1. Introduction
Inherited retinal dystrophies (IRDs) result in irreversible loss of vision due to the progressive degeneration of the retina. IRDs are considered genetically heterogeneous. There are over 280 genes involved in retinal degeneration, which can present with autosomal dominant, autosomal recessive, X-linked, mitochondrial, and complex modes of inheritance [
1,
2,
3,
4,
5] (
https://sph.uth.edu/retnet/home.htm accessed on 1 January 2022). Mouse models have been instrumental in elucidating the mechanisms by which these important genes lead to pathology [
2]. A large number of mouse models of inherited retinal degeneration have been developed by introducing mutations in IRD-associated genes, and several naturally occurring mouse models with retinal pathology have also been characterized [
6,
7,
8,
9,
10]. In this paper, we describe a novel mouse model with a loss of ASRGL1 expression caused by a c.578_579insAGAAA (NM_001083926.2) mutation introduced by editing the genome using CRISPR/Cas9.
A homozygous missense mutation G178R in the human asparaginase like-1 gene (
ASRGL1) is associated with early onset retinal degeneration in a consanguineous Pakistani pedigree [
11].
ASRGL1 encodes an enzyme that is synthesized as an inactive precursor protein that becomes active after undergoing autocatalytic intramolecular processing, with the active form able to catalyze the hydrolysis of L-asparagine and isoaspartyl-peptides [
11,
12]. Expression studies revealed that there are high levels of
ASRGL1 in the optic nerve, retina, heart, and brain, with low to minimal levels in all other tissues in mice [
11]. ASRGL1 localizes to the photoreceptor layer and is known to be highly expressed in the retina at older ages, giving it a putative role in the maintenance of aging retinal tissue. The G178R mutation detected in patients with early onset IRD affects the autocatalytic activity of ASRGL1 [
11]. Despite this progress, the molecular mechanism underlying ASRGL1’s involvement in retinal degeneration is unknown. In this study, we showed that a mouse model with a homozygous null allele provides key insights into the role of
ASRGL1 in retinal degeneration.
2. Methodology
Generation of Asrgl1mut/mut mouse model using CRISPR/Cas9 system: The
Asrgl1 gene ablated mouse model (
Asrgl1mut/mut) was generated under the authorization of the Institutional Animal Care and Use Committee (IACUC) of the University of California San Diego using a CRISPR/Cas9 system (
Figure 1A) [
13,
14]. For genome editing, a concentrated 25 μL cocktail of 100 ng/μL single-stranded oligodeoxynucleotides (ssODNs) as a repair template was used along with 20 ng/μL sgRNAs and 50 ng/μL Cas9 mRNA. The reverse complementary ssODNs sequence is shown as follows: 5′—(~60—nt 5′ homology) CCAACTCGGCCAACCATTTTATTGACAATCCCCCCAGTAGAGGTTGCGTAAGCCAAGTTTCTTCTGCAGTCCAAGGCAACAGCACCCACAGTTCCTGAGTTTctgtcaaaaataagcagaaaagggaaggtaatgaagatgagaaacagcccctaagcaacactgtcccttatgtacc (~60—nt 3′ homology)—3′. This CRISPR/Cas9 cocktail was microinjected into the embryos of C57BL/6 mice.
Genotyping Asrgl1 mut/mut mice: Three unique primers located in the region flanking the mutation or insertion site were generated to amplify either the wildtype or mutant
Asrgl1 allele to determine the genotype of
Asrgl1mut/mut mice (
Figure 2,
Table 1). Amplification reaction with the forward primer “mAsrgl1 wt F” and the reverse primer “mAsrgl1 uni R”, indicated as primer set A, generated a 328 bp PCR product of the wildtype allele (
Table 2), whereas amplification of the
Asrgl1mut/mut allele with primer set B (“mAsrgl1 mut F” and “mAsrgl1 uni R”) generated a 333 bp product. Sanger sequencing of these amplification products enabled the detection of the specific sequence alteration to confirm the genotype (
Figure 1B and
Figure 2). These mice were also genotyped for
Rd1 and
Rd8 as previously described [
15].
Ophthalmic evaluation by in vivo imaging: Mice at desired ages (3, 5, and 10 months) were anesthetized with a mixture of ketamine (93 mg/kg body weight) and xylazine (8 mg/kg body weight) intraperitoneally administered following the UCSD’s Controlled Substances Use Authorization (CSUA). During this procedure, mice were kept on a heating pad at 37 °C to maintain normal body temperature. Prior to imaging, the eyes were dilated with topical proparacaine hydrochloride (0.5%) (NDC: 17478-263-12, Akorn, Inc. Lake Forest, IL, USA) and topical tropicamide (0.5%) (NDC 17478-101-12, Akorn, Inc. Lake Forest, IL, USA), followed by phenylephrine (2.5%) (NDC: 17478-201-15, Akorn, Inc. Lake Forest, IL, USA). For imaging, 3-, 5-, and 10-month-old Asrgl1mut/mut mice and 10-month-old wildtype control mice were used. Six mice of each genotype were used for ophthalmic evaluation at each age point.
A Spectralis™ HRA + OCT machine (Heidelberg Engineering Inc., Heidelberg, Germany) was used for scanning laser ophthalmoscopy, infrared, and spectral domain optical coherence tomography imaging. To improve imaging, a custom-made sterilized PMMA contact lens for mice was used to keep eyes moist and lubricated in order to elude cataractogenesis caused by drying of the surface of the eye. To measure retinal thickness, Heidelberg eye explorer software (Heidelberg Engineering, Inc. Heidelberg, Germany) was used. Three-, five-, and ten-month-old Asrgl1mut/mut mice were compared with age-matched wildtype controls to study retinal changes by using fundoscopy and fundus autofluorescence.
Immunohistochemistry: Cryosections of 3-, 9-, 13-, and 17-month-old
Asrgl1mut/mut and 12-month-old wildtype C57BL/6 mouse eye were used to perform immunohistochemistry as previously described [
15]. Rabbit anti-ASRGL1 antibodies (1:100) (ProteinTech Cat-11400-1-AP), anti-rabbit polyclonal OPN1MW (1:200) (AB5405, Millipore Sigma, Burlington, MA, USA), anti-goat polyclonal OPN1SW (1:200) (Santa Cruz Biotechnology, Dallas, Texas), anti-mouse rhodopsin monoclonal antibody (1:200) (MA1-722, Thermofisher Scientific, Carlsbad, CA, USA), anti-mouse ATP1A1 monoclonal antibody (1: 200) (M7-PB-E9, Thermofisher Scientific, Carlsbad, CA, USA), AlexaFluor-488-conjugated donkey anti-rabbit secondary antibody (1:2000) (Invitrogen, Carlsbad, CA, USA), AlexaFluor-555-conjugated donkey anti-goat secondary antibody (1:3000) (Invitrogen, Carlsbad, CA, USA), and AlexaFluor-555-conjugated donkey anti-mouse secondary antibody (1:3000) (Invitrogen, Carlsbad, CA, USA) were utilized for staining. The nuclei were stained using a Vecta shield DAPI medium (H-1500-10, Vector Laboratories, San Diego, CA, USA). Images were captured using a Nikon confocal microscope system (A1R STORM, Nikon; Melville, NY 11747, USA) at 60× objective lens (0.85 numerical aperture) and processed using Adobe Photoshop CS6 (Adobe Associates, Inc. Santa Rosa, California).
Quantitative transcript expression: The retinal layer was dissected from 3- and 12-month-old
Asrgl1mut/mut and 12-month-old wildtype control mice. Isolation of retinal RNA followed by reverse-transcription reaction and calculation of
Opn1mw,
Opn1sw, and
Rhodopsin expression relative to the housekeeping genes
Gapdh and
Actb were performed as described earlier [
16].
Quantification of asparaginase activity: Asparaginase enzymatic activity of the 11- to 15-month-old Asrgl1mut/mut and 10-month-old wildtype control mice was measured in freshly dissected retinal tissue lysates using an Asparaginase Activity Assay Kit (ab107922, Abcam, Cambridge, MA, USA).
4. Discussion
Patients with the G178R mutation in
ASRGL1, which results in impaired asparaginase activity, show an onset of vision abnormalities in the first decade of life including decreased rod and cone response with progressive vision loss. Night blindness is reported to be among the first symptoms in these patients, suggesting rod photoreceptor abnormalities at early stages [
11]. According to the findings, the rod and cone photoreceptors in the
Asrgl1 ablated mouse model developed progressive and significant structural and functional abnormalities beginning as early as 3 months of age (
Figure 3). Specifically, a significant loss of s-opsin-expressing cones was observed in 3-month-old mice, while significant abnormalities in both m-opsin- and s-opsin-expressing cones were noted from 9-month-old mice (
Figure 3 and
Figure 5). Overall, the retinal phenotype of the
Asrgl1 ablated mouse model recapitulated the cone–rod dystrophy phenotype observed in patients with the G178R mutation in
ASRGL1.
The sequence change introduced in
Asrgl1mut/mut mice is predicted to result in a frameshift mutation leading to nonsense-mediated decay of the transcript (
Figure 1). The minimal levels of the ASRGL1 transcript and protein (
Figure 4) as well as the low levels of asparaginase activity observed in the retina of
Asrgl1 gene ablated mice (
Figure 6) indicate loss of the transcript and, consequently, the enzymatic activity. Earlier in vitro studies revealed a significant loss of asparaginase activity due to the G178R mutation that was detected in patients with retinal degeneration and suggested that appropriate retinal structure and function are dependent on ASRGL1 activity [
11]. The early onset and progressive rod and cone photoreceptor abnormalities observed in the
Asrgl1mut/mut mutant mouse model with minimal to undetectable levels of
Asrgl1 expression and asparaginase activity in the retina further support a critical role for ASRGL1 in the normal function of the retina.
The specific biological role of ASRGL1 and the impact of the loss of this protein in photoreceptors are unknown. The ASRGL1 protein has both L-asparaginase and β-aspartyl peptidase activities [
11,
12,
17,
18]. This protein is suggested to play a role in the production of L-aspartate, which acts as a neurotransmitter in the brain [
12]. The expression of
ASRGL1 was reported in multiple tissues including adipose, brain, mammary glands, cervix, fallopian tube, lung, kidney, prostate, testes, cervix, and uterus (
https://gtexportal.org/home/ accessed on 1 January 2022). Several studies demonstrated a strong association between high levels of ASRGL1 expression and tumorigenesis [
12,
19]. L-asparaginase is widely used in chemotherapy to treat adults and children with acute lymphoblastic leukemia [
20,
21]. ASRGL1 is widely expressed in non-ocular tissues, and while tissues other than the retina of the
Asrgl1 gene ablated mouse model have not been investigated in detail, a gross examination of the major organs of these mice did not reveal significant morphological abnormalities. Therefore, a significant decrease in the activity of
ASRGL1 or the expression of the
ASRGL1 transcript at minimal to undetectable levels may not result in major morphological consequential abnormalities in mice other than the retinal abnormalities that were evaluated in depth.
In a recent study, Zhou et al. reported that the deletion of exons 3 and 4 of
Asrgl1 led to late-onset photoreceptor abnormalities beginning from 8 months of age in a mouse model [
22]. The retinal structure and expression of marker genes were normal up to 8 months without increased TUNEL-positive cells compared with controls. The ERG responses were slightly abnormal at 8 months, while they were reported to be normal at 6 months. A progressive and severe reduction in the outer nuclear layer was reported from 9 to 15 months. Overall, these mice were reported to have normal retinal structure and function up to 6 months with a progressive photoreceptor loss starting from 8 months of age. In contrast, the
Asrgl1mut/mut mouse model with the homozygous c.578_579insAGAAA insertion developed photoreceptor abnormalities starting from 3 months of age including abnormal photopic ERG response and a significant decrease in the expression of rod and cone photoreceptor marker genes. The presence of abnormal and significantly short cone outer segments was observed at 9 months in
Asrgl1mut/mut mice, whereas the
Asrgl1 KO mice reported by Zhou et al. developed similar changes in cone morphology at 12 months [
22]. The major genotype difference between these models is the presence of the
Asrgl1 transcript without exons 3 and 4 in
Asrgl1 KO mice vs. the loss of the
Asrgl1 transcript and protein in the
Asrgl1mut/mut mouse model. The impact of shorter
Asrgl1 transcript without exons 3 and 4 to the phenotype in
Asrgl1 KO mice is unknown. Comparison of the retinal changes between the
Asrgl1mut/mut mouse model described in this study and the
Asrgl1 KO model reported by Zhou et al. revealed that the mice in both models developed progressive retinal degeneration involving both rod and cone photoreceptor loss [
22], further supporting a key role for ASRGL1 in the retina and providing proof that a lack of ASRGL1 leads to retinal pathology.
The onset of retinal degeneration is observed to be earlier in the Asrgl1mut/mut mouse model at 3 months compared with the late-onset degeneration reported in the Asrgl1 KO model at 8 months. However, the photoreceptor loss is much more severe in the Asrgl1 KO mice involving the loss of both rods and cones by 15 months, whereas the Asrgl1mut/mut mice developed severe cone abnormalities by 3 months, while the ONL thickness appeared to be normal at 17 months, suggesting a less severe rod loss. The difference observed in the phenotype of these two mouse models, one (Asrgl1mut/mut) with predominant cone loss and mild rod abnormalities and the other (Asrgl1 KO) with severe rod–cone degeneration despite the loss of the Asrgl1 transcript in both, may be due to the contribution of the genetic background. Comparative analysis of these two models may provide an opportunity to understand the underlying cause of severe cone degeneration sparing rods in the Asrgl1mut/mut model even at older age.
Earlier studies on a zebrafish knockdown model injected with either start site morpholino (MO) or splice site MO of
asrgl1 did not reveal gross morphological abnormalities including lack of retinal abnormalities on 6 dpf [
11]. However, the zebrafish model injected with human mutant hASRGL1 mRNA showed a loss of blue-opsin-expressing cones. Evaluation of zebrafish
Asrgl1 knockout and mutant knock-in models developed by editing the genome is needed to compare the impact of
Asrgl1 gene ablation on rod and cone photoreceptors in the retina of zebrafish.
While the
Asrgl1 gene ablated mouse models establish the involvement of this gene in retinal degeneration, the specific role of ASRGL1 and the mechanism underlying the progressive retinal degeneration due to the loss of ASRGL1 expression and/or activity are unknown. This novel
Asrgl1mut/mut mouse will serve as a model to understand the role of ASRGL1 in the retina. Furthermore, this novel
Asrgl1mut/mut mouse recapitulating the phenotype of patients with G178R ASRGL1 mutation will serve as a model for preclinical evaluation of therapeutic strategies to treat patients with mutations specifically in ASRGL1 and cone–rod dystrophy in general [
11]. Additionally, these mice may also serve as a model to study the role of ASRGL1 in tumorigenesis and related pathologies.