)ORULGD6WDWH8QLYHUVLW\/LEUDULHV
2020 The Function of Snare Complex Proteins in Photoreceptor Outer Segment Formation Shows Support for the Classical Evagination Model Carley Marie Huffstetler
Follow this and additional works at DigiNole: FSU's Digital Repository. For more information, please contact [email protected] THE FLORIDA STATE UNIVERSITY
COLLEGE OF ARTS AND SCIENCES
THE FUNCTION OF SNARE COMPLEX PROTEINS IN PHOTORECEPTOR OUTER
SEGMENT FORMATION SHOWS SUPPORT FOR THE CLASSICAL
EVAGINATION MODEL
By
CARLEY MARIE HUFFSTETLER
A Thesis submitted to the
Department of Chemistry and Biochemistry
In partial fulfillment of the requirements for Honors in the Major
Degree Awarded:
Spring, 2020 The Members of the Committee approve the thesis of Carley Marie Huffstetler defended on April 23rd, 2020.
______
James M. Fadool, Ph.D.
Directing Professor
Department of Biological Sciences
______
Bridget A. DePrince, Ph.D
Committee Member
Department of Chemistry and Biochemistry
______
Justin G. Kennemur, Ph.D
Committee Member
Department of Chemistry and Biochemistry
ACKNOWLEDGEMENTS
I would like to give a huge thanks to Dr. James Fadool for all of his teaching,
training, and guidance over the past few years and for giving me the opportunity and tools to have learned so much in his lab. I would also especially like to thank Dr. Bridget
DePrince and Dr. Justin Kennemur for serving on my committee. Thank you to Sixian
Song, Jacob Dilliplane, and Jacob Myhre for friendship and conversation in the lab.
Thank you to Dr. Debra Fadool and all the members of her lab for providing me with
additional guidance, friendship, and a great weekly lab meeting! Finally, thank you to
the FSU-CRC and The Foundation Fighting Blindness for funding our research.
i TABLE OF CONTENTS
L F iii A “ iv
1 INTRODUCTION 1-5 1.1 SNAREs at the Photoreceptor Synapse . . 2-3 1.2 Outer Segment . 3-4 1.3 Inner Segment 4-5
2 TWO MODELS FOR OUTER SEGMENT MORPHOGENESIS 5-13 2.1 The Classical Evagination Model 5-11 2.1.1 Theories for Initiation of Ciliary Membrane Evagination .. 9-11 2.2 The Alternative Model: Vesicular Targeting 11-13
3 EVIDENCE AGAINST THE VESICULAR TARGETING MODEL AND EXPANSION UPON THE EVAGINATION MODEL . 14-19
4 SNARE PROTEIN FUNCTION IN PHOTORECEPTORS: SUPPORT FOR THE EVAGINATION MODEL . 19-25 4.1 Zebrafish as a Model Organism for Eye Genetics and Development . 21-23 4.2 The Function of Syntaxin 3 and Syntaxin Binding Protein 1B in OS Development . 23-25
5 POTENTIAL METHODS FOR TESTING THIS HYPOTHESIS . 25-29 5.1 Cloning of stx3 DNA . 26-27 5.2 In vitro Transcription and Rescue 27 5.3 Anticipated Results .. 27-28
6 CONCLUSIONS 29
7 REFERENCES 29-37
ii LIST OF FIGURES
1. G “ R C . 2
2. R “ E M .. ..9
3. Representation of C V T M . . 14
4. E M EM . 25
iii ABREVIATIONS AND SYMBOLS
1. 4C12 mouse monoclonal antibody that labels rods 2. CC connecting cilium 3. dpf days post fertilization 4. EM - electron microscopy 5. FYVE domain - named after the four proteins it was first found in: Fab1, YOTB/ZK632.12, Vac1, and EEA1. 6. GPCR G-protein coupled receptor 7. hpf hours post fertilization 8. IS inner segment 9. OKR optokinetic response 10. ONL outer nuclear layer 11. OS outer segment 12. PCDH21 - protocadherin 21 13. PCDH21 the gene which encodes protocadherin 21 14. PFA para-formaldehyde 15. PI3P - phosphatidylinositol 3-phosphate 16. RDS - rim protein retinal degeneration slow (also known as peripherin/rds) 17. SARA - Smad anchor for receptor activation 18. SNARE soluble N-ethylmaleimide sensitive factor attachment protein receptor 19. SNAP - soluble N-ethylmaleimide sensitive factor attachment protein 20. stx3 the t-SNARE protein syntaxin3 21. stx3 the gene that encodes the t-SNARE protein syntaxin3
22. stxbp1b the SNARE regulator protein syntaxin binding protein 1b
23. stxbp1b the gene that encodes syntaxin binding protein 1b
24. VAMP - Vesicle associated membrane protein
25. ZPR1 mouse monoclonal antibody that labels red and green cones
iv 1. Introduction
Rod and cone photoreceptors, which are specialized neurons located in the retina, are responsible for transducing light into the chemical and electrical signals of the brain. Rod photoreceptors mediate vision in dim light, whereas cones function in bright light and provide color vision. Both rod and cone cells are composed of an apically extending outer segment (OS), an inner segment (IS), the cell body, and a single axon leading to a terminal (Figure 1).
Phototransduction occurs in the outer segment, and visual information from the photoreceptor is transmitted to second order neurons at the synapse. SNARE proteins are essential for membrane fusion between vesicles and the plasma membrane during post-Golgi transport, ER and ERGIC trafficking, and autophagosome formation (Brennwald et al. 1994, Muppirala et al.
2011, Nair et al. 2011, Purves et al. 2001). In photoreceptors, SNAREs are predictably implicated in synaptic transmission, but they are also associated with vesicles carrying cargo designated for trafficking to the outer segment (Datta et al. 2015, Zulliger et al. 2015,
Kandachar et al. 2018). I I involvement of SNARE complex proteins in the formation of the photoreceptor OS are consistent with the classical evagination model for OS formation. Further I will discuss the putative interaction of the t-SNARE syntaxin 3 and syntaxin binding protein 1b and its involvement in the formation of the photoreceptor OS.
1
Figure 1. General structure of rod and cone photoreceptor cells. Proteins essential for phototransduction are synthesized in the inner segment and trafficked to the outer segment via the connecting cilium. Cote 2019 https://www.routledge.com/Cyclic-Nucleotide-Phosphodiesterases-in-Health-and-Disease/Beavo-Francis- Houslay/p/book/9780849396687?utm_source=crcpress.com&utm_medium=referral
1.1. SNAREs at the Photoreceptor Synapse
Synaptic transmission is facilitated by the fusion of synaptic vesicles with the plasma membrane and the subsequent release of neurotransmitters into the synaptic cleft. SNARE protein complexes are essential for synaptic transmission (Ramakrishnan et al. 2012). They mediate fusion of neurotransmitter-filled vesicles to the cell membrane at the nerve synapse.
SNARE proteins tether the vesicle to the target membrane and assist with the fusion of their lipid bilayers by forming a SNARE complex.
SNARE protein complexes are made up of proteins bound to the membrane of vesicles, v-
“NARE VAMP
2 membrane, t-SNAREs such as SNAPs and syntaxins. Syntaxins are composed of an N-terminal sequence, the Habc domain, the SNARE domain, and the C-terminal domain which spans the target membrane. The SNARE domain contains the portion of the alpha helix that participates in the core SNARE complex. Regulatory syntaxin binding proteins are known to bind at the triple helix of the Habc domain (Dulubova et al. 1999, Liu et al. 2014). Complexins are another important class of integral membrane proteins associated with SNARE complex regulation (Hu et al. 2002). The final protein complex consists of a tight bundle of alpha helix domains that
-coil motif, tethering the vesicle to the target membrane and fusing the lipid bilayers (Gowthaman et al. 2006).
1.2. The Outer Segment
The OS is a modified cilium where phototransduction occurs. Rhodopsin is the photopigment responsible for absorbing light in rods, and cone pigments are found in cones. All light-absorbing visual pigments have similar structure consisting of an opsin protein and a specific cofactor, the retinal chromophore. Opsins are G-protein coupled receptors (GPCRs).
GPCRs are made up of seven transmembrane alpha helices that bind a specific signaling molecule. The GPCR undergoes a conformational change once the signaling molecule binds, subsequently triggering a signaling cascade. The signaling molecule for opsins is retinal, which is the aldehyde of vitamin A. Retinal via a protonated Schiff base when it is in the 11-cis form. Once it absorbs a photon, retinal undergoes an isomerization that results in conversion to its all-trans form. The conversion to the all-trans form of retinal results in the conformational change of opsin and a subsequent signaling cascade that eventually results in an electrical impulse to the visual cortex (Palczewski 2006).
3 The OS is composed of 100s of stacks of tightly packed membranous disks (Pearring et al.
2013). The OS is elongated and extends parallel to the axis of incoming light. Tight packing allows for a high concentration of opsin within a confined space. In rods, each mature disc is a self-contained structure completely detached from the plasma membrane, while in cones each mature disc retains an attachment to the plasma membrane (Sjöstrand 1953).
OS discs undergo cyclic renewal, requiring a constant, unidirectional flow of opsin and other proteins from the IS to the OS (Young 1967). Nearly 2000 opsin molecules are transported to the OS per second in each individual human photoreceptor (Yildiz and Khanna, 2012). Richard
Young demonstrated this in 1967 by injecting [3H] methionine into a rat, mouse, and frog. He then performed autoradiograms of the excised retina on various days after the injection. He found that the radiolabeled band moved up the outer segment as time after injection increased, eventually disappearing at the apex of the cell (Young 1967). In 1969, Young and Bok discovered that rod OS tips lie adjacent to a single layer of retinal pigment epithelial (RPE) cells, from which slender membranous processes extend. These membranous processes sheath up to one-quarter of the full OS length in rods, and participate in the OS renewal process by phagocytosing older disk packets (Young and Bok 1969).
W Y the realization that outer segments must be continually rebuilt, and led to the question: How are OS discs initially formed?
1.3. The Inner Segment
The IS is connected to the OS by the connecting cilium (CC). The IS contains numerous mitochondria and s ribosomes and rough endoplasmic reticulum. Proteins that are essential for phototransduction and synaptic
4 transmission are synthesized in the IS and trafficked along the CC . This nonmotile cilium resembles the transition zone of other primary cilium and contains an axoneme consisting of a
9 + 0 bundle of microtubule doublets at its base (Röhlich 1975, Horst et al. 1990). A unique aspect of the CC is that the number of microtubules in the axoneme diminishes toward the distal end when the doublets transition into a 9 + 0 arrangement of single microtubules.
2. Two Models for Outer Segment Morphogenesis
Two models have been proposed to answer this question. The first, proposed by Steinberg et al. in 1980, involves evaginations of plasma membrane at the base of the outer segment followed by fusion of adjacent evaginations to form a discrete disk. In the evagination model, fusion of post-Golgi vesicles containing opsins, as well as additional lipids and proteins for OS formation, occurs at the IS plasma membrane (Steinberg et al. 1980, Burgoyne et al. 2015,
Kandachar et al. 2018). Therefore, disruption of vesicular fusion would cause vesicles to accumulate in the IS if OS morphogenesis follows the evagination model. The second involves successive fusion events between intracellular rhodopsin-bearing vesicles at the base of the OS in order to grow nascent disks. The second model is known as the vesicular targeting model and was proposed by Chuang et al. in 2007. Disruption of vesicular fusion would cause vesicles to accumulate in the OS axoneme and at the apex of the CC if OS morphogenesis follows the vesicular targeting model. Following the publication of the Chuang et al. paper, further
“ likely correct.
5 2.1. The Classical Evagination Model
In 1976, Steven Fisher and Don Anderson examined the bases of adult squirrel cone photoreceptors and observed what appeared to be partially formed disks (Anderson and Fisher
1976). They later postulated that these disks formed by a process of outgrowth or evagination of the photoreceptor plasma membrane (Anderson et al. 1978). In 1978, observation of disk production in developing frog photoreceptors by Kinney and Fisher led to the hypothesis that these evaginations were occurring in the CC plasma membrane specifically (Kinney and Fisher
1978). Finally in 1980, Steinberg, Fisher, and Anderson studied the photoreceptors of rhesus monkeys and ground squirrels using electron microscopy following intracardiac perfusion of a glutaraldehyde-formaldehyde mixture. They first observed that the disks of a rod are stacked from the base of the OS (which they call the vitreal end) to the apex (called the scleral end) and are surrounded by the OS plasma membrane (Steinberg et al. 1980).
As mentioned earlier, mature disks in the rod photoreceptor OS are completely detached from the plasma membrane. Rods are often used for research regarding OS morphogenesis because they have this concrete distinction between mature disks that are fully detached and nascent disks that are still attached to the plasma membrane in various capacities. Proximally to the vitreal end of the rod OS, Steinberg et al. observed multiple folds in the plasma membrane of the connecting cilium. They proposed that these were the beginnings of new OS disks formed by outpouchings of the cilium plasma membrane known as evaginations
(Steinberg et al. 1980). This was in contrast to earlier models that suggested nascent disks were formed by invaginations, but was supported experimentally by a disruption of actin filaments in cones using cytochalasin D that resulted in excessive growth of nascent disks (Williams et al.
6 1988, Vaughan and Fisher 1989). If the nascent disks had been formed by invagination, they would have been surrounded by plasma membrane as they pushed out the opposite side of the
OS. There was no such extra plasma membrane present, and the overgrown disks remained open (Williams et al. 1988). This indicated that they were not formed by invagination and provided convincing support for the evagination model.
Steinberg et al. designated the upper surface of each evagination as the scleral surface, and the lower surface as the vitreal surface. The interior of each evagination is filled with cytoplasm.
As older evaginations are displaced in the scleral direction by new evaginations, they have a progressively increasing diameter and also become progressively thinner until they reach the approximate width and diameter of a mature disk (Figure 2). Scleral and vitreal surfaces of two adjacent evaginations must be joined at their perimeters in a process known as rim formation in order to complete a disk. Rim formation results in a fully internalized disk and is mediated by rim protein retinal degeneration slow (RDS, also known as peripherin/rds) (Chakraborty et al.
2014). I advancing around the circumference of the scleral and vitreal surface of the newest evagination and the evagination formed just prior (respectively), finally zippering the two surfaces together to form a fully internalized disk (Steinberg et al. 1980). The disks of cone photoreceptors are not fully internalized. EM (electron microscopy) of cones indicated the established unique hairpin shape of disk rims on the internalized side of the disk, but also suggested the presence of a second type of disk boundary on the side remaining open to the extracellular space
(Corless and Fetter 1987). The second type of boundary is U-shaped and known as an edge.
Corless and Fetter concluded that cone disks, as well as immature rod disks, have edges
7 occurring on the side of the disk that is in continuity with the plasma membrane, while rims occur on the internalized side of the disk, closest to the axoneme. Furthermore the rim formation step, specifically its resultant advancement of the ciliary plasma membrane, accounts for the needed continuous renewal of the OS plasma membrane (Steinberg et al. 1980).
In summary, the Steinberg et al. evagination model for OS disk morphogenesis can be viewed to occur in two steps. The first step involves growth of the ciliary plasma membrane into an evagination growing away from the axoneme. Anchoring of the vitreal surface of the evagination to the axoneme allows a second evagination to be initiated, which is essential for the formation of a disk, as each disk is formed by the scleral surface of one evagination and the vitreal surface of another. The second step involves rim formation by plasma membrane outgrowth at the point where the neighboring evaginations meet, facilitating formation of an internalized disk in the case of rod photoreceptors and a partially internalized disk in cone photoreceptors. Additionally, rim formation facilitates provision of new plasma membrane to allow for membrane loss during constant OS disk renewal.
8
Figure 2. Diagram of disk surface formation. Each disk has a scleral and vitreal surface. Each surface is formed by a separate evagination. Older evaginations are displaced in the scleral direction as new evaginations are formed. Lower segment of the figure depicts a tangential section. Steinberg et al., 1980 https://onlinelibrary.wiley.com/doi/abs/10.1002/cne.901900307?sid=nlm%3Apubmed
2.1.1. Theories for Initiation of Ciliary Membrane Evagination
Initiation of the ciliary plasma membrane evaginations was previously attributed to
actin polymerization, due to the fact that inhibition of actin polymerization results in the
formation of fewer mature disks than normal, and compared to the lamellipodia extension
that is known to be driven by actin polymerization (Hale et al. 1996, Nemet et al. 2014).
Burgoyne et al. challenge this claim based on comparisons between the depth of the
evaginations leading to rod disk biogenesis which is 11 nm, versus the depth for
lamellipodia, which is a minimum of 100 nm (Burgoyne et al. 2015). Lamellipodia contain a
large group of cytoskeletal and signaling proteins and their extension is driven by multiple
layers of actin filaments. It seems very unlikely that there is space for layers of these within
the 11 nm depth of an evagination when we consider that each actin filament has a
9 diameter of about 10 nm (Burgoyne et al. 2015). Alternatively, Burgoyne et al. suggest that evaginations are facilitated by a form of blebbing, which is driven by a localized detachment of the plasma membrane from the cortical actin cytoskeleton, followed by a protrusion of the plasma membrane as a result of hydrostatic pressure inside the cell. They explain the prior findings that inhibition of actin polymerization results in the formation of fewer mature disks than normal by suggesting that F-actin is more involved in the assembly of the disks following evagination (Burgoyne et al. 2015). This theory aligns somewhat with that of
Williams et al. who found that with deregulation of actin filament networks, excessive growth of evaginations occurred. This led to the prediction that instead of being essential for the evagination itself to occur, the networks are essential for regulation of how far the nascent disks grow (Williams et al. 1988). Burgoyne et al. also suggest that the bleb is regulated similarly to how Steinberg et al. proposed that elongation and slimming of older evaginations would occur while they were displaced by new evaginations (Steinberg et al.
1980, Burgoyne et al. 2015). Additionally, analysis of a 3D
OS created by Volland et al. showed invaginations of varied shapes in a bulge of the ciliary plasma membrane, which led them to postulate that, prior to evagination, the ciliary plasma membrane may undergo an invagination in order to help shape the lamellae
(Volland et al. 2015). Both processes would still culminate in disk rim formation and subsequent disk internalization.
Burgoyne et al. also hypothesize that PCDH21 containing contacts are made between the edge of evaginations and the IS plasma membrane and that these stabilize and control evaginations. In support of this theory, deletion of PCDH21 results in disorganized OS disk
10 development (Rattner et al. 2001). Furthermore, Burgoyne et al. observed that peripherin
was present at the base of evaginations on the axonemal side, but absent from the leading
edges of the ciliary plasma membrane evaginations, which was consistent with its
hypothesized role in rim formation (Chakraborty et al. 2014, Burgoyne et al. 2015). The
locations of both PCDH21 and peripherin were confirmed using preembedding labeling.
PCDH21 labeling exhibited fibers emanating from PCDH21 on the edges of evaginations at
the base of the OS to the periciliary ridge of the IS (Burgoyne et al. 2015). A subsequent
study by Salinas et al. found that photoreceptor disk formation is dependent on the
suppression of ciliary ectosome release by peripherin. They demonstrated that in the
absence of peripherin, the photoreceptor cilium will release massive amounts of ectosomes
instead of retaining ectosome membranes at the cilium to be morphed into disks (Salinas et
al. 2017). Additionally, they suggested that this second function of peripherin is performed
independently from its previously discussed role in disk rim development and is fulfilled by
a different portion of the peripherin molecule (Salinas et al. 2017).
2.2. The Alternative Model: Vesicular Targeting
In 2007, Chuang, Zhao, and Sung proposed a competing model for outer segment morphogenesis known as the vesicular targeting model. While screening for protein interactions with the C-terminus of rhodopsin, they observed an interaction between the rhodopsin C-terminal and Smad anchor for receptor activation (SARA), which was initially recognized as an important regulator of transforming growth factor, but has recently been recognized as serving an important role in regulation of membrane dynamics (Tsukazaki et al.
1998, Chuang et al. 2007, Rozés-Salvador et al. 2018). The interaction was further supported by
11 a successful coimmunoprecipitation. Subsequent immunofluorescent labeling of mouse retinal sections showed SARA distribution on vesicular structures in the apical cytoplasm of the IS, synaptic terminals, and basal portion of the OS axoneme (Chuang et al. 2007). SARA contains a
FYVE domain that is known to bind to phosphatidylinositol 3-phosphate (PI3P) with high specificity (Stenmark et al. 2002). Immunolabeling of Vps34, the primary enzyme responsible for catalyzing the production of PI3P, revealed a high probability that PI3P is synthesized locally on nascent disks at the base of the OS (Chuang et al. 2007). This finding, coupled with the expression of SARA on the vesicular structures in the IS suggested that the FYVE domain interaction with PI3P on nascent disks may facilitate docking. Additionally, Chuang et al. hypothesized that a SNARE protein complex would be required in order to fuse the axonemal vesicles with the nascent disks. Subsequent immunolabeling showed that syntaxin 3 was present at the base of the OS in addition to the IS and synaptic terminal. Further immunolabeling indicated similar localization for SARA, leading to the hypothesis that rhodopsin- “ARA interaction with PI3P, rhodopsin, and syntaxin 3 (Chuang et al. 2007).
Chuang et al. also utilized EM to study mouse rods for better membrane preservation, that entailed transcardial perfusion with saline prior to fixed perfusion using 4% PFA/3.75% acrolein (Chuang et al. 2007). The resultant electron micrographs exhibited a lack of the open disks at the base of the OS that would have supported the evagination model. Instead the Chuang et al. micrographs indicated vesicles that were gathered at vitreal end of OS. The basal vesicles, along with the stack of mature disks, appeared to be completely enclosed by the plasma membrane (Chuang et al. 2007). These results led
12 Chuang et al. to reject the evagination model and instead propose that mature disks result from a process in which rhodopsin-containing vesicles at the base of the OS fuse in order to grow nascent disks which possess high levels of phosphatidylinositol 3-phosphate (PI3P). This fusion
“ARA direct interaction with PI3P, rhodopsin, and syntaxin 3 (Chuang et al. 2007). For additional support of this model, Chuang et al. cited that OS targeting of rhodopsin was defective in rods when FYVESARA was overexpressed or suppressed, and when syntaxin 3 was suppressed, which resulted in disorganized OSs filled with unfused vesicles.
These unfused vesicles would accumulate in the OS axoneme, connecting cilium, and IS (Chuang et al. 2007).
In 2015 Chuang, Hsu, and Sung sought to expand upon the vesicular targeting model by investigating the pathway by which ciliary membrane proteins, specifically rhodopsin, translocate through the CC transition zone. Performing immuno-electron microscopy in mouse rods facilitated this investigation. A polyclonal antibody was used to recogniz cytoplasmic C-terminus. In addition to the expected heavy labeling in the OS disks, it showed that rhodopsin was present on both the plasma membrane of the CC and in vesicles on the lumen of the CC (Chuang et al. 2015). These observations led to the conclusion that rhodopsin transported across the CC on the plasma membrane would diffuse into the OS plasma membrane, while the rhodopsin in the vesicles on the CC lumen would directly fuse onto nascent disks, facilitating their growth and maturation (Chuang et al. 2015). In support of these conclusions, EM showed many of these rhodopsin-laden vesicles in close proximity to the nascent disk membranes at the base of the OS.
13 In summary, the vesicular targeting model was proposed based on observations that syntaxin 3 and SARA are enriched at the base of the OS, where they could facilitate rhodopsin vesicle fusion to nascent disks, that OS targeting of rhodopsin was defective in FYVESARA - overexpressed, SARA-suppressed, and syntaxin 3-suppressed rods which resulted in a disorganized OS and vesicular accumulation, and that EM failed to show open disks at the base of the OS (Chuang et al. 2015). Therefore, the vesicular targeting model states that multiple nascent disks are formed simultaneously at the base of the OS via fusion of rhodopsin-laden vesicles transported from the inner segment which allow the nascent disks to grow to their mature size (Figure 3).
Figure 3. Depiction of the vesicular targeting model. New discs are assembled by vesicular trafficking and membrane fusion events mediated by SARA, PI3P, and Syntaxin 3. Figure shows axonemal vesicles pre-fusion and while fusing to grow nascent disks. Chuang et al., 2007. https://www.cell.com/cell/fulltext/S0092-8674(07)00827- 6?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0092867407008276%3Fshowall%3Dtrue
14 3. Evidence Against the Vesicular Targeting Model and Expansion Upon the Evagination
Model
Further investigation into the mechanism of OS disk morphogenesis occurred after the introduction of the vesicular targeting model in 2007. These studies, notably those of Ding et al,
Volland et al, and Borgoyne et al. in 2015, resulted in overwhelming support for the classical evagination model proposed by Steinberg et al. in 1980. Their findings challenge two pivotal claims of the vesicular targeting model: that the basal OS shows no evidence of open disks or evagination and that rhodopsin-containing vesicles can be observed in the CC lumen.
Ding, Salinas, and Arshavsky, aimed to test the key predictions of each model using three experiments that would not rely on the visualization of new discs on EM alone. The first experiment entailed treatment of sections of mouse retinas with a mixture of tannic acid and uranyl acetate, followed by EM (Ding et al. 2015). Tannic acid was used because it penetrates intact membranes poorly, thus allowing it to distinguish between membranes that are exposed to the extracellular space (such as newly formed disks and plasma membrane) and intracellular membrane structures (such as mature rod OS disks). Tannic acid staining of spaces that are open to the extracellular space is intense, while the staining of intracellular membrane structures is weak. Ding et al. observed that a small number of basal rod disks (6-12) showed intense staining, but staining of fully internalized, mature disks was weak and no vesicular structures were observed. Additionally, as a control experiment, retinal membranes were permeabilized using 0.5% saponin, allowing the tannic acid to infiltrate and stain intracellular as well as extracellular membranes. As expected, this additional treatment showed staining that
15 was uniformly intense for both new and mature OS disks, and even extended to other intracellular membrane structures, such as the mitochondria of the IS (Ding et al. 2015). These results confirmed that the newly formed rod disks had been open to the extracellular space.
The second experiment by Ding et al. involved analysis of rhodopsin membrane topology.
This analysis was conducted based on a major distinction in the orientation of rhodopsin molecules in newly formed disks between the two competing models. The evagination hypothesis suggests that rhodopsin in new disks has its N-terminus exposed to the cytoplasm, while mature disks have the rhodopsin C-terminus exposed to the cytoplasm and the N- terminus faces the disc lumen (Ding et al. 2015). This change is due to an inversion of the transmembrane rhodopsin protein, which occurs during rim formation upon fusion of disk edges. In contrast, the vesicular fusion hypothesis insinuates that the C-terminus will be exposed to the cytoplasm for both nascent and mature disks because there is no membrane inversion during vesicular fusion (Chuang et al. 2007, 2015). Ding et al. used postembedment immunogold labeling of rhodopsin with antibodies that would differentiate between the N or C terminus in order to determine which model matched the rhodopsin membrane topology of the OS disks. Results showed that labeling for the N-terminal antibody was approximately 2.2- fold higher over the cytoplasmic surface than the C-terminal antibody labeling, indicating that rhodopsin N-terminus faces the extracellular space in nascent disks which is consistent with the evagination model (Ding et al. 2015).
The third and final experiment involved the use of fixation techniques proven successful by other investigators in order to prove that artifacts of inadequate tissue fixation were
C EM that showed fully internalized disks at the base of the
16 rod OS. Previous EM studies had indicated that cellular structure in neurons is subject to rapid
, and that change is further facilitated by delayed fixation such as the method used by Chuang et al. in which the animal was perfused with saline prior to administering the fixative (Hasty and Hay 1978, Tao-Cheng et al. 2007). Ding et al. replicated the Chuang et al. method of perfusing first with saline, and observed a population of rods that contained the purported vesicular structures at the base of the OS, but also observed that new disks in most of the rods remained open to the extracellular space as indicated by the evagination model. They postulated that this inconsistency further emphasized the fragility of the OS base (Ding et al. 2015). Another facet of Chuang et al. of acrolein rather than glutaraldehyde. They claimed that the monoaldehyde acrolein exhibits rapid tissue penetration and consequently would preserve disk membranes closer to their native state. Although use of acrolein has been limited as a result of a study indicating that it solubilizes membranes, when Ding et al. replaced their glutaraldehyde with acrolein for an
C acrolein perfusion following saline perfusion), the new disks of the rod OS remained open and highly exposed to the tannic acid (Ding et al.
2015). These results provided proof that the structural artifacts seen in the Chuang et al. EM were caused by the delayed fixation rather than the use of acrolein. Further testing of the
Chuang et al. method of perfusion in comparison to other well-proven methods resulted in similar conclusions regarding the unreliability of delayed fixation (Burgoyne et al. 2015, Volland et al. 2015).
Volland et al. used a combination of fixed cardiac perfusion and high-pressure freezing followed by freeze substitution to generate a 3D model of a tomogram of basal
17 OS. This 3D model made it possible to analyze z-plane images of rod OSs from mouse, cat, and monkey retinas in which the disks at the base of the OS first appeared to be closed, but upon analysis in other z-planes could be plainly observed to have an opening to the extracellular space (Volland et al. 2015). Volland et al. also makes an interesting observation regarding the number of open nascent disks at different times of day for mice versus monkeys. They found that though open nascent disks are present at all times, their quantity was elevated during the first part of the daily light cycle in monkey rods, but decreased during this same period for mice
(Volland et al. 2015). They postulate that this may be the result of the respective diurnal and nocturnal habits of monkeys and mice (Volland et al. 2015).
An important aspect of the vesicular targeting model is that the rhodopsin-filled vesicles located on the CC lumen directly fuse onto nascent disks, facilitating their growth and maturation (Chuang et al. 2015). Burgoyne et al. call into question C claim that there were rhodopsin-filled vesicles in the lumen of the CC, and they instead postulate that rhodopsin is primarily transported on the ciliary plasma membrane (Burgoyne et al. 2015).
CryoimmunoEM of mouse retinal sections was performed using antibodies for rhodopsin in order to test for the presence of vesicles containing rhodopsin in the CC lumen. This staining revealed the expected rhodopsin-containing cytoplasmic vesicles in the IS, but rhodopsin staining in the CC was only observed on the ciliary plasma membrane (Burgoyne et al. 2015). No staining was observed in the lumen of the CC to indicate that there may have been rhodopsin- filled vesicles there. Further, rhodopsin staining in the OS was only observed in the disks and the plasma membrane (Burgoyne et al. 2015). EM showed very small vesicles or particles in the ciliary lumen that exhibited sticklike projections, but these did not stain positive for rhodopsin
18 (Burgoyne et al. 2015). Cumulatively, these results led to the conclusion that most rhodopsin is transported via the ciliary plasma membrane; however it is also mentioned that components of intraflagellar transport complexes and anterograde motor kinesin II have consistently been identified as essential for opsin trafficking and subsequently OS development (Marszalek et al.
2000, Pazour et al. 2002, Keady et al. 2011, Trivedi et al. 2012, Burgoyne et al. 2015). Notably, the only rhodopsin-containing vesicles observed were in the IS near the base of the CC
(Burgoyne et al. 2015).
The Burgoyne et al. group also investigated Ch
Munc-18 are present in the OS and their hypothesis that syntaxin 3 has a role in the fusion of rhodopsin-filled vesicles at the base of the OS for nascent disk formation (Chuang et al. 2007).
Contrary to Ch reembedding labeling showed that syntaxin 3 is not present at the base of the OS and that it, along with Munc-18, is instead localized to the plasma membrane of the IS and periciliary ridge (Burgoyne et al. 2015). This localization indicates that syntaxin 3 and Munc-18 are likely to be involved in the fusion of vesicles to the IS plasma membrane rather than to OS nascent disks. Previously, a study by Mazelova et al. also localized syntaxin 3 to the IS plasma membrane at the base of the cilium, and suggested that Munc-18 and syntaxin 3 are involved in rhodopsin-filled vesicle fusion there (Mazelova et al. 2009). The evidence from these two studies does not support the vesicular targeting model due to the lack of syntaxin 3 labeling on the base of the OS, where it would be needed to regulate vesicle fusion with the forming disks by forming a complex with rhodopsin and SARA (Burgoyne et al.
2015). Alternatively, Burgoyne et al. suggest that syntaxin 3 and SARA may have indirect roles in
OS development by facilitating rhodopsin transport to the cilium.
19
4. SNARE Protein Function in Photoreceptors: Support for the Evagination Model
It has been well established that SNARE proteins have a function in vesicular fusion for neurons during synaptic transmission at the nerve terminal (Rizo and Südhof 2002). In 2007,
Chuang et al. postulated an additional function for SNARE proteins. In their vesicular targeting model they claimed that the evagination model did not provide a sufficiently robust explanation for the distinct protein expression profile of the membranous disks of the OS in comparison to the OS plasma membrane. Instead they postulated that multiple nascent disks are grown simultaneously at the base of the OS via fusion of rhodopsin-laden vesicles which are
“ARA PI P the SNARE syntaxin 3. This theory has been largely disproven by studies such as that of Ding et al., Volland et al., and
Burgoyne et al. which were discussed above. In particular, Burgoyne et al. verified the presence of syntaxin 3 and its binding protein Munc-18 in the plasma membrane of the IS and in the periciliary ridge, and did not find evidence to support C present in the basal OS. Vesicle fusion at the IS periciliary ridge is essential for delivery of proteins and phospholipids to the rod OS via the ciliary membrane (Besharse and Pfenninger,
1980; Papermaster et al., 1985), and remains in alignment with the evagination model.
Other studies have indicated roles for SNARE proteins in rhodopsin trafficking. Mazelova et al. localized syntaxin 3 to the IS plasma membrane at the base of the cilium and found that
SNAP-25 is uniformly distributed on the IS plasma membrane and the synapse (Mazelova et al.
2009). Subsequently, they found that treatment with omega-3 docosahexaenoic acid caused
20 increased co-immunoprecipitation of the two proteins, colocalization at the base of the cilium, and an increased delivery of membrane to the OS. These results indicated that syntaxin 3 and
SNAP-25 pairing is regulated by omega-3 docosahexaenoic acid, and controls the delivery of rhodopsin for the biogenesis of the rod outer segment by mediating docking and fusion of rhodopsin transport carriers (Mazelova et al. 2009).
Another experiment by Kandachar et al. studied the role of the Arf4-rhodopsin complex in the sorting of VAMP7 into rhodopsin transport carriers. VAMP7 was consistently observed on rhodopsin transport carriers and was found to interact with a Rab network before forming a complex with syntaxin 3 and SNAP-25 at the ciliary base, illustrating the importance of the
SNARE motif in intracellular trafficking (Kandachar et al. 2018).
Finally, a study by Zulliger et al. sought to uncover the binding partners of photoreceptor protein peripherin-2 (RDS), discussed earlier in regard to rim formation, and rod outer segment protein 1 (ROM-1), which are synthesized in the inner segment and then trafficked to the OS via fusion with the IS plasma membrane. They used various methods such as in vitro co-expression followed by pull-down and in vivo pull-down from mouse retinas to show that SNARE proteins
Syntaxin 3B (Syn3B) and SNAP-25 are novel binding partners of RDS and ROM-1 (Zulliger et al.
2015). Zulliger et al. hypothesized that Syn3B and SNAP-25 mediate the necessary fusion of vesicles containing RDS and ROM-1 to the IS plasma membrane after finding that all four proteins interact and are localized to the IS plasma membrane, although they also concede that the interaction between SynB and RDS may instead indicate that they are trafficked together
(Zulliger et al. 2015).
21 I propose that these earlier findings, viewed alongside genetic data and imaging from our lab indicate that vesicle fusion to the IS plasma membrane for trafficking to the OS is reliant on a SNARE protein complex involving syntaxin 3 and syntaxin binding protein 1b, and this complex is essential for OS disk morphogenesis under the evagination model.
4.1. Zebrafish as a Model Organism for Eye Genetics and Development
Zebrafish have been long-established as a valuable model organism for vision research. Eye morphology, circuitry and biochemistry is conserved with other vertebrates including humans.
A mating pair of fish can provide clutches containing 100-200 fertilized eggs. Fertilization is external and the eggs and embryos are transparent, which allows for visualization of early organogenesis and ease of embryological manipulation.
Zebrafish embryo development is rapid. At 24 hours post-fertilization (hpf), all major organ systems are present and spontaneous muscle contraction can be observed. By 50 hpf, opsin expression can be detected (Raymond et al. 1995). At 55 hpf, photoreceptor outer segments and synaptic terminals are present (Schmitt and Dowling 1999). This corresponds to the 10th week of human gestation, when photoreceptor cell bodies can be observed in the outer nuclear layer of the retina. At 72 hpf, zebrafish display visually induced startle responses (Fadool and
Dowling 2008). By 5 days post fertilization (dpf), larvae are free-swimming and the optokinetic response (OKR) can be reliably detected (Brockerhoff et al. 1995).
Approximately 70% of human genes have at least one zebrafish orthologue, and 84% of the genes known to cause disease in humans have a zebrafish orthologue (Howe et al. 2013).
Targeted genetic screens are commonly used to isolate mutations that alter visual system development. The results of a genetic screen conducted in our lab will be discussed below.
22 4.2. The Function of Syntaxin 3 and Syntaxin Binding Protein 1B in OS Development
In 2005, the Fadool lab conducted a forward genetic screen in zebrafish in order to identify genes essential for photoreceptor development and survival (Morris and Fadool 2005, Alvarez-
Delfin et al. 2009, Sotolongo-Lopez et al. 2016). A mutation was isolated that resulted in rapid degeneration of photoreceptors and blindness. Immunohistochemistry using antibodies specific for rods and cones showed severe degeneration of photoreceptors in 4 day post-fertilization
(dpf) dou larvae (Robinson 2008). Gene mapping and genome sequencing identified the gene as syntaxin binding protein 1b (stxbp1b). Stxbp1b is the zebrafish orthologue of unc-18 in C. elegans, ROP in D. melanogaster, and STXBP1/Munc-18 in mammals. STXBP1/Munc-18 facilitates function of SNARE proteins in synaptic transmission (Rizo and Südhof 2002).
Furthermore, Munc-18 binding to syntaxin is essential for neurotransmission in mammals (Han et al. 2011). Injection of in vitro transcribed mRNA encoding stxbp1b into zebrafish embryos rescued the photoreceptor degeneration phenotype and restored the photoreceptor OS structure. However, the rescue was incomplete as none of the mRNA injected homozygous mutant embryos displayed visual responses, which was likely due to a lack of properly formed synaptic terminals. Although the rescue was only partial, these data provided the first genetic evidence for a role of stxbp1b in photoreceptor morphology in addition to its role in synaptic transmission, and suggested an additional role for the SNARE complex in OS formation.
To further test the role of SNAREs in photoreceptor survival, syntaxin 3 was mutagenized using CRISPR/Cas9 (Allen, 2018). Syntaxin3 (Stx3) is a t-SNARE expressed in photoreceptors
(Datta et al. 2015). In situ hybridization of embryos at 48 hours post-fertilization (hpf) showed stx3 expression in the outer nuclear layer of the retina, which suggests that stx3 has a role in
23 photoreceptor function. Larvae that were homozygous for a indel mutation that disrupted the second exon of stx3 were blind and showed photoreceptor degeneration (Allen,
2018). I conducted further genotyping and OKR testing of additional alleles of stx3.
Heterozygous 13+9 stx3 fish were mated to fish heterozygous for the additional mutant alleles
9 and 5. Additionally, fish that were both heterozygous for the 9 deletion were bred. At
6dpf all larvae were tested for optokinetic response. Larvae were euthanized and genotyped following OKR testing. Findings indicated that 9 alleles display no mutant phenotype, and they have a positive optokinetic response at 6dpf, as do fish homozygous for the 9 deletion. Compound heterozygotes for 5 and also exhibit a wild-type phenotype and no behavioral deficits.
At 6 dpf, immunolabelling of retinal cryosections for wild type zebrafish using the 4C12 antibody, which labels rods, and the ZPR1/ARR3A antibody, which labels cones, shows strong labeling in the ONL. In homozygous stx3 ONL nearly nonexistent other than in the photoreceptor cells located at the periphery of the ONL, which is a site of continual neurogenesis. EM of retinas from homozygous stx3 s a near complete absence of OS (Figure 4). Furthermore, in the distal IS a large buildup of vesicles can be observed under the apical plasma membrane, and the mitochondria are displaced toward the nucleus. Interestingly, a similar ultrastructural phenotype is observed in the stxbp1b (dou) mutant retinas. Based on the similarities between the phenotypes that result from these two mutations, I hypothesized that stx3 and stxbp1b interact and play an essential role in photoreceptor morphogenesis.
24
Figure 4. (A) Normal OS development and mitochondrial placement in WT zebrafish photoreceptors. (B) Homozygous stx3 mutants display a near complete absence of OS, and in the distal IS a large buildup of vesicles can be observed under the apical plasma membrane. The mitochondria are displaced toward the nucleus. F D A
5. Potential Methods for Testing this Hypothesis
First, in order to show that stx3 is necessary for both outer segment formation and neurotransmission at the photoreceptor synaptic terminal, a rescue of the stx3 mutation by mRNA injection could be attempted. Injected embryos would be tested for recovery of visual functionality and photoreceptor survival. Secondly, given the overlapping expression patterns and similarities of the mutant phenotypes, interaction of stx3 and stxbp1b could be tested in a heterologous expression system and in vivo.
25 5.1. Cloning of stx3 DNA
A full-length cDNA encoding wildtype stx3 was cloned from adult zebrafish retina. Retinas were dissected from adult fish, and total RNA was isolated using Trizol according to the manufacturer instructions. Following dissection, the tissue was homogenized in 500µL Trizol.
100µL of chloroform was then added to the homogenate. The aqueous phase was extracted after 15 seconds of shaking, room temperature incubation for 3 minutes, and centrifugation at
12,000 rpm for 15 minutes at 4°C. Then, 250µL of Isopropanol was added to the sample to precipitate the RNA, using 2µL of 20µL/µL glycogen as a carrier. The sample was then incubated at room temperature again for ten minutes and centrifuged again at 4°C for 10 minutes. The pellet was washed with 75% Ethanol, disrupted, then centrifuged again at 4°C for 5 minutes.
The RNA was solubilized in Invitrogen UltraPure water. Quality and quantity of the total RNA was verified by spectrophotometry and agarose gel electrophoresis. After verification of RNA isolation, cDNA was generated by a reverse transcription polymerase chain reaction using oligo dT primers and Superscript IV Reverse Transcriptase (ThermoFisher Scientific) following the ma T stx3 cDNA was amplified by PCR. Primers were
stx3 based upon sequence downloaded from Ensemble (Zebrafish
GRCz11), and were E O A T PCR was then conducted using
T F “ H F P PCR “ M T reaction was verified using gel electrophoresis. A Zero Blunt® TOPO® PCR Cloning Kit protocol was used with high efficiency NEB 5 alpha competent cells in order to clone the stx3 sequence.
Colonies were selected, and liquid cultures were screened for insert size using PCR and gel electrophoresis. Plasmid DNA was purified using the E.Z.N.A. Plasmid Mini Kit (Omega Bio-Tek)
26 and sequenced by the Sequencing Core Facility in the Department of Biological Science using
M13 forward and reverse primers.
5.2. In vitro Transcription and Rescue
A full length stx3 coding sequence with no deleterious mutations was cloned into a pCs2+ vector for in vitro transcription of mRNA using the SP6 promoter. mRNA was injected into one cell stage zebrafish embryos from a cross between two fish that were heterozygous for the
13+9 stx3 mutation. Half of the clutch was injected with mRNA, and half were used as a control group. At 6 days post fertilization (6dfp), vision was tested using the behavioral test optokinetic response (OKR). OKR is tested by immobilizing zebrafish larvae in a petri dish containing 3.0% methylcellulose gel. The dish is then placed on the stage of a microscope fitted with a rotating drum. The interior of the drum is labeled with black and white vertical stripes.
W dark stripes with their eyes and then the eyes snap back in a saccade (Brockerhoff et al. 1995).
T - It was expected that if the mRNA rescue was successful, then all of the injected larvae would be responders, while approximately one fourth of the control group would not respond as they are homozygous for the 13+9 stx3 mutation. After OKR, larvae would be euthanized and genotyped by PCR using DNA isolated from their tails. Heads would be fixed in paraformaldehyde for histology. Following fixation, tissue would be cryoprotected by incubation immersed in 10% and 30% sucrose solution. Tissue would be positioned and frozen in OCT compound, and 10µm thick cryosections would be mounted on slides. Sections would be labeled using the antibody 4C12 to label rods and ZPR1 for cones (Sotolongo-Lopez et al. 2016).
27 Further immunolabeling would be conducted in order to view ganglion cells, Mueller glia, amacrine cells, and bipolar cells.
5.3. Anticipated Results
Based on previous results, it would be expected that larvae that were homozygous for the
13+9 stx3 mutation in the uninjected control group would show a lack of immunolabeling of photoreceptors and no OKR. It is anticipated that the injection of mRNA would rescue vision and the wildtype OS phenotype. Based on the partial rescue observed for stxbp1b mutations, it is possible that stx3 mRNA injection may rescue the immunolabeling and show intact photoreceptors but an absence of the OKR (Sotolongo-Lopez et al. 2016).
Direct interaction between stx3 and stxp1b would be tested in a heterologous expression system. Expression vectors encoding stx3 or stxbp1b would be used to transfect human embryonic kidney cells (HEK cells). Briefly, stx3 would be cloned into pcDNA4 with the addition
-tag. Stxbp1b would be cloned into pcDNA4 with the addition of a myc-tag. These tags enable immunoprecipitation and western blot analysis with commercially available antibodies. Homogenates from transfected cells would be incubated with anti-flag or anti-myc antibodies and precipitated in protein A agarose. Precipitates would be subjected to SDS-page and immunoblot analysis using the reciprocal antibody. It is anticipated that precipitation of anti-myc would also pull down flag-tagged stx3, and precipitation of anti-flag would coprecipitate myc-tagged stxbp1b (Tucker et al. 2010).
To test for interactions in vivo, embryos from a mating between adult zebrafish that were heterozygous for both the stx3 stxbp1b mutation would be injected with in vivo transcribed mRNA encoding myc and flag-tagged stx3 and stxbp1b. The embryos
28 would be allowed to develop for 24 hpf and then would be homogenized. Immunoprecipitation would be completed as described above. It would again be anticipated that precipitation of anti-myc would pull down flag-tagged stx3, and precipitation of anti-flag would coprecipitate myc-tagged stxbp1b (Tucker et al. 2010).
6. Conclusions
In conclusion, I postulate that the mutant phenotype of vesicle build-up in the IS occurs because while the functioning IS continues to produce and package the proteins, such as rhodopsin, that are essential for phototransduction and OS morphogenesis, these vesicles cannot fuse with the IS plasma membrane due to lack of the needed SNARE protein complex involving the interaction of functional stx 3 and stxbp1b. This prevents the transport of rhodopsin to the periciliary ridge for incorporation into newly forming OS disks. In rods, rhodopsin constitutes more than 50% of the OS and therefore takes on a significant structural role in addition to its primary role as a phototransduction protein. The OS cannot be formed without delivery of this structural element and other proteins and lipids needed for OS disk biogenesis via molecular motor transport along the ciliary plasma membrane. The vesicles have nowhere to go, resulting in a build-up of vesicles in the IS that causes displacement of the mitochondria and eventually apoptosis. Our results support the evagination hypothesis for outer segment morphogenesis because the observed vesicle build-up is present in the IS rather than in the OS axoneme or in the apex of the CC.
29 7. References
Allen, D.M. (2018, July 17) Syntaxin 3 Is Essential For Photoreceptor Survivial In Zebrafish.
DigiNole.
http://purl.flvc.org/fsu/fd/FSU_libsubv1_scholarship_submission_1533830132_1751809e
Alvarez-Delfin, K., A. C. Morris, C. D. Snelson, J. T. Gamse, T. Gupta, F. L. Marlow, M. C. Mullins,
H. A. Burgess, M. Granato, and J. M. Fadool. 2009. Tbx2b is required for ultraviolet
photoreceptor cell specification during zebrafish retinal development. Proceedings of the
National Academy of Sciences of the United States of America. 106(6): 2023 2028.
Anderson, D. H., and S. K. Fisher. 1976. The photoreceptors of diurnal squirrels: Outer segment
structure, disc shedding, and protein renewal. Journal of Ultrasructure Research.
55(1):119-41.
Anderson, D. H., S. K. Fisher, and R. H. Steinberg. 1978. Mammalian cones: disc shedding
phagocytosis, and renewal. Investigative Ophthalmology and Visual Science. 17(2):117-33.
Brennwald, P., B. Kearns, K. Champion, S. Keränen, V. Bankaitis, and P. Novick. 1994. Sec9 is a
SNAP-25-like component of a yeast SNARE complex that may be the effector of Sec4
function in exocytosis. Cell. 79(2):245-58.
Brockerhoff, S. E., J. B. Hurley, U. Janssen-Bienhold, S. C. F. Neuhauss, W. Driever, and J. E.
Dowling. 1995. A behavioral screen for isolating zebrafish mutants with visual system
defects. Proceedings of the National Academy of Sciences of the United States of America.
92(23):10545-9.
Burgoyne, T., I. P. Meschede, J. J. Burden, M. Bailly, M. C. Seabra, and C. E. Futter. 2015. Rod
30 disc renewal occurs by evagination of the ciliary plasma membrane that makes cadherin-
based contacts with the inner segment. Proceedings of the National Academy of Sciences
of the United States of America. 112(52):15922-7.
Chakraborty, D., S. M. Conley, M. R. Al-Ubaidi, and M. I. Naash. 2014. Initiation of rod outer
segment disc formation requires RDS. PLoS ONE. 9(6):e98939.
Chuang, J. Z., Y. C. Hsu, and C. H. Sung. 2015. Ultrastructural visualization of trans-ciliary
rhodopsin cargoes in mammalian rods. Cilia. 4(4) https://doi.org/10.1186/s13630-015-
0013-1
Chuang, J. Z., Y. Zhao, and C. H. Sung. 2007. SARA-Regulated Vesicular Targeting Underlies
Formation of the Light-Sensing Organelle in Mammalian Rods. Cell. 130(3):535-47.
Corless, J. M., and R. D. Fetter. 1987. Structural features of the terminal loop region of frog
retinal rod outer segment disk membranes: III. Implications of the terminal loop complex
for disk morphogenesis, membrane fusion, and cell surface interactions. Journal of
Comparative Neurology. 257(1):24-38.
Cote, R. H. 2019. Photoreceptor Phosphodiesterase (PDE6): A G-Protein-Activated PDE
Regulating Visual Excitation in Rod and Cone Photoreceptor Cells. Cyclic Nucleotide
Phosphodiesterases in Health and Disease. (1)165-193.
Datta, P., C. Allamargot, J. S. Hudson, E. K. Andersen, S. Bhattarai, A. V. Drack, V. C. Sheffield,
and S. Seo. 2015. Accumulation of non-outer segment proteins in the outer segment
underlies photoreceptor degeneration in Bardet-Biedl syndrome. Proceedings of the
National Academy of Sciences of the United States of America. 112(32):E4400-9.
Ding, J. D., M. B. Kennedy, and R. J. Weinberg. 2013. Subcellular organization of camkii in rat
31 hippocampal pyramidal neurons. Journal of Comparative Neurology. 521(15):3570-83.
Ding, J. D., R. Y. Salinas, and V. Y. Arshavsky. 2015. Discs of mammalian rod photoreceptors
form through the membrane evagination mechanism. Journal of Cell Biology. 211(3):495
502.
Dulubova, I., S. Sugita, S. Hill, M. Hosaka, I. Fernandez, T. C. Südhof, and J. Rizo. 1999. A
conformational switch in syntaxin during exocytosis: Role of munc18. EMBO Journal.
18:4372 4382.
Fadool, J. M., and J. E. Dowling. 2008. Zebrafish: A model system for the study of eye genetics.
Progress in Retinal and Eye Research. 27(1):89-110.
Gowthaman, R., A. J. Silvester, S. K, R. Kanya K.S, and K. K.S. 2006. Modeling of the potential
coiled-coil structure of snapin protein and its interaction with SNARE complex.
Bioinformation. 1(7):269 275.
Hale, I. L., S. K. Fisher, and B. Matsumoto. 1996. The actin network in the ciliary stalk of
photoreceptors functions in the generation of new outer segment discs. Journal of
Comparative Neurology. 376(1):128-42.
Han, G. A., N. T. Malintan, N. M. N. Saw, L. Li, L. Han, F. A. Meunier, B. M. Collins, and S. Sugita.
2011. Munc18-1 domain-1 controls vesicle docking and secretion by interacting with
syntaxin-1 and chaperoning it to the plasma membrane. Molecular Biology of the Cell
22:4134 4149.
Hasty, D. L., and E. D. Hay. 1978. Freeze-fracture studies of the developing cell surface: II.
particle-free membrane blisters on glutaraldehyde-fixed corneal fibroblasts are artefacts.
Journal of Cell Biology. 78(3):756-68.
32 Horst, C. J., L. V. Johnson, and J. C. Besharse. 1990. Transmembrane assemblage of the
photoreceptor connecting cilium and motile cilium transition zone contain a common
immunologic epitope. Cell Motility and the Cytoskeleton. 17(4): 329-344
Howe, K., M. D. Clark, C. F. Torroja, J. Torrance, C. Berthelot, M. Muffato, J. E. Collins, S.
Humphray, K. McLaren, L. Matthews, S. McLaren, I. Sealy, M. Caccamo, C. Churcher, C.
Scott, J. C. Barrett, R. Koch, G. J. Rauch, S. White, W. Chow D. L. Stemple. 2013. The
zebrafish reference genome sequence and its relationship to the human genome. Nature.
496(7446):498-503.
Hu, K., J. Carroll, C. Rickman, and B. Davletov. 2002. Action of complexin on SNARE complex.
Journal of Biological Chemistry. 277(44):41652-6.
Kandachar, V., B. M. Tam, O. L. Moritz, and D. Deretic. 2018. An interaction network between
the SNARE VAMP7 and Rab GTPases within a ciliary membrane-targeting complex. Journal
of Cell Science. 131(24).
Keady, B. T., Y. Z. Le, and G. J. Pazour. 2011. IFT20 is required for opsin trafficking and
photoreceptor outer segment development. Molecular Biology of the Cell. 22(7):921-30.
Kinney, M. S., and S. K. Fisher. 1978. The photoreceptors and pigment epithelium of the larval
Xenopus retina: morphogenesis and outer segment renewal. Proceedings of the Royal
Society of London - Biological Sciences. 201(1143):149-67.
Liu, X., R. Heidelberger, and R. Janz. 2014. Phosphorylation of syntaxin 3B by CaMKII regulates
the formation of t-SNARE complexes. Molecular and Cellular Neuroscience. 60:53-62.
Marszalek, J. R., X. Liu, E. A. Roberts, D. Chui, J. D. Marth, D. S. Williams, and L. S. B. Goldstein.
2000. Genetic evidence for selective transport of opsin and arrestin by kinesin-II in
33 mammalian photoreceptors. Cell. 102(2):175-87.
Mazelova, J., N. Ransom, L. Astuto-Gribble, M. C. Wilson, and D. Deretic. 2009. Syntaxin 3 and
SNAP-25 pairing, regulated by omega-3 docosahexaenoic acid, controls the delivery of
rhodopsin for the biogenesis of cilia-derived sensory organelles, the rod outer segments.
Journal of Cell Science. 122(12):2003-13.
Morris, A. C., and J. M. Fadool. 2005. Studying rod photoreceptor development in zebrafish.
Physiology and Behavior. 86(3):306-13.
Muppirala, U. K., V. G. Honavar, and D. Dobbs. 2011. Predicting RNA-Protein Interactions Using
Only Sequence Information. BMC Bioinformatics. 12:489.
Nair, U., A. Jotwani, J. Geng, N. Gammoh, D. Richerson, W. L. Yen, J. Griffith, S. Nag, K. Wang, T.
Moss, M. Baba, J. A. McNew, X. Jiang, F. Reggiori, T. J. Melia, and D. J. Klionsky. 2011.
SNARE proteins are required for macroautophagy. Cell. 146(2):290-302.
Nemet, I., G. Tian, and Y. Imanishi. 2014. Submembrane assembly and renewal of rod
photoreceptor cGMP-gated channel: Insight into the actin-dependent process of outer
segment morphogenesis. Journal of Neuroscience. 34(24):8164-74.
Palczewski, K. 2006. G Protein Coupled Receptor Rhodopsin. Annual Review of Biochemistry.
75:743 767.
Pazour, G. J., S. A. Baker, J. A. Deane, D. G. Cole, B. L. Dickert, J. L. Rosenbaum, G. B. Witman,
and J. C. Besharse. 2002. The intraflagellar transport protein, IFT88, is essential for
vertebrate photoreceptor assembly and maintenance. Journal of Cell Biology. 157(1):103-
13.
Pearring, J. N., R. Y. Salinas, S. A. Baker, and V. Y. Arshavsky. 2013. Protein sorting, targeting and
34 trafficking in photoreceptor cells. Progress in Retinal and Eye Research. 36:24-51.
Ramakrishnan, N. A., M. J. Drescher, and D. G. Drescher. 2012. The SNARE complex in neuronal
and sensory cells. Molecular and Cellular Neuroscience. 50(1):58-69.
Rattner, A., P. M. Smallwood, J. Williams, C. Cooke, A. Savchenko, A. Lyubarsky, E. N. Pugh, and
J. Nathans. 2001. A photoreceptor-specific cadherin is essential for the structural integrity
of the outer segment and for photoreceptor survival. Neuron. 32(5):775-86.
Raymond, P. A., L. K. Barthel, and G. A. Curran. 1995. Developmental patterning of rod and cone
photoreceptors in embryonic zebrafish. Journal of Comparative Neurology. 359(4):537-50.
Rizo, J., and T. C. Südhof. 2002. Snares and munc18 in synaptic vesicle fusion. Nature Reviews
Neuroscience. 3(8):641-53.
Röhlich, P. 1975. The sensory cilium of retinal rods is analogous to the transitional zone of
motile cilia. Cell and Tissue Research. 161(3):421-30.
Rozés-Salvador, V., S. O. Siri, M. M. Musri, and C. Conde. 2018. New Player in Endosomal
Trafficking: Differential Roles of Smad Anchor for Receptor Activation (SARA) Protein.
Molecular and Cellular Biology. 38(24):e00446-18.
Salinas, R. Y., J. N. Pearring, J. D. Ding, W. J. Spencer, Y. Hao, and V. Y. Arshavsky. 2017.
Photoreceptor discs form through peripherindependent suppression of ciliary ectosome
release. Journal of Cell Biology. 216(5):1489-1499.
Schmitt, E. A., and J. E. Dowling. 1999. Early retinal development in the Zebrafish, Danio rerio:
Light and electron microscopic analyses. Journal of Comparative Neurology. 404(4):515-36.
Sjöstrand, F. S. 1953. The ultrastructure of the outer segments of rods and cones of the eye as
revealed by the electron microscope. Journal of Cellular and Comparative Physiology.
35 42(1):15-44.
Sotolongo-Lopez, M., K. Alvarez-Delfin, C. J. Saade, D. L. Vera, and J. M. Fadool. 2016. Genetic
Dissection of Dual Roles for the Transcription Factor six7 in Photoreceptor Development
and Patterning in Zebrafish. PLoS Genetics. 12(4):e1005968.
Steinberg, R. H., S. K. Fisher, and D. H. Anderson. 1980. Disc morphogenesis in vertebrate
photoreceptors. Journal of Comparative Neurology. 190(3):501-18.
Stenmark, H., R. Aasland, and P. C. Driscoll. 2002. The phosphatidylinositol 3-phosphate-binding
FYVE finger. FEBS Letters. 513(1):77-84.
Tao-Cheng, J. H., P. E. Gallant, M. W. Brightman, A. Dosemdeci, and T. S. Reese. 2007. Structural
changes at synapses after delayed perfusion fixation in different regions of the mouse
brain. Journal of Comparative Neurology. 501(5):731-40.
Trivedi, D., E. Colin, C. M. Louie, and D. S. Williams. 2012. Live-cell imaging evidence for the
ciliary transport of rod photoreceptor opsin by heterotrimeric kinesin-2. Journal of
Neuroscience. 32(31):10587-93.
Tsukazaki, T., T. A. Chiang, A. F. Davison, L. Attisano, and J. L. Wrana. 1998. SARA, a FYVE
“ TGF Cell. 95(6):779-91.
Vaughan, D. K., and S. K. Fisher. 1989. Cytochalasin D disrupts outer segment disc
morphogenesis in situ in rabbit retina. Investigative Ophthalmology and Visual Science.
30(2):339-42.
Volland, S., L. C. Hughes, C. Kong, B. L. Burgess, K. A. Linberg, G. Luna, Z. H. Zhou, S. K. Fisher,
and D. S. Williams. 2015. Three-dimensional organization of nascent rod outer segment
disk membranes. Proceedings of the National Academy of Sciences of the United States of
36 America. 112(48):14870-14875.
Williams, D. S., K. A. Linberg, D. K. Vaughan, R. N. Fariss, and S. K. Fisher. 1988. Disruption of
microfilament organization and deregulation of disk membrane morphogenesis by
cytochalasin D in rod and cone photoreceptors. Journal of Comparative Neurology.
272(2):161-76.
Young, R. W. 1967. The renewal of photoreceptor cell outer segments. The Journal of cell
biology. 33(1):61-72.
Young, R. W., and D. Bok. 1969. Participation of the retinal pigment epithelium in the rod outer
segment renewal process. The Journal of cell biology. 42(2):392-403.
Zulliger, R., S. M. Conley, M. L. Mwoyosvi, M. W. Stuck, S. Azadi, and M. I. Naash. 2015. SNAREs
interact with retinal degeneration slow and rod outer segment membrane protein-1
during conventional and unconventional outer segment targeting. PLoS ONE.
10(9):e0138508.
37