by a null mutation in Irpb, demonstrating that ocular axial elongation in Irbp-/- mice is fully dependent on PRSS56 and MFRP functions. Collectively, our findings provide insight into the molecular network involved in ocular axial growth regulation and refractive development and support the notion that relay of the signal between the retina and RPE could be critical for promoting ocular axial elongation.
INTRODUCTION
Nanopthalmos is a rare developmental disorder characterized by significantly smaller but structurally normal eyes and extreme hyperopia resulting from compromised ocular growth [1]. Also, nanophthalmic individuals are highly susceptible to developing blinding conditions including secondary angle-closure glaucoma, spontaneous choroidal effusions, cataracts, and retinal detachment [1]. Both sporadic and familial forms of napophthalmos with autosomal dominant or recessive inheritance have been reported [2]. To date, six genes (PRSS56, MFRP, TMEM98, CRB1, BEST1, and MYRF) have been implicated in familial forms of nanophthalmos, with PRSS56 and MFRP mutations accounting for the most frequent causes among multiple cohorts [1, 3-10]. Furthermore, the eyes of nanophthalmic individuals with biallelic mutations in PRSS56 or MFRP were found to be significantly smaller compared to those carrying dominant mutations in TMEM98 or MYRF. Interestingly, common variants of PRSS56 and MFRP have also been found to be associated with myopia, a condition phenotypically opposite to nanophthalmos that is characterized by increased ocular elongation [11]. Together, these findings underscore the importance of PRSS56 and MFRP in ocular size regulation[2].
Ocular growth can be broadly divided into two distinct phases that take place pre- and postnatally[12]. Prenatal ocular growth occurs in the absence of visual stimulation and is primarily dictated by genetic factors [13]. In contrast, postnatal ocular growth also referred to as emmetropization, is a vision-guided process modulated by the refractive status of the eye to ensure that the axial length matches the optical power of the eye to achieve optimal focus and clear vision. Abnormal postnatal ocular axial growth leading the increased axial length constitutes a major cause of myopia, a condition characterized by blurred vision caused by focused images falling in front of the retina [12, 14, 15]. Nanophthalmos is generally attributed to impaired prenatal ocular growth as individuals with this condition are born hyperopic [1, 16]. Interestingly, in addition to being responsible for nanophthalmos, common variants of MFRP and PRSS56 have also been found to be associated with myopia in the general population (an opposite condition) that primarily results from alterations in postnatal ocular axial growth [11, 17]. Thus, the association of PRSS56 and MFRP with nanophthalmos and myopia support a role for these factors in the regulation of embryonic and postnatal ocular growth development and suggest that the molecular mechanisms underlying pre- and postnatal ocular growth are shared.
It is generally accepted that postnatal ocular growth is regulated by a cascade of signaling events by which information is relayed from the retina to the sclera to induce scleral extracellular matrix (ECM) remodeling to promote ocular axial elongation and [14, 18]. Notably, PRSS56 expression is specifically detected in the retina [13], which is consistent with a central role for the retina in ocular growth regulation. MFRP is predominantly expressed in the retinal pigment epithelium (RPE) and ciliary epithelium[16] and is implicated in the transmission of molecular cues between retina and sclera during ocular growth. Using a genetic mouse model, we have recently demonstrated that the genetic ablation of Prss56 from retinal Müller glia leads to a significant reduction in ocular axial length and hyperopia[13]. Similarly, mice and zebrafish lacking MFRP exhibit ocular axial length reduction, and MFRP variants in humans are associated with myopia[19-21].
Although current shreds of evidence support a key role of MFRP and PRSS56 in ocular axial length determination, the underlying mechanisms remain elusive. In this study, we use Prss56 and Mfrp mutant mouse model in combination with complementary genetic approaches to gain insights into the molecular network involved in ocular size regulation. Importantly, we identified characteristic changes in retinal gene expression in response to impaired ocular growth. Specifically, we show that Adamts19 mRNA levels are significantly increased in the retina of Prss56 and Mfrp mutant mice and provide evidence that the upregulation of retinal Adamts19 expression is part of an adaptive molecular response to impaired ocular growth. Furthermore, we demonstrate that loss of PRSS56 or MFRP function prevents excessive ocular axial elongation in a mouse model of early-onset myopia caused by a mutation in Irbp. Collectively, our finding hints at a potential molecular link between Müller glia and RPE involved in ocular axial growth regulation.
RESULTS
Adamts19 expression is upregulated in the retina of Prss56 mutant mice
To begin addressing the molecular processes underlying PRSS56-mediated ocular size regulation, we performed RNA-Seq analysis on the retina from Prss56gclr4 mutant mice and their wild-type littermates. We recently demonstrated that the ocular size reduction that we originally described in mice with a Prss56gclr4 mutation (causing PRSS56 protein truncation) result from a loss of function mechanism, hence, Prss56gclr4/gclr4 mice will be referred to as Prss56-/- throughout the manuscript for simplicity[13]. Our transcriptome analysis identified Prss56 and Adamts19 as the top two differentially expressed genes between Prss56 mutant (Prss56-/-) and control (Prss56+/-) retina. Consistent with the RNA-Seq data, the qPCR analysis revealed that Prss56 and Adamts19 mRNA levels were significantly upregulated in the retina of Prss56 mutant mice (Prss56-/-) compared to their Prss56+/- and Prss56+/+ littermates at both ages examined (postnatal day (P) 15 and P30) (Fig. 1A-B). Importantly, Prss56 and Adamts19 retinal expression levels in heterozygous Prss56+/- mice were comparable to those detected in Prss56+/+ mice, which is consistent with the absence of an ocular phenotype in Prss56+/- mice [13]. Prss56+/- mice were therefore used as controls for all experiments presented in this study. As described previously [13], we detected a progressive upregulation of Prss56 mRNA levels in Prss56-/- retina from P15 to P60 (Fig. S1). The increase in Adamts19 retinal expression of was found to precede that of Prss56 in Prss56-/- retina, and was detected as early as P10 and gradually increased to reach peak expression levels by P30 (Fig. S1). Notably, qPCR-Ct values suggested that the expression of Adamts19 was minimal or negligible in Prss56+/+ and Prss56+/- retina. Furthermore, the upregulation of retinal Prss56 and Adamts19 expression was also observed in mice carrying a null allele of Prss56 (Prss56Cre), which we had described previously [13]. Thus, confirming that the increase in Prss56 and Adamts19 expression results from a loss of PRSS56 function (Fig. 1C). To determine the spatial distribution of Adamts19 mRNA, we next performed in situ hybridization on ocular sections from Prss56-/- and Prss56+/- mice. Despite using the highly sensitive QuantiGene View RNA in situ hybridization method, Adamts19 expression was only detected in Prss56-/- retina, indicating that Adamts19 expression was below the threshold of detection in control Prss56+/- retina (Fig. 1D). In Prss56 mutant retina, Adamts19 expression was predominantly observed in the inner nuclear layer (INL), a region containing the cell bodies of Müller glia, a cell type in which Prss56 is normally expressed. Collectively, these findings demonstrate that in addition to causing ocular size reduction, loss of PRSS56 function leads to alterations in retinal gene expression marked by increased Prss56 and Adamst19 mRNA levels.
Retinal Prss56 and Adamts19 mRNA levels are upregulated in response to ocular size reduction in Prss56 mutant mice
To determine whether the upregulation in retinal Adamts19 and Prss56 mRNA levels correlates with ocular size reduction in Prss56 mutant mice, we took advantage of the Egr1; Prss56 double mutant mouse model (Egr1-/-; Prss56-/-) that we described previously [13]. EGR1 (early growth response1) is a major regulator of ocular growth and Egr1-/- mice exhibit increased ocular axial length[13, 22]. We have previously shown that Egr1 inactivation rescues the reduction in ocular axial length and vitreous chamber depth (VCD) in Prss56 mutant mice as the ocular size of Egr1-/-;Prss56-/- mice is comparable to that of control Egr1+/-;Prss56+/- mice [12]. Using qPCR analysis, we show that in addition to rescuing ocular axial elongation, Egr1 inactivation also prevented the increase in retinal expression of Prss56 and Adamts19 in Prss56 mutant mice (compare Egr1-/-; Prss56-/- to Egr1+/-; Prss56-/- in Fig. 2). These findings suggest that the upregulation of retinal Prss56 and Adamts19 does not result from loss of PRSS56 function per se, but rather from its effect on ocular size.
ADAMTS19 is not required for ocular growth during normal development
PRSS56 and ADAMTS19 are both secreted serine proteases, raising the possibility that they might have overlapping functions in ocular growth regulation. To test this possibility, we first generated Adamts19 knockout mice by crossing a conditional Adamts19 mutant mouse line with the ubiquitous β-actin-Cre line (Fig. S2). To determine if the loss of ADAMTS19 function lead to ocular defects, we performed optical coherence tomography (OCT) to assess various ocular biometric parameters. We found that all the ocular parameters examined, including ocular axial length, VCD, and retinal thickness were indistinguishable between Adamts19+/+, Adamts19+/- and Adamts19-/- mice (Fig. 3 and Fig. S3). These findings demonstrate that ADAMTS19 is not required for ocular growth during normal development.
Loss of ADAMTS19 function exacerbates ocular axial length reduction in Prss56-/- mice
In light of our findings, we hypothesized that the upregulation of retinal Adamts19 expression might be part of an adaptive molecular response to compensate for the loss of PRSS56 function and promote ocular axial growth. To this end, we tested the effect of Adamts19 inactivation in Prss56-/- mice by crossing Prss56 mutant mice to the Adamts19 mutant line to generate Prss56-/- mice that are wild-type, heterozygous or homozygous for the Adamts19 null allele (Adamts19+/+, Adam19+/- or Adamts19-/-). Notably, since all ocular biometric parameters of Adamts19+/-; Prss56+/- mice were comparable to those of wild-type (Adamts19+/+; Prss56+/+) littermates (Fig. S4A), Adamts19+/-;Prss56+/- mice were used as controls. As expected, axial length and VCD were significantly reduced in all three groups of mice lacking Prss56 (Prss56-/-) compared to the control mice (Fig. 4). However, the axial length and VCD were significantly reduced in Adamts19;Prss56 double mutant mice (Adamts19-/-;Prss56-/-) compared to Prss56 single mutants (Adamts19+/-; Prss56-/- or Adamts19+/+;Prss56-/-) at both age examined (P18 and P30) (Fig. 4). As reported previously [13], ocular axial length reduction in Prss56-/- mice was associated with an increase in retinal thickness (Fig. S5). Notably, a modest but significant increase in retinal thickness was observed in Adamts19-/-;Prss56-/- mice compared to Prss56 mutant mice (Adamts19+/-;Prss56-/- and Adamts19+/+; Prss56-/-) at P18 (Fig. S5).
Besides, we found that Adamts19 expression was significantly increased in the retina from both Adamts19+/+;Prss56-/- and Adamts19+/-;Prss56-/- mice compared to that of Adamts19+/-;Prss56+/- control mice, which is consistent with the observation that exacerbation of the ocular axial length reduction in Prss56 mutant mice is only observed when Adamts19 is completely knocked out (Fig. S5C). Together, these results demonstrate that Adamts19 inactivation exacerbates ocular size reduction in Prss56-/- mice and is consistent with the upregulation of retinal Adamts19 expression being part of an adaptive molecular response triggered by impaired ocular growth in Prss56 mutant mice.
Adamts19 inactivation exacerbates ocular axial length reduction in Mfrp mutant mice
Interestingly, elevated retinal levels of Prss56 expression has recently been reported in another mouse model of nanophthalmos caused by a mutation in the gene coding for membrane frizzed related-protein (Mfrp) [23]. Increased Adamts19 expression was also observed in Mfrp-/- eyes but the specific ocular tissue/cell type in which Adamts19 was expressed was not addressed [23]. Since Adamts19 expression was specifically detected in the retina of the Prss56-/- mice, we performed a qPCR analysis to confirm that the levels of Prss56 and Adamts19 were upregulated in the retina of Mfrp-/- mice compared to control Mfrp+/- littermates (Fig. 5A). To determine if Adamts19 inactivation also exacerbates the ocular size reduction caused by Mfrp deficiency, we crossed Mfrp mutant mice with the Adamts19 mutant line and conducted OCT analyses on the progeny. Since the ocular biometric parameters of Adamts19+/-; Mfrp+/- were comparable to those of wild-type (Adamts19+/+;Mfrp+/+), they were used as controls (Fig. S6). As expected, Mfrp mutant mice (Adamts19+/-;Mfrp-/-) exhibited reduced ocular axial length and VCD compared to Adamts19+/-; Mfrp+/- control mice (Fig. 5C-E). Importantly, the ocular axial length and VCD of Adamts19-/-;Mfrp-/- mice were significantly reduced compared to Mfrp mutant mice (Adamts19+/-;Mfrp-/-) (Fig. 5C-E). In addition, retinal thickness was increased in Adamts19+/-;Mfrp-/- and Adamts19-/-;Mfrp-/- mice compared to control Adamts19+/-; Mfrp+/- mice (Fig. S7). These findings further support a role for the upregulation of retinal Adamts19 expression being part of a compensatory mechanism triggered by impaired ocular axial growth.
Inactivation of Prss56 or Mfrp prevents excessive ocular axial elongation in Irbp mutant mice
To further establish the role of PRSS56 and MFRP in ocular elongation, we tested the effects of Prss56 and Mfrp inactivation in a mouse model of early-onset developmental myopia associated with excessive ocular axial growth caused by a null mutation in the gene coding for IRBP (Interphotoreceptor retinoid-binding protein)[24]. To this end, each of the Prss56 and Mfrp mutant lines were crossed to Irbp mutant mice and biometric ocular assessment was conducted on their progeny. As expected, OCT analyses revealed that ocular axial length and VCD were significantly increased in Irpb single mutant mice (Irbp-/-;Prss56+/- or Irbp-/-:Mfrp+/-) and significantly reduced in Prss56 or Mfrp single mutant mice (Irbp+/-;Prss56-/- or Irbp+/-;Mfrp-/-,) compared to their respective controls (Irbp+/-;Prss56+/- and Irbp+/-:Mfrp+/- mice) (Fig. 6A-D). Inactivation of either Prss56 or Mfrp prevented ocular axial elongation in Irbp mutant mice (Irbp-/-;Prss56-/- and Irbp-/-;Mfrp-/-, respectively) (Fig. 6A-D). Notably, ocular axial length and VCD were significantly reduced in both double mutant lines (Irbp-/-;Prss56-/- and Irbp-/-;Mfrp-/-) compared to their respective control littermates (Irbp+/-;Prss56+/- and Irbp+/-;Mfrp+/-, respectively) and were comparable to those observed in Prss56 and Mfrp single mutant mice (Irbp+/-;Prss56-/- and Irbp+/-;Mfrp-/-, respectively) (Fig. 6A-D and Fig. S8 A, C). In addition, while retinal thickness was increased in both Prss56 and Mfrp single mutant mice (Irbp+/-;Prss56-/- and Irbp+/-;Mfrp-/-), it was significantly reduced in Irpb single mutant mice (Irbp-/-;Prss56+/- or Irbp-/-;Mfrp+/-) compared to control littermates (Irbp+/-;Prss56+/- and Irbp+/-:Mfrp+/-, respectively) (Fig. S8B, D). Interestingly, the retinal thickness of Irbp-/-; Prss56-/- and Irbp-/-;Mfrp-/- mice was comparable to that of their Prss56 and Mfrp single mutant littermates (Irbp+/-;Prss56-/- and Irbp+/-;Mfrp-/-, respectively) (Fig. S8B, D). Together, these findings demonstrate that the excessive ocular elongation observed in Irbp-/- mice is dependent on PRSS56 and MFRP functions.
DISCUSSION
The molecular and cellular mechanisms involved in ocular axial growth and emmetropization are poorly understood. Previous studies have identified PRSS56 and MFRP mutations as a major cause of nanophthalmos, a condition characterized by severe ocular size reduction and extreme hyperopia, suggesting that these factors play a critical role in ocular axial growth[3-6]. Consistent with this, Prss56 and Mfrp mutant mice recapitulate the characteristic pathophysiological features of nanophthalmos, i.e. exhibit reduced ocular axial length and hyperopia[3, 13, 21]. Here, we use complementary genetic approaches in Prss56 and Mfrp mutant mouse models as a first step to elucidate the molecular and cellular factors playing a role in the ocular size regulation. Notably, we identified ADAMTS19 as a novel factor involved in ocular size regulation and demonstrate that the upregulation of retinal Adamts19 expression is part of a protective molecular response to impaired ocular growth. Also, we use a complementary strategy to show that inactivation of Prss56 or Mfrp prevents excessive ocular elongation in a mouse model of early-onset developmental myopia caused by a null mutation in Irpb. Overall, our findings suggest that PRSS56 and MFRP are not only necessary for supporting ocular axial elongation under normal conditions, but also in the context of childhood-onset high myopia.
Gene expression profiling led us to the identification of PRSS56 and ADAMTS19 two secreted serine proteases, whose expression is altered in the retina of mouse model recapitulating features of nanophthalmos. We have previously reported that increased retinal expression of Prss56 was a key molecular feature of Prss56 mutant mice exhibiting a reduction in ocular axial length [13]. Here, we show that Adamts19 expression is also upregulated in the retina of Prss56 mutant mice. Importantly, taking advantage of the Egr1;Prss56 double mutant mouse model in which Egr1 inactivation rescues the ocular size reduction caused by loss of PRSS56 function, we demonstrate that the increased expression of retinal Adamts19 results from ocular size reduction and is not a direct consequence of Prss56 mutation per se. Further to support this finding, we show that retinal Prss56 and Adamts19 mRNA levels are also upregulated in an independent mouse model of nanophthalmos caused by a null Mfrp mutation. Importantly, we show that ADAMTS19 is not required for ocular axial growth during normal development, however, Adamts19 inactivation exacerbates the reduction in ocular axial length and VCD in both Prss56 and Mfrp mutant mouse models. Collectively, these findings indicate that the upregulation of retinal Prss56 and Adamst19 expression constitutes a protective response to overcome impaired ocular axial growth in two distinct mouse models of nanophthalmos. Since both PRSS56 and ADAMTS19 belong to the family of secreted serine-protease, it raises the possibility that they likely have overlapping or redundant function(s) and share the same substrate(s), which might explain the compensatory effect of ADAMTS19 on ocular elongation in mutant mice lacking PRSS56. Interestingly a recent study has found an association between genetic variant near Adamts19 and ocular axial length, making our findings in mice relevant to human ocular size regulation [25].
Using a complementary genetic approach, we demonstrate that Prss56 and Mfrp inactivation prevents the excessive ocular axial growth observed in a mouse model of early-onset high myopia caused by a null mutation in Irbp [24]. In the currently accepted model of ocular axial elongation, signals originating from the retina must first be relayed to the RPE before being transmitted to the choroid and subsequently to the sclera to induce scleral ECM remodeling and ocular axial growth [14, 18]. Importantly, the expression pattern of Prss56 and Adamts19 in the retina and that of Mfrp in the RPE are consistent with a central role for the retina and RPE in promoting ocular axial growth [13, 16]. More specifically, the cellular localization of Prss56 and Adamts19 highlights the importance of Müller glia in mediating crosstalk between the retina and RPE. Interestingly, IRBP is localized in the interphotoreceptor matrix, a layer occupying the subretinal space juxtaposing the retinal photoreceptor cells and RPE. Thus, IRBP along with PRSS56 and MFRP may be part of a signaling network that not only connects the retina and RPE but also facilitates the flow of information, which are integral to ocular growth regulation. The increased expression of retinal Prss56 in Mfrp mutant mice further lends support to the existence of potential crosstalk between Müller glia and RPE in the regulation of ocular axial growth. Overall, our findings suggest that PRSS56 and MFRP are critical for ocular axial growth during early developmental stages. Furthermore, as PRSS56 and MFRP are localized in the Müller glia and RPE respectively, they point towards a role for the interplay between Müller glia and RPE in ocular axial growth regulation.
We have shown previously that the ocular expression of Prss56 is restricted to the retina, and predominantly observed in Müller glia [13]. Also, Prss56 upregulation was seen in retinal Müller glia of Mfrp mutant mice [23]. Here, we show that Adamst19 expression is specifically detected in the INL of the Prss56 mutant retina, a region where the cell body of Müller glia soma is found. These findings suggest that impaired ocular growth triggers the activation of a transcriptional program in retinal Müller glia leading to increased expression of Prss56 and Adamts19, two genes encoding secreted serine proteases. Müller cells have been postulated to play a role in the detection of subtle changes in retinal structure due to mechanical stretching of their long processes or side branches[26, 27]. Reduction in ocular size may alter the structural and mechanical properties of the retina that are sensed by Müller glia triggering transcriptional activation of factors participating in the regulation of ocular axial growth [27].
Since genetic variants of PRSS56 and MFRP are also associated with common forms of myopia[11, 17], it raises the possibility of whether a nexus between Müller glia and RPE may have a broader role that contributes to vision-guided postnatal ocular growth. Also, given that loss of PRSS56 and MFRP function causes a reduction in ocular length[6, 13], it is plausible their noncoding variants associated with myopia may cause an increase in gene expression and thus, act via gain of function mechanism thereby contributing to an opposite phenotype characterized by an increase in ocular axial length. Furthermore, supporting the role of PRSS56 in myopia pathogenesis, a recent study in marmoset has shown an increase in retinal expression of Prss56 in response to minus lens-induced axial elongation/myopic compared to the control eyes[28]. Future efforts will focus on determining the cellular and molecular basis of the potential crosstalk between Müller glia and RPE in the regulation of ocular axial growth and their relevance to axial elongation in the context of myopia.
In summary, we identify ADAMTS19 as a novel factor involved in ocular size regulation and use a distinct mouse model of hyperopia/reduced ocular size and myopia/ excessive ocular growth to describe a regulatory genetic network playing a central role in regulating eye growth during development and disease. Collectively, these findings raise the possibility that modulation of Adamts19 expression could be part of the general adaptive mechanism needed for regulating ocular axial growth. Furthermore, they suggest that PRSS56 and MFRP are indispensable for normal and aberrant ocular axial growth as a consequence of mutation in Irbp and point towards Prss56 and Mfrp likely being part of a sequential pathway necessary for supporting ocular elongation.
CONFLICT OF INTEREST STATEMENT
All authors declare for no conflict of interests in the study
AUTHOR CONTRIBUTIONS
SK and KSN conceived and designed the study. SK, CL-D, SP, and YZ performed experiments. SK, CL-D, YZ, and KSN interpreted the results of analyses on mouse study.SK, CL-D, KSN, contributed to the drafting of the original manuscript. SK, CL-D, and KSN critically reviewed the manuscript.
MATERIALS AND METHODS
Animals
All experiments were conducted in compliance with protocols approved by the Institutional Animal Care and Use Committee at University of California San Francisco (IACUC) (Protocols # AN181358-01D) and following the guidelines from the Association for Research in Vision and Ophthalmology’s statement on the use of animals in ophthalmic research. Animals were given access to food and water ad libitum and housed under controlled conditions including 12-h light/dark cycle per the National Institutes of Health guidelines. Both male and female mice were used in all experiments and no differences were observed between sexes, all comparisons were made between littermates to minimalize variability.
Mouse lines
Prss56-/- :(C57BL/6, Cg-Prss56glcr4/SjJ) – Mice carrying ENU induced mutation in Prss56 causing truncation of PRSS56 protein at its C-terminal region [3].
Prss56cre/cre : (C57BL/6.Cg-Prss56tm(cre)) – Mice carrying a null allele of Prss56 in which the exon1 of Prss56 is replaced by CRE recombinase sequence[13, 29].
Egr1-/- :(C57BL/6. Egr1tm1Jmi/J) - Egr1 mutant mice: C57BL/6. Egr1tm1Jmi/J, the targeted mutation by insertion of a PGK-neo cassette introduces stop codon resulting in protein truncation upstream of the DNA-binding domain[30].
Adamts19-/- :(Adamts19tm4a(EUCOMM)Wtsi)) - A conditional Adamts19 knockout mouse with LoxP sites flanking exon3. Excision of the LoxP sites by ubiquitously expressed CRE recombinase driven by beta-actin promoter leads to the generation of a knockout allele of Adamts19 (Supplementary Fig.2 A&B).
Mfrp-/- :(B6.C3Ga-Mfrprd6/J): The mouse strain is homozygous for rd6 exhibiting retinal degeneration around four weeks during retinal developmental phase [23].
Irbp-/- :(B6.129P2-Rbp3tm1Gil/J): A knockout mouse model Irbp(Interstitial retinal binding protein 3) gene. This mouse line carries a targeted mutation for the Rbp3 gene where the promoter and Exon 1 have been replaced by a NEO selection cassette rendering Irbp protein inactive.
PCR genotyping of all mouse strains was performed on genomic DNA obtained from tail biopsies digested with Proteinase K (Sigma, St. Louis, MO, USA) using primers listed in Table S1.
Ocular Biometry
Ocular biometry was performed using Envisu R4300 spectral-domain optical coherence tomography (SD-OCT, Leica/Bioptigen Inc., Research Triangle Park, NC, USA). Measurements of various ocular parameters including axial length, vitreous chamber depth (VCD), anterior chamber depth (ACD), lens diameter and retinal thickness were performed on mice anesthetized with ketamine/xylazine (100 mg/kg and 5mg/kg, respectively; intraperitoneal (IP)) following pupil dilation as described previously[13].
Quantitative polymerase chain reaction (qPCR)
For qPCR analysis of gene expression, eyes were enucleated and retinas were immediately dissected and total RNA was extracted from mouse retinal tissue using Qiagen RNeasy Mini Kit as per manufacturers protocol (Qiagen, Valencia, CA, USA) and reverse-transcribed using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA) and primer sets listed in Table S2. qPCR was performed on Bio-Rad C1000 Thermal Cycler/CF96 Real-Time System using SSOAdvanced™SYBR Green® Supermix (Bio-Rad, Hercules, CA, USA). Briefly, 100ng of cDNA and 10uM primers were used per reaction in a final volume of 20ul. Each cycle consisted of denaturation at 95°C for 15s, followed by annealing at 60°C for 15s, extension 72°C for 30s for a total of 39 cycles. All the experiments were run as technical duplicates and a minimum of three biological replicates were used per group. The relative expression level of each gene was normalized to housekeeping genes (Actinβ and Mapk1) and analyzed using the CFX Maestro software (Bio-Rad, Hercules, CA, USA).
In situ hybridization
Mice were transcardially perfused with ice-cold RNase-free PBS followed by 4% PFA (in RNase-free PBS). Eyes were enucleated post-fixed in RNAse-free 4% PFA, cryoprotected in 20% sucrose, and embedded in OCT and sectioned within 24 hours for in situ hybridization. QuantiGene View RNA (Affymetrix, Santa Clara, CA, USA). In situ hybridization was performed according to the manufacturer protocol. Briefly, 12µm cryosections were fixed overnight in 4% PFA, dehydrated through a graded series of ethanol, were subjected to 2X protease digestion for 10 minutes, fixed and hybridized with probe sets against Adamts19 (NM_175506 (Adamts19), TYPE1, high sensitivity with 40∼50 bp DNAs) for 3 hours at 40°C using a ThermoBrite system (Abbott Molecular, Des Plaines, IL, USA). Cryosections were then washed and subject to signal amplification and detection using a fast red substrate, counterstained and mounted for subsequent imaging. Fluorescent images were acquired using an AxioImager M1 microscope equipped with an MRm digital camera and AxioVision software, with an LSM700 confocal microscope and Zen software (Carl Zeiss Microscopy, LLC, Germany).
Statistical analyses
Statistical comparisons between mutant and control groups or between multiple experimental groups at a given age were performed using two-tailed unpaired Student’s t-test and one-way ANOVA, respectively, using Prism statistical software (version 6.02, GraphPad Software, San Diego, CA). A p-value of < (0.05, 0.01 and 0.001) was considered significant.
ACKNOWLEDGMENTS
This work was made possible, in part, by NEI P30 EY002162 - Core Grant for vision research (UCSF, Ophthalmology), Research to Prevent Blindness unrestricted grant (UCSF, Ophthalmology) and William and Mary Greve Special Scholar Award, That Man May See Inc, Research Evaluation and Allocation Committee (REAC)-Tidemann fund, and Marin Community Foundation-Kathlyn McPherson Masneri and Arno P. Masneri Fund (KSN) as well as by the Knight Templar Eye Foundation Career Starter Award (SK). The authors would like to acknowledge Ms. Vivian Chi and Mr. Yusef Seymens with mouse genotyping assay and general laboratory care.
Footnotes
↵* Joint first authors