Wiechmann AF et al. (2020), CRISPR/Cas9 mediated mutation of the mtnr1a mel...

XB-ART-57259

Sci Rep 2020 Aug 13;101:13757. doi: 10.1038/s41598-020-70735-2.

Show Gene links Show Anatomy links

CRISPR/Cas9 mediated mutation of the mtnr1a melatonin receptor gene causes rod photoreceptor degeneration in developing Xenopus tropicalis.

Wiechmann AF , Martin TA , Horb ME .

Abstract
Nighttime surges in melatonin levels activate melatonin receptors, which synchronize cellular activities with the natural light/dark cycle. Melatonin receptors are expressed in several cell types in the retina, including the photon-sensitive rods and cones. Previous studies suggest that long-term photoreceptor survival and retinal health is in part reliant on melatonin orchestration of circadian homeostatic activities. This scenario would accordingly envisage that disruption of melatonin receptor signaling is detrimental to photoreceptor health. Using in vivo CRISPR/Cas9 genomic editing, we discovered that a small deletion mutation of the Mel1a melatonin receptor (mtnr1a) gene causes a loss of rod photoreceptors in retinas of developing Xenopus tropicalis heterozygous, but not homozygous mutant tadpoles. Cones were relatively spared from degeneration, and the rod loss phenotype was not obvious after metamorphosis. Localization of Mel1a receptor protein appeared to be about the same in wild type and mutant retinas, suggesting that the mutant protein is expressed at some level in mutant retinal cells. The severe impact on early rod photoreceptor viability may signify a previously underestimated critical role in circadian influences on long-term retinal health and preservation of sight. These data offer evidence that disturbance of homeostatic, circadian signaling, conveyed through a mutated melatonin receptor, is incompatible with rod photoreceptor survival.

PubMed ID: 32792587
PMC ID: PMC7426423
Article link: Sci Rep
Grant support: [+]

Species referenced: Xenopus tropicalis
Genes referenced: mtnr1a mtnr1b pam psmd6 rpe
GO keywords: photoreceptor cell development
Antibodies: Mtnr1b Ab1
gRNAs referenced: mtnr1a gRNA1 mtnr1a gRNA2 mtnr1a gRNA3 mtnr1a gRNA4

Phenotypes: Xtr + mtnr1a sgRNAs (Fig 4 a b ) [+]

Article Images: [+] show captions


	Figure 2. Identification of heterozygous F1 animals with CRISPR/Cas9 indel mutations of the mtnr1a gene. (a) Representative RFLP analysis shows that PCR products of mtnr1a VIL heterozygotes display two predicted cleavage bands (arrows), whereas WT and non-VIL mutant DNA is not cleaved. The creation of a BseRI site by the VIL deletion is illustrated in Supplementary Fig. S3. (b) Direct PCR Sanger sequencing chromatogram of F1 WT progeny display prominent peaks for all bases in the T2 sgRNA target area. The T2 PAM sequence (CCT) is boxed in black, and the downstream 9-bp sequence of the mtnr1a VIL deletion mutation is boxed in blue. The corresponding amino acid sequence is shown below the nucleotide codon sequence. (c) In an F1 tadpole heterozygous for the mtnr1a 9-bp (VIL) deletion, the trace decomposes (i.e., mixed base calls) at the predicted target site 4-bp downstream of the T2 PAM site, and continues to the end of the trace, as expected for a heterozygous indel mutation. The decomposed sequence represents the overlapping peaks of the mtnr1a WT and VIL alleles. (d) Partial view of alignments from Poly Peak Parser (https://yosttools.genetics.utah.edu/PolyPeakParser/) output revealing the heterozygous 9-bp mtnr1a VIL deletion mutation. The WT allele is the top line and the mutant allele is the bottom line.


	Figure 5. Rods are dystrophic in stage 54–56 (4-month-old) F2 heterozygous but not homozygous mtnr1a VIL mutant tadpoles (n = 5). F2 heterozygous and homozygous tadpoles were generated by crossing two F1 VIL mutant heterozygotes. (a) WT siblings of F2 mutants exhibit healthy-appearing rod and cone photoreceptors. (b) Central retinas of VIL mutant heterozygous tadpoles display ROS dystrophy and abnormal distribution of RPE cells, whereas cones appear relatively normal (white arrows). (c) Retinas of homozygous VIL mutant siblings display morphologies similar to WT siblings. (d) Quantitative measurements of rod dystrophy in X. tropicalis retinas show that rod outer segment (ROS) length is diminished significantly by 24% in VIL heterozygous 4-month-old tadpoles, compared to WT when analyzed by an unpaired two-tailed t-test (*; p = 0.0267, n = 3). The 15% decline in ROS length in VIL homozygous tadpoles compared to WT was statistically insignificant (p = 0.0867, n = 3). No significant differences in rod (i.e., ROS) or cone abundance were observed among the three genotype groups. Measurements were made of digital images of groups of 4-month-old tadpoles represented in a–c. Error bars indicate standard error of the mean. RPE retinal pigment epithelium, ROS rod outer segments, ONL outer nuclear layer, INL inner nuclear layer, cone inner segments; white arrows. Magnification bars = 20 µm.
	Figure 6. Retinas of heterozygous and homozygous mtnr1a VIL mutant tadpoles display Mel1a protein expression at intensities similar to WT siblings. Double-label confocal immunohistochemistry (IHC) of (a,b) WT, (c,d) F2 VIL heterozygous, and (e,f) F2 VIL homozygous tadpole head paraffin sections was performed with antibodies to X. laevis Mel1a (red label) and Mel1b (green label) melatonin receptors. Panels on the right are higher magnifications of areas in panels on the left side of the figure. Mel1a receptor fluorescent signal intensity and location appears to be similar in animals of all three genotypes. The Mel1a receptor is expressed in RPE, rod and cone photoreceptors (arrows), and inner retinal neurons. The Mel1a and Mel1b receptors display an expression pattern different from each other, although there are areas in which the red and green labels merge as a yellow label, suggestive of co-expression and/or very close proximity to each other. (g,h) WT retina sections processed for IHC with Mel1b antibodies, but without Mel1a antibodies (No 1° red) display specific immunofluorescence for Mel1b (green), but not for Mel1a (red). RPE retinal pigment epithelium, ROS rod outer segments, ONL outer nuclear layer, INL inner nuclear layer, IPL inner plexiform layer, arrows; rod photoreceptor inner segments. Magnification bars = 20 µm.
	Figure 7. Heterozygous mtnr1a VIL mutant young adult (juvenile) frogs display normal rods and cones. (a) WT juvenile frogs display normal rods (ROS) and cones. (b) F1 mtnr1a VIL heterozygous mutant juvenile sibling frogs display healthy-looking rods and cones, although the ROS density appears to be lower than in WT (asterisks). Mtnr1a 56-bp deletion mutant tadpoles do not display rod dystrophy. (c) Four-month-old WT tadpole retinas and (d) F1 mtnr1a heterozygous 56-bp deletion (potential knockout) siblings exhibit healthy-appearing rods (ROS) and cones (white arrows). RPE retinal pigment epithelium, ROS rod outer segments, ONL outer nuclear layer, INL inner nuclear layer. Magnification bars = 20 µm.
	Figure 8. Amino acid alignment of the endofacial portion of the first transmembrane domain of human and X. tropicalis melatonin receptors. The hydrophobic residues of transmembrane domain 1 (TM-1) located near the membrane cytoplasmic surface is highly conserved among GPCRs, as exemplified by the alignment of this region in the human MT1 and MT2 receptors and the homologous X. tropicalis Mel1a and Mel1b receptors. The human MT1 and Xenopus Mel1a receptors show a 100% amino acid identity in the GNLLVILSV 9-bp region in TM-1. The first intracellular loop (ICL-1) is also highly conserved, as indicated in yellow. The mtnr1a T2 sgRNA most effectively targeted the hydrophobic LLVIL residues (see Supplementary Fig. S6) in the center of the Xenopus Mel1a endofacial TM-1 region illustrated here. This region is the site of a human MT1 nonsynonymous point mutation (I49N) associated with autism59 and a human MT2 mutation (L60R) linked to type 2 diabetes57. The amino acid point mutations associated with each disorder are indicated by a blue box.
	Figure S2. Cas9 in vitro cleavage assay with sgRNAs. Genomic DNA from a Nigerian strain wild type (WT) X. tropicalis adult was used as the PCR template. DNA PCR products were incubated with sgRNA and Cas9 protein, and then separated on an agarose gel. PCR products of WT gDNA were cleaved into the predicted sizes of DNA fragments (white arrows) when incubated in vitro with Cas9 protein (T1, T2, and T3 lanes). The T1 and T2 target sites are in close proximity to each other on exon 2, so the Cas9 cleavage could be detected using the same PCR primer pairs (T12). The T3 target site was in exon 3, so a different set of PCR primers was used to amplify that region. PCR products that were not incubated with the Cas9 enzyme (--) displayed no cleavage bands.
	Figure S4. Genotyping assays identify adult F0 X. tropicalis sibling crispants with mtnr1a gene VIL mutations. (a) Representative T7 endonuclease I (T7EI) mismatch assay identifies mosaic F0 animals that express T2 mtnr1a indels. Denatured/reannealed amplicons are cleaved by T7E1 at the site of mismatches due to mutations in one strand, which generate DNA cleavage fragments on agarose gels. The crispants in this assay are assigned the same numbers as in supplementary figures S5 and S6, in which crispant #2 is the male founder of the F1 mtnr1a mutant progeny. All of the crispants shown here are mutants, based on the presence of PCR cleavage fragments when incubated with T7EI (+; arrows). None of the samples are cleaved in the absence of the T7E1 enzyme (--), and the DNA of normal homozygous (WT) progeny is not cleaved by T7EI (+). The image was enhanced digitally in Photoshop for contrast and brightness, and to mask some undesirable flecks of ethidium bromide precipitate. (b) Direct PCR Sanger sequencing chromatogram of WT progeny displays prominent peaks for all bases in the T2 sgRNA target area. The T2 PAM sequence (CCT) is boxed in black. (c) Direct PCR sequencing chromatogram of an F0 mosaic founder (crispant #2 in supplementary figures 5 and 6) shows pronounced sequence decomposition beginning 4-bp downstream of the T2 PAM site (arrow), representing the presence of a mixture of WT and mutant mtnr1a T2 target cDNAs from the gDNA of the mosaic tissue sample.
	Figure S5. Comparison of the editing efficiency of the mtnr1a T2 sgRNA in F0 mutant animals. Trace decomposition yields the spectrum of indel size and frequency occurring within a 0 to 12- bp indel size range. (a) The F0 crispant #2, which is the male founder of all of the F1 progeny described in this study, does not display a 9-bp deletion in this TIDE analysis of PCR products from gDNA extracted from a pair of web clips. (b-d) F0 sibling crispants #3-5 all display a 9-bp deletion at various frequencies, with crispant #3 having the highest frequency. (e) For comparison, PCR of web clip gDNA from an F1 heterozygous VIL mutant frog displays the expected indels only at 0 and -9-bp. (f) A sibling F1 homozygous VIL mutant frog displays a -9-bp indel only, as anticipated. TIDE analysis was performed using http://tide.nki.nl [81].
	Figure S7. Identification of heterozygous F1 animals with CRISPR/Cas9 indel mutations of the mtnr1a gene. Representative T7E1 assay of two heterozygous mutant mtnr1a F1 and two F1 WT sibling tadpoles. PCR products amplified from gDNA extracted from F1 progeny were denatured to separate the cDNA strands, then re-annealed slowly to promote formation of mismatched WT/mutant double-stranded cDNA. Mismatched sites of re-annealed WT and mutant DNA strands in heterozygous tadpoles are cleaved by T7E1 (+), resulting in two cleavage products (arrows) in a tadpole with the 9-bp (non-frameshift VIL) mtnr1a deletion, and for comparison, in another tadpole with a 3-bp substitution (predicted amino acid change is L→S; see supplementary figure S7). The DNA of normal homozygous (WT) progeny is not cleaved by T7E1 (+), and none of the samples are cleaved in the absence of the enzyme (--).
	Figure S8. A novel restriction fragment length polymorphism (RFLP) assay uniquely identifies VIL mutants. Left panel: Schematic representation of exon 2 of the mtnr1a gene, with the TM-1 domain flanked by the amino-terminus (N-Term) and first intracellular loop (ICL-1) domains. The nucleotide sequence targeted by the T2 sgRNA at the lower area of the panel is indicated in red, with the concomitant amino acid sequence below. The protospacer-adjacent motif (PAM) sequence, which directs Cas9 targeting, is indicated in green. Some flanking nucleotide sequence is indicated in blue. Right panel: The VIL deletion merges the upstream CTCCT sequence with the downstream CT sequence to create a unique BseRI restriction site.
	Figure S9: Sequence alignment of plasmid clones of mtnr1a F1 progeny. Sanger sequencing of 22 plasmid clones of F1 with mtnr1a T2 progeny revealed that 54.5% (12/22) of clones from mutant animals expressed the VIL deletion. No frameshift mutations were observed in these samples.

	Figure S11. Identification of F2 heterozygous and homozygous mtnr1a VIL mutant tadpoles. (a,b) T7E1 assay of ten random sibling F2 tadpole progeny from a crossing of two heterozygous mtnr1a VIL mutant frogs. PCR products amplified from gDNA were denatured and re-annealed slowly to promote formation of mismatched WT/mutant double-stranded cDNA. Mismatched sites are cleaved by T7E1 (+; arrows) in VIL mtnr1a heterozygotes. The DNA of WT (homozygous; lanes 1 and 6) or homozygous VIL mutant (lanes 2, 4, and 9) siblings is not cleaved by T7E1 (+), and none of the samples are cleaved in the absence of the enzyme (--). In this assay, tadpoles 3, 5, 7, 8, and 10 are identified as VIL heterozygotes (arrows). (c) BseRI RFLP analysis of the same PCR products as in (a,b). Mtnr1a VIL heterozygotes (lanes 3, 5, 7, 8, and 10) and homozygotes (lanes 2, 4, and 9) display two predicted cleavage bands (arrows), whereas WT DNA (lanes 1 and 6) is not cleaved. Note that the amount of uncleaved PCR template of the VIL homozygous mutants is much lower than in the homozygotes and WT, whereas the amount of cleavage product is higher than the others, illustrating that both alleles in homozygous mutants are cleaved in this assay. Illustration of the creation of a BseRI site by the VIL deletion is in Supplementary Figure S8. (d) Direct PCR Sanger sequencing chromatogram (DSP assays) of F1 WT progeny display prominent peaks for all bases in the T2 sgRNA target area. The T2 PAM sequence (CCT) is boxed in black, and the downstream 9-bp sequence of the mtnr1a VIL deletion mutation is boxed in red. The corresponding amino acid sequence is shown below the nucleotide sequence. (e) The boxed 9-nucleotide sequence in (d) is missing in the F2 mtnr1a VIL chromatogram, indicating that the 9-bp deletion is present on both alleles. The deleted sequence is boxed in blue below the chromatogram. (f) In F2 mutants that heterozygous for the mtnr1a 9-bp (VIL) deletion, the trace decomposes (i.e., mixed base calls) at the predicted target site 4-bp downstream of the T2 PAM site, and continues to the end of the trace, as expected for a heterozygous indel mutation. The decomposed sequence represents the overlapping peaks of the mtnr1a WT and VIL alleles. (g) Partial view of alignments from Poly Peak Parser (http://yosttools.genetics.utah.edu/PolyPeakParser/) [82] output displaying the heterozygous 9-bp mtnr1a VIL deletion mutation. The WT allele is the top line and the mutant allele is the bottom line.
	Figure S12. Alignment of WT and F1 X. tropicalis mtnr1a heterozygous 56-bp deletion mutant. The upper sequence is WT with hypothetical amino acids in the non-coding intron corresponding to the codon below indicated in blue. The mutant is the lower sequence, with the hypothetical amino acid sequence centered over the corresponding codon indicated in red. The T2 sgRNA target sequence is underlined with the PAM sequence underlined in bold. The 56-bp deletion occurs at the boundary of exon 2 and intron 2 (Exon 2/Intron 2 boundary), and is indicated by red dashes (---). Since 56 is not a multiple of three, the deletion causes a hypothetical shift in the reading frame. Note that the amino acids indicated in blue or red downstream of the mutation are not expected to be expressed, and are shown only to illustrate the misalignment of nucleotides due to the putative frameshift. Since the exon/intron splice site is skipped in the mutant due to the 56-bp deletion, the non-coding region downstream of the deletion could theoretically encode aberrant protein, indicated in red. No stop codons were observed in the mutant sequence downstream of the deletion.

References [+] :

Anders, In vitro enzymology of Cas9. 2015, Pubmed