Syntaxin 3 is a N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) protein that catalyzes fusion between vesicles and their target membranes (1). In the retina, syntaxin 3 is expressed exclusively by photoreceptors and bipolar cells (2–5), where it is required for neurotransmitter release (4, 6, 7). Recently, biallelic loss-of-function mutations in the human retinal-specific syntaxin 3 spliceform, syntaxin 3B (STX3B), have been linked to an early-onset severe retinal dystrophy in young children (8). Furthermore, biallelic postnatal inactivation of the syntaxin 3 gene (Stx3) in mouse photoreceptors has shown to result in the rapid degeneration of photoreceptors and a dramatic thinning of the outer nuclear layer (ONL) of the retina (8, 9). Thus, in addition to catalyzing the exocytic release of neurotransmitter release that underlies chemical synaptic transmission at photoreceptor and bipolar cell synaptic terminals (4, 6, 7), Stx3 also has an essential role that is necessary for photoreceptor survival.

A cellular hallmark of many inherited disorders of vision is the progressive loss of photoreceptors with age. Given the rapidly devastating consequences of biallelic STX3/stx3 loss-of-function in both humans and mice, we wondered whether monoallelic Stx3 loss-of-function might lead to retinal degeneration later in life. As a first step towards addressing this possibility, we examined the outer nuclear layer of older adult mice that were heterozygous for Stx3 with those of age-matched controls. Analysis of outer nuclear layer thickness, number of photoreceptor somata and sprouting of bipolar cell dendrites into the outer nuclear layer (ONL) indicated that inactivation of a single allele of Stx3 does not drive age-related photoreceptor loss in the mouse retina.

In this study, we asked whether having only a single functional Stx3 allele might be a risk factor for age-related photoreceptor death. To address this question, we examined and compared the retinae of older adult mice that were heterozygous for Stx3 with those of similarly-aged controls (control: 22 ± 1 months, n=15; Stx3 het: 20 ± 2 months, n=7; p = 0.6922). In the representative confocal images shown in Figure 1A and in the compiled data from multiple animals ( Figure 1B ), ONL thickness was not diminished in mice that were heterozygous for Stx3 when compared to controls (control: 63 ± 2 µm, n=15; Stx3 het: 73 ± 3 µm, n=7; p = 0.0164). Furthermore, there was no difference in the number of photoreceptor somata per unit area between groups ( Figures 1A, C ; control: 70.5 ± 2.9, n=15; Stx3 het: 73.2 ± 2.8, n=7; p = 0.6173). Together, these results indicate that the outer nuclear layer (ONL) in mice heterozygous for Stx3 is comparable to that of control mice ( Figure 1 ).

When rod photoreceptors die and/or their ribbon-style synapses become non-functional, rod bipolar cells extend their dendrites beyond the outer plexiform layer (OPL) and into the ONL (19–23). We therefore measured and compared the length of PKC-labeled rod bipolar cell dendrites as a proxy of rod photoreceptor loss in older adult mice heterozygous for Stx3 and in similarly-aged control mice. Results show that dendritic lengths were virtually identical amongst the two groups, with each group having a similar percentage of dendritic length distributions that included the occasional longer ONL sprout ( Figure 2 ). Taken together, these results demonstrate that one functional Stx3 allele is sufficient to maintain long-term photoreceptor viability.

Syntaxin 3 is the only plasma membrane syntaxin known to be expressed by photoreceptors (2, 3, 24). In humans, biallelic loss-of-function mutations in syntaxin 3A, a syntaxin 3 splice form expressed widely throughout the body outside of the retina, gives rise to a devastating gastrointestinal disorder that presents in infancy (8, 25, 26). If the mutations are located in exons that are conserved between syntaxin 3A and the retinal-specific syntaxin 3 spliceform, syntaxin 3B (5, 8), the children additionally exhibited an early onset severe retinal dystrophy (8). In mice, the global inactivation of Stx3 is embryonic lethal (10), while inactivation of Stx3 selectively in photoreceptors produced a rapid loss of photoreceptors and a dramatic reduction in ONL thickness (8, 9). Thus, in addition to its role in synaptic transmission at retinal ribbon-style synapses (4, 6, 7), syntaxin 3 is also required for photoreceptor survival.

In this study, we examined the effects of deletion of a single Stx3 allele on the outer retina. We found no difference in the thickness of the outer nuclear layer between older adult controls and older adult mice that were heterozygous for Stx3. In addition, we did not observe a decrease in the number of photoreceptor somata per unit measure or an increase in the sprouting of rod bipolar cell dendrites. The latter might be expected if rods had died or retracted their spherules at a higher rate in older adult Stx3 heterozygous mice relative to age-matched controls or if the Stx3 heterozygous rod to rod bipolar cell synapses were non-functional (19, 21–23). We did note that most of the rod bipolar cell dendrites in older adult Stx3 heterozygous mice appropriately contacted rod terminals, suggesting that the primary reason for a lack of exuberant dendritic sprouting is that the dendritic targets, the rod terminals, demarcated by Stx3 immunolabeling, were still present and located close to the OPL/ONL border.

One of the motivations for conducting this study was to predict whether loss-of-function mutations in one STX3 allele might increase the risk of photoreceptor loss in human subjects later in life. Our results suggest that Stx3 is haplosufficient for photoreceptor survival, even at older ages. However, the situation could be very different if, rather than a loss-of-function mutation, there were a monoallelic dominant negative mutation. Indeed, SNAREopathies have been reported in which dominant mutations in one gene negatively affect the functionality of the wild-type transcript (27). Interestingly, of the identified human STX3 mutations associated with visual impairment to-date, all have been biallelic loss-of-function mutations (8). Thus, for these patients, the introduction of a wild-type gene could be sufficient to rescue the remaining photoreceptors and prevent further photoreceptor loss.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Animal Welfare Committee of the University of Texas Health Science Center at Houston.

RH conceived and directed the project. MP-H and AS performed the experiments. MP-H, CD, and RH analyzed data. All authors contributed to data interpretation. RH and MH-P prepared the first draft. All authors contributed to the article and approved the submitted version.