Multimodal Study of PRPH2 Gene-Related Retinal Phenotypes

PRPH2 gene mutations are frequently found in inherited retinal dystrophies (IRD) and are associated with a wide spectrum of clinical phenotypes. We studied 28 subjects affected by IRD carrying pathogenic PRPH2 mutations, belonging to 11 unrelated families. Functional tests (best-corrected visual acuity measurement, chromatic test, visual field, full-field, 30 Hz flicker, and multifocal electroretinogram), morphological retino-choroidal imaging (optical coherence tomography, optical coherence tomography angiography, and fundus autofluorescence), and clinical data were collected and analyzed. Common primary complaints, with onset in their 40s, were visual acuity reduction and abnormal dark adaptation. Visual acuity ranged from light perception to 20/20 Snellen. Visual field peripheral constriction and central scotoma were found. Chromatic sense was reduced in one third of patients. Electrophysiological tests were abnormal in most of the patients. Choroidal neovascular lesions were detected in five patients. Three novel PRPH2 variants were found in four different families. Based on the present multimodal study, we identified seven distinct PRPH2 phenotypes in 11 unrelated families carrying either different mutations or the same mutation, both within the same family or among them. Fundus autofluorescence modality turned out to be the most adequate imaging method for early recognition of this dystrophy, and the optical coherence tomography angiography was highly informative to promptly detect choroidal neovascularization, even in the presence of the extensive chorioretinal atrophy phenotype.


Introduction
Mutations in the peripherin-2 (PRPH2) gene are frequently found in inherited retinal diseases (IRD) [1,2]. This gene is located on chromosome 6p21.2 and is also known as retinal degeneration slow (RDS) gene.
The gene product, the PRPH2 protein, is a member of the tetraspanin family, a transmembrane structural glycoprotein with an integral role in the formation and structure of both rod and cone photoreceptor outer segments [3,4]. The protein, containing four transmembrane domains and an intracellular domain, forms intramolecular disulfide bonds [5][6][7] and mediates assembly of peripherin-2/retinal outer segment membrane protein 1 (PRHP2/ROM1) tetramers into covalently linked higher-order complexes [7]. The formation of this protein complex is quite important for the functional activity of the protein, that is to create and maintain the rim region of rod discs and cone lamellae and to regulate disc size and alignment [7].
Although the mechanism of action of PRPH2 gene alterations is still not completely understood [8], different related clinical pathological presentations have been described:

Materials and Methods
All research procedures described in this work adhered to the tenets of the Declaration of Helsinki. The study protocol (NEU_01-2014) was approved by the local ethical committee (Comitato Etico Centrale IRCCS Lazio, Sezione IFO/Fondazione Bietti, Rome, Italy) and informed consent after full explanation of the procedures was obtained from each subject included in the study.
We retrieved from our IRD registry, all patients with a pathogenic mutation of the PRPH2 gene and collected clinical and instrumental examinations performed during their visits. Data presented in the present study refer to the last visit.

Genetic Testing
Genetic testing was performed at MAGI's laboratory (MAGI'S Lab, Rovereto, Italy, and MAGI Euregio, Bolzano-Bozen, Italy) from salivary samples, after genetic counseling to reconstruct the family pedigree and after obtaining informed consent and explaining the characteristics of a genetic test. We assumed the mode of inheritance as autosomal dominant if two generations or more were affected; autosomal recessive if there was parental consanguinity or siblings from normal parents were affected; patients not reporting parental consanguinity and not having any evidence of other affected family members were defined as "sporadic".
The patients were tested between 2014 and 2021 via targeted NGS performed on a MiSeq personal sequencer (Illumina, San Diego, CA, USA), using panels that include genes associated with RP, MD, and pattern dystrophy. Family members of the proband were analyzed only for the variants already known and for genes modifier, as associated with the phenotype in the first family member tested with PCR. The pathogenicity of variants was evaluated according to the American College of Medical Genetics and Genomics (ACMG) guidelines [28].

Clinical Findings
Demographic and functional data of the study cohort are reported on Table 1. From a total of 63 patients belonging to 11 unrelated families, we found 34 patients carrying a PRPH2 mutation. Among them, two subjects, deceased after the genetic test, (done for completing the family segregation study) and four subjects were unavailable to come to our center to be studied. Thus, we had the opportunity to collect clinical and genetic data from 28 affected subjects. All these latter patients had some visual complaints or some clinical findings typical of bilateral and symmetric IRD.
All the family's pedigrees are available in Figure S1 in Supplementary Material. The autosomal dominant inheritance pattern was verified in 6 out of 11 families, the remnants were defined as sporadic.
Our cohort included 14 females and 14 males. The age of the patients ranged from 37 to 79 years with a mean (±SD) of 58 ± 1235 and the mean age of onset of symptoms was 41 ± 1283 years old, similar to data already reported in previous studies [7,9,20]. BCVA of the patients ranged from light perception to 20/20 Snellen. Common primary complaints were reduction in VA (10 patients, 35%), difficulty in dark adaptation (7 patients, 25%), metamorphopsia and photophobia (both 5 patients, 17%) in accordance with other previous reports [7,9].
Of note, six subjects (21%) had no symptoms, and they were detected only because of sibilants of other patients.
The most frequent visual field defects were peripheral constriction (8 patients, 28%), central scotoma (7 patients, 25%) and ring scotoma (4 patients, 14%); in a small percentage of patients no abnormalities were detected at the visual field (4 patients, 14%). We found abnormal chromatic test in nine patients (32%) of our cohort, data not reported in other studies, except for one reported patient [29]. The a-b wave amplitude of the scotopic ffERG and the amplitude of the 30 Hz flicker ERG were reduced similarly in the majority of patients (22 patients, 78%). The mfERG RAD was found reduced between 0-20 degrees in 15 patients (53%), whereas localized dysfunction was found within 0-5 degrees in another six patients (21%) and reduced RAD within 10-20 degrees was found in only two patients (7%); mfERG RAD was found normal in four patients (14%) and in one patient this data was not available.
As above mentioned, and based on previous clinical reports of PRPH2 families, we identified seven different phenotypes associated with PRPH2 mutation in our cohort. Patients displayed clinical features varying from RP to MD. For instance, we identified a total of five patients (17%) with autosomal dominant RP (ADRP), two patients with MD (7%), three patients (10%) with AVMD, six patients (21%) with PD, two patients (7%) with CACD, four with ECA (14%), and six patients (21%) with PDSFF. Fundus aspect and retinal morphological features detected by SD-OCT and FAF are reported in Table 2. Representative examples of the seven different phenotypes are illustrated in Figure 1.       An unusual feature already discussed and reported in the literature [20,30,31] was the presence of monocular CNV in five affected patients (17%) in our group presenting with different phenotypes (PD, CACD, ECA).

Genetic Findings
We studied 11 families with seven distinct PRPH2 genetic variants. Among these, we found three novel PRPH2 variants not previously reported: the same variant c. Concerning the modifier genes, we found a total of five variants (four on ABCA4 gene and one on ROM1 gene) in five patients of five unrelated families.
Genotype data including a detailed list of PRPH2 variants, genetic modifiers and correlated clinical diagnosis are presented on Table 3.

Discussion
We performed a retinal multimodal study in a cohort of patients carrying causative mutations of the PRPH2 gene that, to our knowledge, represents at the present time the biggest study in Italy.
The study was conducted in a cohort of 28 PRPH2 patients with the aim of describing the clinical variability of the wide spectrum phenotype, which was classified in seven main types. The present work also described the presence of monolateral choroidal neovascularization in five patients, as an unexpected but relevant feature, unusually associated with other IRD.
Among the PRPH2 variants found to be pathogenic in our cohort, we also described three novel mutations, one of which was found in members of two unrelated families.

Phenotype-Genotype Variability of PRPH2 Disease Related Spectrum
We found the clinical diagnosis and the classification of the disease considering the variable clinical spectrum at presentation to be challenging. Despite it appearing that only one gene was involved in the pathogenesis of the disease, the retinal dystrophy presented in almost seven different phenotypes involving the peripheral retina (i.e., retinitis pigmentosa, extensive chorio-retinal atrophy, pattern dystrophy-simulating fundus flavimaculatus), or the central macula (i.e., macular dystrophy, AVMD, CACD, pattern dystrophy), as illustrated in Figure 1.
Indeed, we observed in our cohort that phenotype variability was present: (1) in unrelated families carrying different mutations (inter-familiar genetic variability), as expected from already reported studies [7,13,14] and depicted in Figure 1. (2) in different unrelated families carrying the same mutation (inter-familiar phenotype variability), as reported in Figure 2.
The relationships between the clinical features and genetic variants are still unclear because the same genetic variant can affect rods and cones differently. [3,7] Therefore, without consistent genotype-phenotype correlations, the accepted view is that a single mutation in PRPH2 may cause a spectrum of phenotypes, impacting on both the central photopic system and peripheral scotopic cellular elements. In other reports it is evidenced that many genetic variants are mostly found in the D2 loop [11,20], which is critical for protein-protein interactions. In agreement, we found that most of our patients have a mutation in this domain, except for family 2 (Trp97*), family 6 (Arg46*), and family 10 (Ser301A). vimaculatus), or the central macula (i.e., macular dystrophy, AVMD, CACD, pattern dystrophy), as illustrated in Figure 1.
Indeed, we observed in our cohort that phenotype variability was present: (1) in unrelated families carrying different mutations (inter-familiar genetic variability), as expected from already reported studies [7,13,14] and depicted in Figure 1. (2) in different unrelated families carrying the same mutation (inter-familiar phenotype variability), as reported in Figure 2. We found it valuable to acquire FAF imaging for all patients. By analyzing the 50 • and 30 • images (Figure 1) we classified the PRPH2 retinal dystrophy spectrum and found common characteristics of seven different patterns. To explain the uncommon and variable presentation of this monogenic disease, as already hypothesized, we accounted for other factors such as genetic background, genetic modifiers, and/or environmental factors that may affect phenotypes and outcomes [7,8,[15][16][17]. As recently reported, it is likely that also mRNA and protein expression levels and/or post transcriptional regulatory mechanisms are intermediate factors between gene expression and clinical phenotypes [32,33].
Of interest, we documented different presentations in unrelated families carrying the same identical gene mutation either already described, as found in families 1, 3, 5, and 9 (c.499G > A) or novel (c.734dup), as reported in families 4 and 8 (as illustrated in Figure 2). About the cases with the c.734dup mutation, the phenotype variability could be given by the genetic modifiers ABCA4, which resulted, however, differently mutated in both families, and could influence the prognosis.
In addition to this interfamilial variability, an interesting feature that we found is the important intrafamilial variability identified in family 2, where the same identical mutation (c.290G > A) produced five different presentations (PD, PDSFF, ECA, ADVM, and ADRP), as illustrated in Figure 3. This example enabled us to think that there could be unidentified factors, more than the classic genetic one, that can influence the translation of the phenotype. Moreover, we could further observe a follow-up of 6 years of a member of Family 2 (F2-III-7) showing AVMD ( Figure 3A,B) feature at the onset of the disease (age of 38 y/o) with a subsequent reabsorption of the vitelliform lesion without progression of the outer retinal layer to atrophy, as usually expected [34], which can justify the good BCVA at the present. This example enabled us to think that there could be unidentified factors, more than the classic genetic one, that can influence the translation of the phenotype. Moreover, we could further observe a follow-up of 6 years of a member of Family 2 (F2-III-7) showing AVMD ( Figure 3A,B) feature at the onset of the disease (age of 38 y/o) with a subsequent reabsorption of the vitelliform lesion without progression of the outer retinal layer to atrophy, as usually expected [34], which can justify the good BCVA at the present.

Unreported Clinical Functional and Morphologic Characteristics of PRPH2-Related Disease
Although previous authors have tried to find out some distinctive traits of this spectrum disorder, as above-mentioned, this is one of the most variable IRD. In addition to that already reported in other IRD, we found an alteration of the chromatic sense, which was interestingly reduced only in those PRPH2 patients with a reduction of BCVA. This observation was not confirmed by Sonia H et al. [35] who reported an alteration in chromatic perception even with good BCVA, however the study cohort was made of patients with only best maculopathy.
On the functional assessment by electroretinographic signals, Rola Ba-Abbad et al. [36] described a case series of six patients (51.6 ± 11.86 years old), all with PRPH2 mutations but with different retinal involvement, all with an electronegative electroretinogram waveform (full field scotopic and photopic ERG), later confirmed only by one more report [37]. In our cohort, which appears slightly older (58 ± 12.35 years old) none of our patients displayed an electronegative ERG. It is likely that the electronegative ERG is not pathognomonic of PRPH2 related dystrophy, contrary to that previously hypothesized [36]. In addition, since we found similarly reduced signals derived from scotopic and photopic cellular systems of the outer retina, we were not able to establish whether PRPH2 related disorder mimics a cone-rod or a rod-cone dystrophy Concerning the electrofunctional assessment of the macular region, we were able to record mfERG in almost all subjects and found reduced RAD in the majority [17]. Interestingly of these, eight had preserved BCVA (20/20 Snellen) and only slightly macular involvement evident at the SDOCT. As far as we know, there are only a few mfERG studies [38,39] in patients with this phenotype and one of these included only four patients with evident macular involvement. Our finding of reduced mfERG responses describes a dysfunction of photoreceptors and bipolar cells in this retinal degeneration.
Of interest, the presence of CNV was relevant in our cohort. A recent report by Yousra Falfoul et al. [40] assessed a frequent macular involvement with CNV in RP patients, enough to consider the research of PRPH2 gene mutation, when a CNV is observed. In agreement with this observation, we found monocular CNV in five subjects, presenting with RP and PD phenotypes, as already reported [20,30,31], and associated with ECA phenotype, not previously reported (F3-I-2, Figure 1 and F2-II-4, Figure 3). The follow-up of CNV found in our cohort was not complete, as only one patient underwent anti-vascular endothelial growth factor (VEGF) intravitreal injection in our center.
Our findings are relevant because, as previously suggested [40], all PRPH2 patients displaying PD or RP phenotypes should be followed by SD-OCT and OCTA for the possibility of developing CNV as a complication of the disease. This alert should be extended to PRPH2 patients displaying ECA features. This agrees with a previous OCTA study that highlighted the importance of assessing vascular retino-choroidal alterations, such as the already described increase in the size of the foveal avascular zone (FAZ) at the superficial vascular plexus (SVP) or deep capillary complex (DCP), appreciable in PRPH2 patients presenting a phenotype of CACD and ECA [12].
Concerning the imaging, FAF was revealed as a useful tool to detect and observe the seven phenotypes of this IRD and especially PDSFF. This phenotype was characterized by the appearance of a speckled point of hypo-and hyper-autofluorescence in the posterior pole and beyond the vascular arcades (see Figures 1 and 3) Only the thorough acquisition of the FAF modality in all patients allowed the seven patterns (see Figure 1) to be distinguished and the PRPH2 IRD to be easily differentiated from any others that could have been misdiagnosed using only the SD-OCT scans. All this let us propose the FAF modality as the most appropriate morphological method to categorize the retinal prototypical charac-teristics of the PRPH2 disease spectrum, especially the PDSFF type which presents peculiar abnormalities along the vascular arcades and otherwise not detectable easily by SD-OCT. In agreement with this observation, a previous study [41] described that quantitative fundus autofluorescence (qAF) may help to distinguish patients with PRPH2 gene mutations. In this group, qAF values were lower than in patients with ABCA4 gene mutations but higher with respect to control subjects [22].

Conclusions
The identification of PRPH2 IRD is challenging, and the rate of affected population may be underestimated because of the clinical variability of the different phenotypes, and thus the numerous misdiagnosed cases with limited access to genetic testing. We reported three novel PRPH2 variants: the c.734dup associated with PD, PDSFF and CACD in two unrelated families, the c.903del associated with RP, and the c.742C > A associated with PD in another two distinguished families. We observed new electrophysiological features of the PRPH2 spectrum phenotypes, consisting of an impairment of the mfERG, even in those patients with preserved BCVA and only slightly macular SD-OCT alterations. We propose in PRPH2 patients FAF modality as the most suitable and accessible imaging method to identify the disease phenotypes and OCTA acquisition to promptly detect CNV, even in patients with ECA phenotype, and for a correct diagnosis, advocating the programing of a correct follow up for appropriate management of this complication.