Abstract

Inherited retinal dystrophies are clinically and genetically heterogeneous with significant number of cases remaining genetically unresolved. We studied a large family from the West Indies islands with a peculiar retinal disease, the Martinique crinkled retinal pigment epitheliopathy that begins around the age of 30 with retinal pigment epithelium (RPE) and Bruch's membrane changes resembling a dry desert land and ends with a retinitis pigmentosa. Whole-exome sequencing identified a heterozygous c.518T>C (p.Leu173Pro) mutation in MAPKAPK3 that segregates with the disease in 14 affected and 28 unaffected siblings from three generations. This unknown variant is predicted to be damaging by bioinformatic predictive tools and the mutated protein to be non-functional by crystal structure analysis. MAPKAPK3 is a serine/threonine protein kinase of the p38 signaling pathway that is activated by a variety of stress stimuli and is implicated in cellular responses and gene regulation. In contrast to other tissues, MAPKAPK3 is highly expressed in the RPE, suggesting a crucial role for retinal physiology. Expression of the mutated allele in HEK cells revealed a mislocalization of the protein in the cytoplasm, leading to cytoskeleton alteration and cytodieresis inhibition. In Mapkapk3−/− mice, Bruch's membrane is irregular with both abnormal thickened and thinned portions. In conclusion, we identified the first pathogenic mutation in MAPKAPK3 associated with a retinal disease. These findings shed new lights on Bruch's membrane/RPE pathophysiology and will open studies of this signaling pathway in diseases with RPE and Bruch's membrane alterations, such as age-related macular degeneration.

Introduction

Inherited retinal dystrophies are a group of clinically and genetically heterogeneous disorders, a number of which remains in part genetically unresolved, as, for example, retinitis pigmentosa. Increasing knowledge on genes specifically expressed in the retina and on their associated pathogenic mechanisms is a permanent challenge for a better understanding of these disorders and for the design of therapies. Following the identification of the major genes mutated in inherited retinal dystrophies during the last three decades using classical genetic approaches, techniques of whole-exome sequencing (WES) recently allowed accessing more private genes and mutations.

We report here the genetic and pathogenic data of an inherited dystrophy characterized by alterations of the Bruch's membrane and the retinal pigment epithelium (RPE) resembling a dry desert land on fundus examination (Fig. 1). This disease, the Martinique crinkled retinal pigment epitheliopathy (MCRPE), is an autosomal dominant retinal dystrophy that we clinically described in 2013 in three families of Martinique, one of the French West Indies islands (1). These three families have indeed the same ancestor background. Briefly, the disease appears at 30–40 years of age and worsens progressively featuring extending lesions from the posterior pole to the periphery of the retina (stages 1 and 2, Fig. 1) and later polypoidal choroidal vasculopathy (PCV), choroidal neovascularization or atrophic/fibrous macular scar causing a decrease in visual acuity after the age of 50 (stage 3, Fig. 1). MCRPE ends with a retinitis pigmentosa-like phenotype (stage 4, Fig. 1) related to RPE and photoreceptor cell death. A specific hallmark of the disease is the dry desert land pattern on fundus examination. This pattern corresponds to the irregular thickness of the Bruch's membrane and the RPE, with a scalloped elevation of the RPE, noted by spectral domain-optical coherence tomography analysis (SD-OCT, Fig. 2). Full-field electroretinogram (ERG) can be normal at pre-clinical and early stages of the dystrophy, but later, cone and rod responses become severely reduced, in keeping with the progressive photoreceptor cell dysfunction and death at the final stage. No other associated symptom has ever been reported for MCRPE individuals.

Figure 1.

Phenotypic characteristics of a unique retinal dystrophy, the MCRPE. Fundus pictures and disease progression with age, from pre-clinical stage with a normal fundus to retinitis pigmentosa. After a pre-clinical stage with a normal fundus, a dry desert land pattern appears. It first involves the posterior pole in the temporal part of the fovea in the case presented (stage 1, blue arrows) and slowly progresses with age to the peripheral retina with a bilateral involvement (stage 2). Stage 3 is defined by a macular and peripheral dry desert land pattern combined with a retinal complication. In the left eye of this patient aged 50 years, there is a large fibrous macular scar (blue arrows). Stage 4, fundus photography of an 86-year-old woman. She has a retinitis pigmentosa with peripheral bone spicules and a dry desert land pattern in the temporal part in the right eye (black arrows).

Figure 1.

Phenotypic characteristics of a unique retinal dystrophy, the MCRPE. Fundus pictures and disease progression with age, from pre-clinical stage with a normal fundus to retinitis pigmentosa. After a pre-clinical stage with a normal fundus, a dry desert land pattern appears. It first involves the posterior pole in the temporal part of the fovea in the case presented (stage 1, blue arrows) and slowly progresses with age to the peripheral retina with a bilateral involvement (stage 2). Stage 3 is defined by a macular and peripheral dry desert land pattern combined with a retinal complication. In the left eye of this patient aged 50 years, there is a large fibrous macular scar (blue arrows). Stage 4, fundus photography of an 86-year-old woman. She has a retinitis pigmentosa with peripheral bone spicules and a dry desert land pattern in the temporal part in the right eye (black arrows).

Figure 2.

Spectral domain coherent tomography in stages 1 and 2 of MCRPE. Normal segmentation from the inner to the outer layers, the retinal nerve fiber (RNFL), the retinal ganglion cell (RGC), the inner (INL) and outer nuclear (ONL) layers, the external limiting membrane (ELM), the ellipsoid zone (inside inner segment of photoreceptor cell), the RPE and Bruch's membrane complex (RPE + MB) and the choroid. In comparison to normal image, the dry desert land pattern corresponds to irregular, thickened and scalloped elevation of the RPE (blue arrows).

Figure 2.

Spectral domain coherent tomography in stages 1 and 2 of MCRPE. Normal segmentation from the inner to the outer layers, the retinal nerve fiber (RNFL), the retinal ganglion cell (RGC), the inner (INL) and outer nuclear (ONL) layers, the external limiting membrane (ELM), the ellipsoid zone (inside inner segment of photoreceptor cell), the RPE and Bruch's membrane complex (RPE + MB) and the choroid. In comparison to normal image, the dry desert land pattern corresponds to irregular, thickened and scalloped elevation of the RPE (blue arrows).

WES was applied to this large autosomal dominant pedigree of MCRPE spanning three generations. A missense mutation was identified in MAPKAPK3 (mitogen-activated protein kinase-activated protein kinase 3), a gene never associated with any disease. MAPKAPK3, as MAPKAPK2, is a kinase of the p38 kinase pathway that participates in stress responses. We subsequently investigated the structural consequences of MAPKAPK3 mutation, the expression of MAPKAPK3 and its ortholog MAPKAPK2 in RPE cells and fibroblasts and analyzed the phenotype of HEK cells overexpressing the wild-type and mutant proteins. In addition, age-related retinal alterations of Mapkapk3−/− mice were investigated.

Results

Clinical findings

Forty-two family members were included (Fig. 3). MCRPE diagnosis was established upon the observation of the characteristic dry desert land pattern on fundus examination. Fourteen affected and 28 unaffected subjects were available for genetic testing. All the 14 clinically affected persons were older than 30 at presentation. Stage 1 was noted in younger individuals (V:3, V:5 and V:8, 30–37 years old). Patient V:3 aged 30 developed a PCV in the abnormal dry desert land area of the right eye. Four affected persons with a mean age of 54.75 years and a visual acuity ranging from 20/30 to 20/20 had a stage 2. Six patients older than 50 years (mean age: 56.8 years) had a stage 3, one of them was legally blind (IV:2), and visual acuity varied from 20/1000 to 20/20 for the others. The oldest patient, aged 86 years, had a retinitis pigmentosa (stage 4, III:2) and was legally blind. Clinical features of affected individuals are summarized in Table 1. Twenty-eight patients were asymptomatic with a normal fundus; 12 of them were older than 30 years and according to the age of onset of MCRPE, they should remain unaffected.

Table 1.

Clinical data of patients with the MAPKAPK3 mutation c.518T>C (p.Leu173Pro)

Patient Age at presentation Visual acuity RE-LE Fundus examination Complications 
III:2 86 20/2000-20/400 Retinitis pigmentosa + macular atrophy 
IV:2 59 20/200-20/400 Macular atrophy RE-LE 
IV:3 58 20/20-20/20  
IV:7 54 20/25-20/30  
IV:10 51 20/20-20/400 PCV LE 
IV:11 50 20/25-20/1000 PCV LE + macular fibrosis 
IV:18 63 20/20-20/20 PCV RE 
IV:19 60 20/30-20/25  
IV:23 55 20/100-20/50 PCV LE + macular scar 
IV:25 63 20/20-20/20 PCV RE 
IV:28 57 20/20-20/20  
V:3 30 20/50-20/20 PCV RE with a stage 1 dry desert land pattern 
V:5 37 20/20-20/20  
V:8 30 20/20-20/20  
V:18 21 20/20-20/20 Pre-clinical stage with normal fundus 
Patient Age at presentation Visual acuity RE-LE Fundus examination Complications 
III:2 86 20/2000-20/400 Retinitis pigmentosa + macular atrophy 
IV:2 59 20/200-20/400 Macular atrophy RE-LE 
IV:3 58 20/20-20/20  
IV:7 54 20/25-20/30  
IV:10 51 20/20-20/400 PCV LE 
IV:11 50 20/25-20/1000 PCV LE + macular fibrosis 
IV:18 63 20/20-20/20 PCV RE 
IV:19 60 20/30-20/25  
IV:23 55 20/100-20/50 PCV LE + macular scar 
IV:25 63 20/20-20/20 PCV RE 
IV:28 57 20/20-20/20  
V:3 30 20/50-20/20 PCV RE with a stage 1 dry desert land pattern 
V:5 37 20/20-20/20  
V:8 30 20/20-20/20  
V:18 21 20/20-20/20 Pre-clinical stage with normal fundus 

RE, right eye; LE, left eye.

Stage 1: dry desert land pattern emerging in the temporal of the macula or the juxtapapillary zone; stage 2: the pattern progresses to the posterior pole and the peripheral retina; stage 3: stage 2 with complications, i.e. choroidal neovascularization (CNV), PCV or macular atrophy or fibroglial scar; stage 4: retinitis pigmentosa with the characteristic dry desert land pattern at the posterior pole. All patients are symptomatic and have characteristic fundus abnormalities, except patient V:18, aged 21 (pre-clinical stage).

Figure 3.

Pedigree of family originating from Martinique, one of the French West Indies islands. The dystrophy is inherited as an autosomal dominant trait. Forty-two individuals from three generations have been examined. Fourteen patients older than 30 have a typical dry desert land pattern. All affected individuals (black symbols) have the c.518T>C (p.Leu173Pro) mutation (M/+) in MAPKAPK3. The 12 patients older than 30 with a normal fundus (open symbols) do not have the mutation (+/+). Electropherogram showing the affected sequence surrounding the c.518T>C mutation in exon 8 of MAPKAPK3. Alignment of eukaryote protein sequences (amino acids 164–180 from MAPKAPK3) including the mutated amino acid Leu173Pro (H.s., Homo sapiensMAPKAPK3 and K2; M.m., Mus musculus ; X.l., Xenopus laevis; D.m., Drosophila melanogaster; C.e., Caenorhabditis elegans; S.c ., Saccharomyces cerevisiae ; A.t., Arabidopsis thaliana and O.s., Oriza sativa) showing conserved amino acids across species. Protein kinase subdomain VIb (HRDxKPEN) containing the aspartate (D) necessary for catalysis is indicated (boxed).

Figure 3.

Pedigree of family originating from Martinique, one of the French West Indies islands. The dystrophy is inherited as an autosomal dominant trait. Forty-two individuals from three generations have been examined. Fourteen patients older than 30 have a typical dry desert land pattern. All affected individuals (black symbols) have the c.518T>C (p.Leu173Pro) mutation (M/+) in MAPKAPK3. The 12 patients older than 30 with a normal fundus (open symbols) do not have the mutation (+/+). Electropherogram showing the affected sequence surrounding the c.518T>C mutation in exon 8 of MAPKAPK3. Alignment of eukaryote protein sequences (amino acids 164–180 from MAPKAPK3) including the mutated amino acid Leu173Pro (H.s., Homo sapiensMAPKAPK3 and K2; M.m., Mus musculus ; X.l., Xenopus laevis; D.m., Drosophila melanogaster; C.e., Caenorhabditis elegans; S.c ., Saccharomyces cerevisiae ; A.t., Arabidopsis thaliana and O.s., Oriza sativa) showing conserved amino acids across species. Protein kinase subdomain VIb (HRDxKPEN) containing the aspartate (D) necessary for catalysis is indicated (boxed).

Exome sequencing

WES analysis was applied on DNAs from four affected individuals (III:2, IV:3, IV:10 and V:8) and one 58-year-old unaffected sibling (IV:5). Considering MCRPE autosomal dominant inheritance and full penetrance, affected individuals must share the same heterozygous causal mutation, whereas the mutation must be absent in the unaffected person. Twenty-eight single nucleotide variants (SNV) and one intronic small insertion on chromosome 17 in UBE2Z (MIM: 611362) intron 3, 41 nucleotides from the splice site, fulfilled these conditions. Seventeen SNVs were excluded because of their allelic frequency in public databases higher than 1/1000 (data on request). Eight others were excluded because they did not segregate with the disease in the family (data on request). Three SNVs on chromosome 3 (3p21) remained in terms of familial segregation (c.88G>A, p.Asp30Asn in ELP6, rs200181192—c.1634A>G, p.Lys545Arg in USP4, rs61754208—c.518T>C, p.Leu173Pro in MAPKAPK3, 3:50681853, Table 2) with a significant Lod-score of 3.6 (locus two-point analysis at theta = 0). SNVs in ELP6 and USP4 were eliminated after analysis of their bioinformatic prediction (Table 2) and their presence in databases (ELP6, ExAC database: 19/120 622; USP4, ExAC database: 109/121 356, Table 2). Conversely, the c.518T>C in exon 8 of MAPKAPK3 (MIM 602 130) is not reported in any public human SNP database [1000 genomes (http://www.1000genomes.org), Exome variant server (http://evs.gs.washington.edu/EVS), dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP, date last accessed, December 5, 2015) and ExAC database Version 2.0 (exac.broadinstitute.org, date last accessed, December 5, 2015)] or in the literature. The daughter V:18 of the affected person IV:11 carried this variant and had a normal fundus, but considering her age of 21 years, she is most probably at an asymptomatic pre-clinical stage. The c.518T>C was not found in 140 control alleles from Martinique individuals older than 40 years with normal eye fundi. The four affected subjects analyzed by WES had no other pathogenic mutation in the known retinal dystrophy genes [Retinal Information Network, RetNet (https://sph.uth.edu/retnet, date last accessed, December 5, 2015)]. The copy number variant analysis did not reveal other candidate variants.

Table 2.

Data of the SNVs on chromosome 3

 ELP6
MIM: 615 020 
USP4
MIM: 603 484 
MAPKAPK3
MIM: 602 130 
Variation c.88 G>A
p.Asp30Asn
rs: 200 181 192 
c.1634 A>G
p.Lys545Arg
rs: 61 754 208 
c.518 T>C
p.Leu173Pro
rs: unknown 
Ensembl location
GRCh38 
3:47 511 193 3:49 297 927 3:50 642 346 
ExAc location 3:47 552 683 3:49 335 360 3:50 681 852 
ExAC allelic frequencies Total: 19/120 622
African: 5/9776 
Total: 109/121 356
African: 16/10 400 
Never reported 
Polyphen 2 Damaging Tolerated Damaging 
SIFT Tolerated Tolerated Damaging 
Provean Damaging Tolerated Damaging 
Mutation Taster Damaging Damaging Damaging 
Align-GVDG Tolerated Tolerated Damaging 
 ELP6
MIM: 615 020 
USP4
MIM: 603 484 
MAPKAPK3
MIM: 602 130 
Variation c.88 G>A
p.Asp30Asn
rs: 200 181 192 
c.1634 A>G
p.Lys545Arg
rs: 61 754 208 
c.518 T>C
p.Leu173Pro
rs: unknown 
Ensembl location
GRCh38 
3:47 511 193 3:49 297 927 3:50 642 346 
ExAc location 3:47 552 683 3:49 335 360 3:50 681 852 
ExAC allelic frequencies Total: 19/120 622
African: 5/9776 
Total: 109/121 356
African: 16/10 400 
Never reported 
Polyphen 2 Damaging Tolerated Damaging 
SIFT Tolerated Tolerated Damaging 
Provean Damaging Tolerated Damaging 
Mutation Taster Damaging Damaging Damaging 
Align-GVDG Tolerated Tolerated Damaging 

In silico analysis of p.Leu173Pro MAPKAPK3 mutation

MAPKAPK3 and its ortholog MAPKAPK2 are serine/threonine kinases activated by p38 kinase α and β isoforms in response to stress factors (2–7). The c.518T>C mutation causes a leucine-to-proline substitution at codon 173 (p.Leu173Pro) in the 382 amino acid-long MAPKAPK3 sequence (ENST00000446044, Q16644). This unknown variant is predicted to be probably damaging by PolyPhen 2 with a score = 1 (http://genetics.bwh.harvard.edu/pph2) and damaging according to SIFT with a score = 0 (http://sift.jcvi.org), to interfere most likely with the function of the protein by the align-GVDG program with a GD score = 97.78 (http://agvgd.iarc.fr/agvgd_input.php), disease causing for Mutation Taster with a probability of 0.99 (www.mutationtaster.org) and deleterious by Provean with a score of −6.994 (provean.jcvi.org) (last search in databases performed in September 2015). Amino-acid sequence alignment of MAPKAPK3 orthologs showed the conservation of the leucine at position 173 in all eukaryotic sequences available (Fig. 3). According to our modeling study using the server @TOME-2, the p.Leu173Pro substitution, but not the conservative mutation p.Leu173Phe, another variant present in the ExAC db at low frequency (3/121 384), affects one central β-strand (residues 172–175) of the protein kinase domain (residues 44–304) within the ATP-binding site (Fig. 4B and C) (8). This mutation is predicted to destabilize not only the interaction with adenosine moiety but also the global structure of the enzyme (8–10).

Figure 4.

Model of ADP-bound MAPKAPK3 and mutation impacts. (A) The overall structure of MAPKAPK3 is shown in cyan ribbon with the bound ADP in orange stick and the leucine Leu173 in violet stick. (B) Side view of the ADP binding with the wild-type or mutant side chain drawn as in (A). (C) Top view of the β-sheet containing a central β-strand harboring the mutated residue. The red crosses in (B) and (C) indicate bad or lost contacts. The 3D models were built using as template the crystal structure of MAPKAPK2 bound to an ADP molecule. The corresponding models can be found at http://atome1.cbs.cnrs.fr/AT2B/MAPKAPK3_MUTANTS/. The figure was drawn using Pymol.

Figure 4.

Model of ADP-bound MAPKAPK3 and mutation impacts. (A) The overall structure of MAPKAPK3 is shown in cyan ribbon with the bound ADP in orange stick and the leucine Leu173 in violet stick. (B) Side view of the ADP binding with the wild-type or mutant side chain drawn as in (A). (C) Top view of the β-sheet containing a central β-strand harboring the mutated residue. The red crosses in (B) and (C) indicate bad or lost contacts. The 3D models were built using as template the crystal structure of MAPKAPK2 bound to an ADP molecule. The corresponding models can be found at http://atome1.cbs.cnrs.fr/AT2B/MAPKAPK3_MUTANTS/. The figure was drawn using Pymol.

MAPKAPK3 expression and MAPKAPK3/MAPKAPK2 ratio in retinal and fibroblast cells

As detailed in the databases (Genecards, www.genecards.org; the human protein Atlas, www.proteinatlas.org), MAPKAPK3 is expressed ubiquitously including in the choroid and the retina. We confirmed by reverse transcription (RT)–quantitative polymerase chain reaction (q-PCR) that MAPKAPK3 mRNA is found in the RPE derived from human-induced pluripotent stem cells (iPSCs). We also found similar amounts of MAPKAPK3 mRNA in control (n = 3) and patient (n = 2) fibroblasts (Fig. 5), thus excluding mRNA decay linked to the mutation. Interestingly, relative expression of MAPKAPK3 compared with MAPKAPK2 was proportionally higher in RPE (ratio MAPKAPK3/MAPKAPK2 0.59) than in fibroblasts (ratio 0.11), suggesting that MAPKAPK3 is critical for RPE physiology (Fig. 5).

Figure 5.

Wild-type and mutated MAPKAPK3 expression, localization and effects on cells. (A) Ratio of MAPKAPK2/MAPKAPK3 mRNA abundance in fibroblasts (Fib.) and retinal pigmentary epithelium (RPE). (B) MAPKAPK3 (left) and MAPKAPK2 (right) mRNA relative abundance in control (Ctl.; n = 3) and patient (Pat.; n = 2) fibroblasts, determined by RT–q-PCR. Results are given as mean ± SEM. (C) Intracellular localization and effect on nuclear segregation of wild-type and mutated MAPKAPK3-GFP fusion protein expression in HEK cells. Assessment of the nuclear localization of the GFP fluorescence (left) and of the percentage of multi-nucleated transfected cells (right). Results are given as mean ± SEM. (**P < 0.01). (D) Fluorescent pictures of the GFP and DAPI staining in HEK cells transfected with the wild-type (left) and Leu173Pro (right) MAPKAPK3-GFP fusion proteins (bar: 10 µm). (E) Fluorescent pictures of the tubulin-associated, GFP and DAPI staining in HEK cells transfected with the wild-type (left) and Leu173Pro (right) MAPKAPK3-GFP fusion proteins. Top: merge of the three fluorescent signals; middle: merge of the tubulin and DAPI staining; bottom: enlargement of the former pictures showing the nuclear localization of the wild-type GFP fusion protein and normal cytoskeleton (left) and the cytoplasmic localization of the mutated Leu173Pro GFP fusion protein with disorganized cytoskeleton (right) (bar: 10 µm).

Figure 5.

Wild-type and mutated MAPKAPK3 expression, localization and effects on cells. (A) Ratio of MAPKAPK2/MAPKAPK3 mRNA abundance in fibroblasts (Fib.) and retinal pigmentary epithelium (RPE). (B) MAPKAPK3 (left) and MAPKAPK2 (right) mRNA relative abundance in control (Ctl.; n = 3) and patient (Pat.; n = 2) fibroblasts, determined by RT–q-PCR. Results are given as mean ± SEM. (C) Intracellular localization and effect on nuclear segregation of wild-type and mutated MAPKAPK3-GFP fusion protein expression in HEK cells. Assessment of the nuclear localization of the GFP fluorescence (left) and of the percentage of multi-nucleated transfected cells (right). Results are given as mean ± SEM. (**P < 0.01). (D) Fluorescent pictures of the GFP and DAPI staining in HEK cells transfected with the wild-type (left) and Leu173Pro (right) MAPKAPK3-GFP fusion proteins (bar: 10 µm). (E) Fluorescent pictures of the tubulin-associated, GFP and DAPI staining in HEK cells transfected with the wild-type (left) and Leu173Pro (right) MAPKAPK3-GFP fusion proteins. Top: merge of the three fluorescent signals; middle: merge of the tubulin and DAPI staining; bottom: enlargement of the former pictures showing the nuclear localization of the wild-type GFP fusion protein and normal cytoskeleton (left) and the cytoplasmic localization of the mutated Leu173Pro GFP fusion protein with disorganized cytoskeleton (right) (bar: 10 µm).

Effects of mutated MAPKAPK3 in HEK cells

To further characterize the effect of the p.Leu173Pro substitution on MAPKAPK3 functions, wild-type and c.518T>C MAPKAPK3 cDNAs were cloned in fusion to the green fluorescent protein (GFP) cDNA and expressed in HEK cells. We found that the wild-type protein was mostly localized in the nucleus (87 ± 2.6%), whereas the mutated protein was predominantly localized in the cytoplasm (84.5 ± 8.1%, Fig. 5C and D). This suggests that the pathogenic effect of the p.Leu173Pro substitution involves subcellular trafficking, possibly by de-masking the nuclear export signal of MAPKAPK3—a mechanism also reported for the activation of MAPKAPK2 (11). In addition, we observed that in cells transfected with the mutated allele, 32.5 ± 3.5% of the cells were multi-nucleated, whereas only 8.0 ± 1.4% of the cells showed this phenotype when transfected with the wild-type allele (Fig. 5C and D), thus indicating that the expression of the mutated allele leads to an inhibition of cytokinesis. This prompted us to analyze the structure of the tubulin-related cytoskeleton. We observed that the expression of the p.Leu173Pro allele induced a severe depolarization of the tubulin structures, as no co-localization of microtubules with the mutated MAPKAPK3 protein could be detected in transfected cells, whereas the cytoskeleton was thoroughly normal in cells expressing the wild-type allele (Fig. 5E). These results suggested that mislocalization of the mutated protein could lead to a nuclear MAPKAPK3 haploinsufficiency and that the p.Leu173Pro mutated protein could alter the cytoskeleton by a deleterious cytoplasmic gain of function.

Histological analysis of Mapkapk3−/− mice

To gain further insights into the pathophysiological mechanisms, we performed electron microscopy studies on eyes from Mapkapk3−/− (n = 11) and wild-type littermate mice (n = 7) aged 2–6 months. This mouse strain was generated by targeted deletion of exons 3 and 4 of Mapkapk3 in 129 Sv embryonic stem cells as a model of toxic glomerulonephritis (12). We examined only homozygous mice, because in most heterozygous models of dominant retinal diseases, no phenotype is observed. On the basis of the human retinal phenotype, we focussed our observations on the Bruch's membrane and RPE anomalies. Bruch's membrane is an elastin- and collagen-rich extracellular matrix located between RPE and its source of nutrition, the choriocapillaris. In wild-type mice, this membrane, consisting of the RPE basement membrane, intermediate connective tissue collagen fibers and the choriocapillaris basement membrane, had a regular thickness of 451 ± 73.88 nm at 2 months and of 514 ± 98.29 nm at 6 months. In contrast, the Mapkapk3−/− mice developed Bruch's membrane abnormal thickening and thinning progressing with age, 516 ± 204.9 nm at 2 months and 857 ± 663 nm at 6 months (Fig. 6). At 6 months, mice had no additional retinal lesion and no photoreceptor loss, as more than 10–12 lines of nuclei were found all over the photoreceptor layer in both wild-type and mutated mice. Older Mapkapk3−/− or Mapkapk3+/− mice were not available for eye analysis. These data suggest that the structure or composition of Mapkapk3−/− Bruch's membrane is modified, therefore altering exchanges between the RPE and the choroid, affecting their function and that of photoreceptors, and possibly influencing the occurrence of PCV as choroidal neovascularization. All together, these data suggest that nuclear MAPKAPK3 haploinsufficiency is the possible mechanism underlying the c.518T>C mutation.

Figure 6.

Bruch's membrane electron microscopy of wild-type and Mapkapk3−/− mice. Bruch's membrane is a pentalaminated extracellular matrix interposed between RPE with characteristic basal infoldings and choroidal endothelial cells (choriocapillaris) (bar: 1 µm). (A) Bruch's membrane has a regular and normal thickness below 600 nm in 2-month-old wild-type mouse (mean value: 451 ± 73.88 nm, black arrows). (B) Two-month-old Mapkapk3−/− mouse presents a mean Bruch's membrane (black arrows) thickness of 516 ± 204.9 nm. (C) Bruch's membrane thickness (black arrows) is regular in the 6-month-old wild-type mouse (mean 514 ± 98.29 nm). (D) Six-month-old Mapkapk3−/− mouse has a Bruch's membrane thickness of 857 ± 663 nm. The Bruch's membrane architecture is severely disorganized in several parts with abnormal thickening (large black arrow) and thinning (narrow black arrow) portions. These morphological changes are in line with the coherent tomography data of affected subjects in whom RPE and Bruch's membrane are thickened and irregular on SD-OCT.

Figure 6.

Bruch's membrane electron microscopy of wild-type and Mapkapk3−/− mice. Bruch's membrane is a pentalaminated extracellular matrix interposed between RPE with characteristic basal infoldings and choroidal endothelial cells (choriocapillaris) (bar: 1 µm). (A) Bruch's membrane has a regular and normal thickness below 600 nm in 2-month-old wild-type mouse (mean value: 451 ± 73.88 nm, black arrows). (B) Two-month-old Mapkapk3−/− mouse presents a mean Bruch's membrane (black arrows) thickness of 516 ± 204.9 nm. (C) Bruch's membrane thickness (black arrows) is regular in the 6-month-old wild-type mouse (mean 514 ± 98.29 nm). (D) Six-month-old Mapkapk3−/− mouse has a Bruch's membrane thickness of 857 ± 663 nm. The Bruch's membrane architecture is severely disorganized in several parts with abnormal thickening (large black arrow) and thinning (narrow black arrow) portions. These morphological changes are in line with the coherent tomography data of affected subjects in whom RPE and Bruch's membrane are thickened and irregular on SD-OCT.

Discussion

MCRPE exhibits a phenotype featuring a fundus appearance resembling a dry desert land, Bruch's membrane irregular thickening on SD-OCT, frequent occurrence of PCV or choroidal neovascularization and late RPE and photoreceptor cell death, distinct from all other known retinal dystrophies.

WES analysis identified in this large family from the West Indies islands a heterozygous mutation in MAPKAPK3, a gene never associated with any retinal dystrophy or other clinical presentation. MAPKAPK3 is located on chromosome 3 (3p21.2, MIM 602 130) and harbors 11 exons. It encodes a 382 amino acid protein that belongs to serine/threonine kinase family. The protein kinase domain is located from amino acid 47 to 304 and the serine/threonine-protein kinase active site from 162 to 174 amino acids. MAPKAPK3 and its ortholog MAPKAPK2 are phylogenetically closely related serine/threonine kinases with a protein sequence identity of 75%. They are activated by p38 kinase α and β isoforms in response to stress factors such as DNA damage, oxidative and osmotic stresses (13–17). Activation of this p38 pathway leads to cytokine production, cytoskeleton remodeling, cell migration and transcriptional regulations (13–17). MAPK signaling networks exhibit complexity and involve multiple feedback mechanisms. For example, inactive (dephosphorylated) MAPKAPK2 and MAPKAPK3 form a stable complex with inactive p38 MAPK in the nucleus. Upon activation, MKK3/6 displaces MAPKAPK2 from the complex, resulting in the phosphorylation of p38 MAPK and subsequently activation of MAPKAPK2/3 and their exportations into the cytoplasm. In most mammalian tissues, the expression level and activity of MAPKAPK2 are much higher than those of MAPKAPK3 (18). Nevertheless, we found that in RPE cells, MAPKAPK3 is highly expressed in comparison to other cell types such as fibroblasts. This observation is relevant to the specificity of MCRPE clinical presentation, which is restricted to the retina in MAPKAPK3 patients, and suggests that the MAPKAPK2 activity in RPE cells cannot compensate the decrease of MAPKAPK3 abundance in the nucleus or that MAPKAPK3 has a specific function in RPE (19–21). The altered MAPKAPK3 function can therefore impinge on RPE cytoskeleton integrity, gene expression and cell proliferation and can lead to a progressive pathophysiological course affecting the interactions between the RPE and Bruch's membrane. Indeed, RPE alterations might affect Bruch's membrane modeling and renewal, leading to its abnormal structure, and later dysfunction. Bruch's membrane turnover abnormalities are, for example, implicated in age-related macular degeneration, choroidal neovascularization and in autosomal dominant Sorsby retinal dystrophy due to mutations in TIMP3 (22–26). The Bruch's membrane turnover is regulated by a dynamic balance between the matrix metalloproteinases (MMP-1, -2, -3, -9) and the tissue inhibitors of metalloproteinases (TIMPs), two families of enzymes that are secreted by the RPE and choroidal endothelial cells (22–24). Thus, dysregulation of the p38 MAPK pathway by non-functional MAPKAPK3 could alter the balance of MMP and TIMP, leading to a reduction of Bruch's membrane degradation and to its thickening and thinning, as noted in homozygous Mapkapk3−/− mice (Fig. 4) and in SD-OCT data of the family reported here (Fig. 2).

Consequently, progressive Bruch's membrane alterations and the possible imbalance between MMPs and TIMPs may promote neovascularization and choroidal polypoidal vasculopathy, as evidenced in six out of the 15 affected individuals. Ultimately, impairment of RPE cell cycle might affect the lifelong RPE capacity, whereas reduced MAPKAPK3 activity could impinge vascular endothelial growth factor (VEGF) expression and secretion and compromise RPE cell survival (27–30). Together, these mechanisms might lead to RPE cell death, with subsequent photoreceptor loss causing a retinitis pigmentosa as observed in the eldest patient (III:2, 86 years old).

Altogether, these data support the evidence that MAPKAPK3 is responsible for this specific inherited retinal pigment dystrophy. It is the first reported MAPKAPK3-related human disease, and it is also the first implication of the p38 signaling pathway by one of its actors in inherited retinal dystrophies. It will be important to further study the functions of this gene and its pathophysiological mechanisms and screen large cohorts with retinal dystrophy, particularly those presenting primary RPE and Bruch's membrane alterations such as age-related macular disease, and those associated with choroidal neovascularization.

Materials and Methods

Clinical examination

Informed consent was obtained for clinical examination and genetic analysis from all patients, according to approved protocols of the Fort de France and Montpellier University Hospitals, and in agreement with the Declaration of Helsinki. The Ministry of Public Health accorded approval for biomedical research under the authorization number 11018S. For each patient, age at examination, refraction and initial and final visual acuity were noted. The best corrected visual acuity was obtained with Snellen charts. Reading visual acuity was assessed with the current French near vision chart (Parinaud). Color fundus frames were performed with Topcon Imagenet (Ophthalmic Imaging Systems, Japan). Autofluorescence imaging and SD-OCT were performed with Combined Heidelberg Retina Angiograph + OCT Spectralis device (Heidelberg Engineering, Dossenheim, Germany) and Nidek non-mydriatic autofundus camera AFC 330 (Nidek Inc., Japan). The full-field ERG was recorded according to the standards of the International Society of Clinical Electrophysiology of Vision (31).

Molecular genetic and bioinformatic analyses

Genomic DNA was isolated from 10 ml peripheral blood leucocytes using standard salting out procedure (32). WES was performed by the company Integragen (Evry, France). All coding exons of genomic DNA samples were captured using Sure Select Human All Exon Kits Version 3 in-solution enrichment methodology (Agilent, Santa Clara, CA, USA) with the biotinylated oligonucleotides probe library (Human All Exon v3 50 Mb, Agilent), followed by paired-end 75 bp massively parallel sequencing (Illumina HISEQ2000, Illumina, San Diego, CA, USA). Image analysis and base calling were performed using Illumina Real Time Analysis Pipeline version 1.14 with default parameters (Illumina). Bioinformatic analysis of sequencing data was based on the CASAVA 1.8 pipeline [Consensus Assessment of Sequence and Variation (CASAVA) 1.8, Illumina]. For filtering, exonic and splice variants were selected on the basis of their heterozygosity in the four affected individuals and absence in the unaffected sibling, in dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP), the 1000 genome project (http://www.1000genomes.org), the Exome Variant Server (http://evs.gs.washington.edu/EVS) and ExAC v2.0 database (exac.broadinstitute.org). Sequence conservation, PolyPhen 2 (http://genetics.bwh.harvard.edu/pph2), SIFT (http://sift.jcvi.org), the align-GVDG (http://agvgd.iarc.fr/agvgd_input.php), disease causing for Mutation Taster (www.mutationtaster.org) and Provean (provean.jcvi.org) programs were used to predict the pathogenic nature of sequence alterations.

Expression of MAPKAPK3and MAPKAPK2 in fibroblasts and in RPE derived from human iPSCs

Isolation and amplification of skin fibroblasts

After informed consent, a skin biopsy on the inner side of the upper arm of two affected patients and three unaffected with a confirmed molecular diagnosis was performed under sterile conditions. The skin biopsy specimens were rinsed in phosphate-buffered saline (PBS), cut into small pieces and cultured in 35 mm culture dishes (two pieces per dish) in AmnioMax-C100 basal media with l-glutamine (Invitrogen, Life Technologies, St Aubin, France) containing 10% de-complemented fetal calf serum (Lonza, Verviers, Belgium), 1% penicillin–streptomycin–amphotericin B (Lonza) and 2% AmnioMax-C100 supplement (Invitrogen, Life Technologies) at 37°C under 5% CO2. The biopsies were removed to a fresh dish once the emerging fibroblasts reached 80% of confluence. The cells were passaged, and aliquots ranging from P1 to P5 from each culture were frozen in fetal calf serum containing 10% dimethyl sulfoxide (Sigma Aldrich, St Quentin Fallavier, France).

RPE generation

To differentiate iPSCs into RPE, we used a previously described spontaneous differentiation protocol (33). Briefly, iPSC colonies were allowed to grow to confluence on feeder cells, and the basic fibroblast growth factors (bFGFs) were then removed from the ES media. The media continued to be changed daily during the differentiation process. Pigmented foci appeared over the course of the month following bFGF depletion, which were manually dissected. The foci from one plate were pooled, dissociated with 0.25% trypsin, seeded onto 24- or six-well culture dishes coated with Matrigel (diluted 1:30) and cultured in bFGF-depleted ES media. Once a confluent monolayer was reached, the cells showed a pigmented polygonal morphology and could be maintained in long-term culture with media changes every 3–5 days. Cells were passaged by trypsin dissociation and amplified as required. All analyses were performed on RPE at P3. Fluid-filled domes were observed using a SteREO Discovery V2.0 microscope (Carl Zeiss S.A.S., Le Pecq, France).

RT and q-PCR studies

RT–q-PCR was used to analyze the expression of MAPKAPK2 and MAPKAPK3 in human-RPE cells as well as in fibroblasts. Following RNA isolation and cDNA synthesis, q-PCR amplification was performed using gene-specific primers (Refseq NM_001243926, sequences upon request) and the Light Cycler 480 SYBR Green I Master mix on a Light Cycler 480 II thermal cycler (Roche, France). Gene expression was normalized to glyceraldehyde 3-phosphate dehydrogenase expression. Results were analyzed using Light Cycler 480 software and Microsoft Excel. The expression of RPE-specific markers was analyzed using classic RT–PCR amplification with gene-specific primers (sequences upon request), and the amplification products were analyzed on 2% agarose gel in 0.5× TAE buffer.

Cell culture and plasmid transfection in HEK cells

Wild-type and mutated MAPKAPK3 cDNAs were synthetized by Genecust and cloned in pEGFP-N1 in fusion with the GFP open reading frame. HEK cells were cultured in Dulbecco's modified Eagle's medium (DMEM), 10% fetal bovine serum (FBS) and 5% CO2. Transfections of plasmids (0.8 µg of plasmids/well and 2 µl Lipofectamine/well) were performed with Lipofectamine 2000 reagent (Invitrogen). For GFP direct fluorescent microscopy, cells were grown on glass coverslips and then fixed in PBS, 4% paraformaldehyde + DAPI (20 min, 4°C), washed three times in PBS and observed. For immunostaining, cells were washed with DMEM and fixed in 4% paraformaldehyde in PBS for 20 min at 4°C, then washed three times with PBS and permeabilized in 0.1% Triton X-100 in PBS containing 5% FBS for 30 min at room temperature (PBSF), followed by washing three times with PBS. Primary monoclonal anti-alpha-tubulin mouse antibodies (T9026, Sigma) were diluted 1/500 in PBSF and incubated with cells for 2 h at room temperature. After washing three times with PBSF, the cells were incubated for 1 h at room temperature with the fluorescent secondary donkey anti-mouse-Alexa-Fluor 594 antibody (Molecular Probes) and diluted 1/1000 in PBSF. Cells were rinsed twice with PBS for 10 min, and coverslips were mounted using fluorescence Mounting Medium (Dako, S3023). The images were acquired using an Inverse1 Zeiss Axio observer/LSM 5 LIVE DUO Confocal microscope (Carl Zeiss Microscopy), equipped with a 63× oil objective.

Histological analysis of the retina of Mapkapk3−/− mice

Eyes of Mapkapk3−/− (n = 11) and wild-type (n = 7) mice (background C57BL/6) were immersed in a solution of 2.5% glutaraldehyde in PHEM buffer (1×, pH 7.4) overnight at 4°C. They were then rinsed in PHEM buffer and post-fixed in a 0.5% osmic acid for 2 h at dark and room temperature. After two rinses in PHEM buffer, the cells were dehydrated in a graded series of ethanol solutions (30–100%). The cells were embedded in EmBed 812 using an Automated Microwave Tissue Processor for Electronic Microscopy, Leica EM AMW. Thin sections (70 nm, Leica-Reichert Ultracut E) were collected at different levels of each block. These sections were counterstained with uranyl acetate and lead citrate and observed using a Hitachi 7100 transmission electron microscope in the Centre de Ressources en Imagerie Cellulaire de Montpellier (France) and using a Tecnai G20 transmission electron microscope at 200 kV in the CoMET MRI facilities, INM France.

Funding

This work was also supported by INSERM, CNRS and ANR-10-BINF-03-03 (to G.L.).

Acknowledgements

We thank all the family members who participated in this study. We acknowledge the support from the INSERM, the University Hospital of Montpellier and the Imaging and Laser Center of Paris (CIL). We are indebted to a native English speaker, Patrick Carroll, from the Institute for Neurosciences of Montpellier for helpful English lectures and corrections.

Conflict of Interest statement. The authors declare no competing financial interests.

References

1
Jean-Charles
A.
,
Cohen
S.Y.
,
Merle
H.
,
Quentel
G.
,
Legargasson
J.-F.
,
Gaudric
A.
(
2013
)
Martinique (West Indies) crinkled retinal pigment epitheliopathy: clinical description
.
Retina
 ,
33
,
1041
1048
.
2
Sithanandam
G.
,
Latif
F.
,
Duh
F.M.
,
Bernal
R.
,
Smola
U.
,
Li
H.
,
Kuzmin
I.
,
Wixler
V.
,
Geil
L.
,
Shrestha
S.
(
1996
)
3pK, a new mitogen-activated protein kinase-activated protein kinase located in the small cell lung cancer tumor suppressor gene region
.
Mol. Cell. Biol.
 ,
16
,
868
876
.
3
Ludwig
S.
,
Engel
K.
,
Hoffmeyer
A.
,
Sithanandam
G.
,
Neufeld
B.
,
Palm
D.
,
Gaestel
M.
,
Rapp
U.R.
(
1996
)
3pK, a novel mitogen-activated protein (MAP) kinase-activated protein kinase, is targeted by three MAP kinase pathways
.
Mol. Cell. Biol.
 ,
16
,
6687
6697
.
4
Cargnello
M.
,
Roux
P.P.
(
2011
)
Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases
.
Microbiol. Mol. Biol. Rev.
 ,
75
,
50
83
.
5
Garrington
T.P.
,
Johnson
G.L.
(
1999
)
Organization and regulation of mitogen-activated protein kinase signaling pathways
.
Curr. Opin. Cell Biol.
 ,
11
,
211
218
.
6
Kostenko
S.
,
Dumitriu
G.
,
Lægreid
K.J.
,
Moens
U.
(
2011
)
Physiological roles of mitogen-activated-protein-kinase-activated p38-regulated/activated protein kinase
.
World J. Biol. Chem.
 ,
2
,
73
89
.
7
Shiryaev
A.
,
Moens
U.
(
2010
)
Mitogen-activated protein kinase p38 and MK2, MK3 and MK5: ménage à trois or ménage à quatre
?
Cell Signal.
 ,
22
,
1185
1192
.
8
Pons
J.L.
,
Labesse
G.
(
2009
)
@TOME-2: a new pipeline for comparative modeling of protein–ligand complexes
.
Nucleic Acids Res.
 ,
37
,
W485
W491
.
9
Cheng
R.
,
Felicetti
B.
,
Palan
S.
,
Toogood-Johnson
I.
,
Scheich
C.
,
Barker
J.
,
Whittaker
M.
,
Hesterkamp
T.
(
2010
)
High-resolution crystal structure of human Mapkap kinase 3 in complex with a high affinity ligand
.
Protein Sci. Publ. Protein Soc.
 ,
19
,
168
173
.
10
Underwood
K.W.
,
Parris
K.D.
,
Federico
E.
,
Mosyak
L.
,
Czerwinski
R.M.
,
Shane
T.
,
Taylor
M.
,
Svenson
K.
,
Liu
Y.
,
Hsiao
C.L.
et al
. (
2003
)
Catalytically active MAP KAP kinase 2 structures in complex with staurosporine and ADP reveal differences with the autoinhibited enzyme
.
Structure
 ,
11
,
627
636
.
11
Engel
K.
,
Kotlyarov
A.
,
Gaestel
M.
(
1998
)
Leptomycin B-sensitive nuclear export of MAPKAP kinase 2 is regulated by phosphorylation
.
EMBO J.
 ,
17
,
3363
3371
.
12
Zaru
R.
,
Ronkina
N.
,
Gaestel
M.
,
Arthur
J.S.C.
,
Watts
C.
(
2007
)
The MAPK-activated kinase Rsk controls an acute Toll-like receptor signaling response in dendritic cells and is activated through two distinct pathways
.
Nat. Immunol.
 ,
8
,
1227
1235
.
13
Mayer
I.A.
,
Verma
A.
,
Grumbach
I.M.
,
Uddin
S.
,
Lekmine
F.
,
Ravandi
F.
,
Majchrzak
B.
,
Fujita
S.
,
Fish
E.N.
,
Platanias
L.C.
(
2001
)
The p38 MAPK pathway mediates the growth inhibitory effects of interferon-alpha in BCR-ABL-expressing cells
.
J. Biol. Chem.
 ,
276
,
28570
28577
.
14
Neufeld
B.
,
Grosse-Wilde
A.
,
Hoffmeyer
A.
,
Jordan
B.W.
,
Chen
P.
,
Dinev
D.
,
Ludwig
S.
,
Rapp
U.R.
(
2000
)
Serine/threonine kinases 3pK and MAPK-activated protein kinase 2 interact with the basic helix–loop–helix transcription factor E47 and repress its transcriptional activity
.
J. Biol. Chem.
 ,
275
,
20239
20242
.
15
Li
Y.
,
Sassano
A.
,
Majchrzak
B.
,
Deb
D.K.
,
Levy
D.E.
,
Gaestel
M.
,
Nebreda
A.R.
,
Fish
E.N.
,
Platanias
L.C.
(
2004
)
Role of p38alpha Map kinase in Type I interferon signaling
.
J. Biol. Chem.
 ,
279
,
970
979
.
16
Ronkina
N.
,
Menon
M.B.
,
Schwermann
J.
,
Arthur
J.S.C.
,
Legault
H.
,
Telliez
J.-B.
,
Kayyali
U.S.
,
Nebreda
A.R.
,
Kotlyarov
A.
,
Gaestel
M.
(
2011
)
Stress induced gene expression: a direct role for MAPKAP kinases in transcriptional activation of immediate early genes
.
Nucleic Acids Res.
 ,
39
,
2503
2518
.
17
Gaestel
M.
(
2006
)
MAPKAP kinases—MKs—two's company, three's a crowd
.
Nat. Rev. Mol. Cell Biol.
 ,
7
,
120
130
.
18
Ronkina
N.
,
Kotlyarov
A.
,
Dittrich-Breiholz
O.
,
Kracht
M.
,
Hitti
E.
,
Milarski
K.
,
Askew
R.
,
Marusic
S.
,
Lin
L.-L.
,
Gaestel
M.
et al
. (
2007
)
The mitogen-activated protein kinase (MAPK)-activated protein kinases MK2 and MK3 cooperate in stimulation of tumor necrosis factor biosynthesis and stabilization of p38 MAPK
.
Mol. Cell. Biol.
 ,
27
,
170
181
.
19
Ehlting
C.
,
Ronkina
N.
,
Böhmer
O.
,
Albrecht
U.
,
Bode
K.A.
,
Lang
K.S.
,
Kotlyarov
A.
,
Radzioch
D.
,
Gaestel
M.
,
Häussinger
D.
et al
. (
2011
)
Distinct functions of the mitogen-activated protein kinase-activated protein (MAPKAP) kinases MK2 and MK3: MK2 mediates lipopolysaccharide-induced signal transducers and activators of transcription 3 (STAT3) activation by preventing negative regulatory effects of MK3
.
J. Biol. Chem.
 ,
286
,
24113
24124
.
20
Guess
A.J.
,
Ayoob
R.
,
Chanley
M.
,
Manley
J.
,
Cajaiba
M.M.
,
Agrawal
S.
,
Pengal
R.
,
Pyle
A.L.
,
Becknell
B.
,
Kopp
J.B.
et al
. (
2013
)
Crucial roles of the protein kinases MK2 and MK3 in a mouse model of glomerulonephritis
.
PLoS ONE
 ,
8
,
e54239
.
21
Scharf
M.
,
Neef
S.
,
Freund
R.
,
Geers-Knörr
C.
,
Franz-Wachtel
M.
,
Brandis
A.
,
Krone
D.
,
Schneider
H.
,
Groos
S.
,
Menon
M.B.
et al
. (
2013
)
Mitogen-activated protein kinase-activated protein kinases 2 and 3 regulate SERCA2a expression and fiber type composition to modulate skeletal muscle and cardiomyocyte function
.
Mol. Cell. Biol.
 ,
33
,
2586
2602
.
22
Pons
M.
,
Cousins
S.W.
,
Alcazar
O.
,
Striker
G.E.
,
Marin-Castaño
M.E.
(
2011
)
Angiotensin II-induced MMP-2 activity and MMP-14 and basigin protein expression are mediated via the angiotensin II receptor type 1-mitogen-activated protein kinase 1 pathway in retinal pigment epithelium: implications for age-related macular degeneration
.
Am. J. Pathol.
 ,
178
,
2665
2681
.
23
Alge-Priglinger
C.S.
,
Kreutzer
T.
,
Obholzer
K.
,
Wolf
A.
,
Mempel
M.
,
Kernt
M.
,
Kampik
A.
,
Priglinger
S.G.
(
2009
)
Oxidative stress-mediated induction of MMP-1 and MMP-3 in human RPE cells
.
Invest. Ophthalmol. Vis. Sci.
 ,
50
,
5495
5503
.
24
Hou
X.
,
Han
Q.-H.
,
Hu
D.
,
Tian
L.
,
Guo
C.-M.
,
Du
H.-J.
,
Zhang
P.
,
Wang
Y.-S.
,
Hui
Y.-N.
(
2009
)
Mechanical force enhances MMP-2 activation via p38 signaling pathway in human retinal pigment epithelial cells
.
Graefes Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Für Klin. Exp. Ophthalmol.
 ,
247
,
1477
1486
.
25
Felbor
U.
,
Benkwitz
C.
,
Klein
M.L.
,
Greenberg
J.
,
Gregory
C.Y.
,
Weber
B.H.
(
1997
)
Sorsby fundus dystrophy: reevaluation of variable expressivity in patients carrying a TIMP3 founder mutation
.
Arch. Ophthalmol.
 ,
115
,
1569
1571
.
26
Nagai
N.
,
Klimava
A.
,
Lee
W.-H.
,
Izumi-Nagai
K.
,
Handa
J.T.
(
2009
)
CTGF is increased in basal deposits and regulates matrix production through the ERK (p42/p44mapk) MAPK and the p38 MAPK signaling pathways
.
Invest. Ophthalmol. Vis. Sci.
 ,
50
,
1903
1910
.
27
Hecquet
C.
,
Lefevre
G.
,
Valtink
M.
,
Engelmann
K.
,
Mascarelli
F.
(
2002
)
Activation and role of MAP kinase-dependent pathways in retinal pigment epithelial cells: ERK and RPE cell proliferation
.
Invest. Ophthalmol. Vis. Sci.
 ,
43
,
3091
3098
.
28
Klettner
A.
,
Westhues
D.
,
Lassen
J.
,
Bartsch
S.
,
Roider
J.
(
2013
)
Regulation of constitutive vascular endothelial growth factor secretion in retinal pigment epithelium/choroid organ cultures: p38, nuclear factor κB, and the vascular endothelial growth factor receptor-2/phosphatidylinositol 3 kinase pathway
.
Mol. Vis.
 ,
19
,
281
291
.
29
Klettner
A.
,
Roider
J.
(
2009
)
Constitutive and oxidative-stress-induced expression of VEGF in the RPE are differently regulated by different mitogen-activated protein kinases
.
Graefes Arch. Clin. Exp. Ophthalmol.
 ,
247
,
1487
1492
.
30
Hecquet
C.
,
Lefevre
G.
,
Valtink
M.
,
Engelmann
K.
,
Mascarelli
F.
(
2003
)
Activation and role of MAP kinase-dependent pathways in retinal pigment epithelium cells: JNK1, P38 kinase, and cell death
.
Invest. Ophthalmol. Vis. Sci.
 ,
44
,
1320
1329
.
31
McCulloch
D.L.
,
Marmor
M.F.
,
Brigell
M.G.
,
Hamilton
R.
,
Holder
G.E.
,
Tzekov
R.
,
Bach
M.
(
2015
)
ISCEV Standard for full-field clinical electroretinography (2015 update)
.
Doc. Ophthalmol. Adv. Ophthalmol.
 ,
130
,
1
12
.
32
Miller
S.A.
,
Dykes
D.D.
,
Polesky
H.F.
(
1988
)
A simple salting out procedure for extracting DNA from human nucleated cells
.
Nucleic Acids Res.
 ,
16
,
1215
.
33
Liao
J.-L.
,
Yu
J.
,
Huang
K.
,
Hu
J.
,
Diemer
T.
,
Ma
Z.
,
Dvash
T.
,
Yang
X.-J.
,
Travis
G.H.
,
Williams
D.S.
et al
. (
2010
)
Molecular signature of primary retinal pigment epithelium and stem-cell-derived RPE cells
.
Hum. Mol. Genet.
 ,
19
,
4229
4238
.

Author notes

These authors contributed equally to this work.