1. Neuroscience
Download icon

LRRTM1 underlies synaptic convergence in visual thalamus

  1. Aboozar Monavarfeshani
  2. Gail Stanton
  3. Jonathan Van Name
  4. Kaiwen Su
  5. William A Mills III
  6. Kenya Swilling
  7. Alicia Kerr
  8. Natalie A Huebschman
  9. Jianmin Su
  10. Michael A Fox  Is a corresponding author
  1. Virginia Tech Carilion Research Institute, United States
  2. Virginia Tech, United States
  3. Virginia Tech Carilion School of Medicine, United States
  4. Roanoke Valley Governor School, United States
Research Article
  • Cited 0
  • Views 938
  • Annotations
Cite as: eLife 2018;7:e33498 doi: 10.7554/eLife.33498

Abstract

It has long been thought that the mammalian visual system is organized into parallel pathways, with incoming visual signals being parsed in the retina based on feature (e.g. color, contrast and motion) and then transmitted to the brain in unmixed, feature-specific channels. To faithfully convey feature-specific information from retina to cortex, thalamic relay cells must receive inputs from only a small number of functionally similar retinal ganglion cells. However, recent studies challenged this by revealing substantial levels of retinal convergence onto relay cells. Here, we sought to identify mechanisms responsible for the assembly of such convergence. Using an unbiased transcriptomics approach and targeted mutant mice, we discovered a critical role for the synaptic adhesion molecule Leucine Rich Repeat Transmembrane Neuronal 1 (LRRTM1) in the emergence of retinothalamic convergence. Importantly, LRRTM1 mutant mice display impairment in visual behaviors, suggesting a functional role of retinothalamic convergence in vision.

https://doi.org/10.7554/eLife.33498.001

Introduction

Over thirty classes of functionally and morphologically distinct retinal ganglion cells (RGCs) exist in mammals, each responsible for conveying different features of the visual world and each with unique projections to retinorecipient nuclei within the brain (Sanes and Masland, 2015; Martersteck et al., 2017; Baden et al., 2016). As a group, RGCs innervate over 40 retinorecipient brain regions (Morin and Studholme, 2014; Monavarfeshani et al., 2017). However, only a subset of RGCs (~50%) innervate relay cells in the visual thalamus (i.e. the dorsal lateral geniculate nucleus [dLGN]) and provide the principal pathway for image-forming visual information to reach the cerebral cortex (Dhande et al., 2015; Seabrook et al., 2017) (Figure 1A). The recent development of transgenic tools to label these classes of RGCs has revealed that their inputs are segregated into distinct class-specific sublamina within visual thalamus (Huberman et al., 2008; Monavarfeshani et al., 2017; Huberman et al., 2009; Kay et al., 2011; Kim et al., 2008, 2010; Hong and Chen, 2011), supporting the longstanding belief that different features of the visual field are transmitted through the subcortical visual system in parallel, unmixed anatomical channels (Dhande et al., 2015; Cruz-Martín et al., 2014).

Figure 1 with 1 supplement see all
Retinal projections develop into unique terminal types in dLGN.

(A) Schematic of the mouse brain highlighting the main retinorecipient regions including dLGN. (B) Development of VGluT2+ retinal terminals in dLGN and vLGN in wild type mice. (C–G) Cumulative (cum.) distribution of VGluT2+ puncta size in P3 (C), P8 (D), P14 (E), P25 (F) and P62 (G) dLGN (orange) and vLGN (blue). Data are shown as Mean ±SEM. (H) Average total VGluT2+ terminal size in developing dLGN and vLGN. Data are shown as Mean ±SEM, *p<0.0001 by ANOVA. (I) GFP+ retinal terminals in dLGN and superior colliculus (SC) of Mnx1-GFP mice. Red arrowheads highlight gfp-labeled retinal terminals. (J) Cumulative (cum.) distribution of GFP+ puncta size in P25 dLGN (orange) and SC (purple). Data are shown as Mean ±SEM. (K) SBFSEM of retinogeniculate synapses in dLGN of P8 and P14 mice. 3D reconstructions of retinal terminals and relay cell dendrites are depicted below each micrograph. The black arrows denote simple retinogeniculate synapses in P8 and P14 or clusters of retinal terminals originating from multiple RGCs in P14 mouse dLGN. (L) Schematic representation of brainbow-AAV constructs. (M–O) Examples of brainbow-labeled clusters of retinal terminals in dLGN of P8 (M), P10 (N) and P14 (O) Calb2-Cre mice. Arrowheads denote terminals labeled by different colors. Scale bars, 20 µm (B and I), 5 µm (K), 10 µm (M).

https://doi.org/10.7554/eLife.33498.002

In addition to being segregated based on class, retinal projections in dLGN are unique in that they form structurally and functionally distinct synapses compared to their counterparts in other retinorecipient nuclei (Hammer et al., 2014). Retinal terminals in dLGN are prototypic ‘driver’ inputs which are large (compared to adjacent non-retinal inputs) and capable of generating strong excitatory postsynaptic responses in thalamic relay cells. Until recently, it was thought that the level of convergence of retinal inputs onto these relay cells was exceptionally low with only a few (1-5) RGCs innervating each relay cell (Chen and Regehr, 2000; Jaubert-Miazza et al., 2005; Sincich et al., 2007; Hamos et al., 1987; Cleland and Lee, 1985; Cleland et al., 1971; Mastronarde, 1992; Usrey et al., 1999; Yeh et al., 2009; Weyand, 2016; Rathbun et al., 2016, 2010). This low level of retinal convergence allows relay cells to faithfully transfer information from RGCs to visual cortex in an unaltered form, also adding support to the notion that information regarding different features of the visual field flow through the thalamus in parallel channels.

Recently, however, a series of anatomical studies in mice have challenged the concept of feature-specific, parallel visual channels by revealing a level of retinal convergence onto relay cells that is more than an order of magnitude higher than previously described (Hammer et al., 2015; Morgan et al., 2016; Rompani et al., 2017; Howarth et al., 2014). Not only is there a high level of retinogeniculate (RG) convergence in mice, but some relay cells receive input from functionally distinct classes of RGCs (Rompani et al., 2017) raising new questions about the role of thalamus in processing visual information before it reaches visual cortex.

Part of this newly appreciated retinal convergence stems from a set of unique RG synapses (termed complex RG synapses) that contain numerous retinal axons whose terminals aggregate on shared regions of relay cell dendrites (Morgan et al., 2016; Hammer et al., 2015; Lund and Cunningham, 1972). Complex RG synapses have been reported in both rodents and higher mammals (Lund and Cunningham, 1972; Jones and Powell, 1969; So et al., 1985; Campbell and Frost, 1987; Guillery and Scott, 1971; Wilson et al., 1984). Similar to the more classical simple RG synapses (which contain a single retinal terminal on a given portion of a relay cell dendrite), these complex RG synapses are absent from other retinorecipient regions of brain (Hammer et al., 2014) (Figure 1—figure supplement 1). Since branches of dLGN-projecting RGCs also innervate other retinorecipient nuclei (Dhande et al., 2015), we interpret this to suggest that target-derived signals must be generated in dLGN to pattern the unique transformation of retinal axons into simple and complex RG synapses.

In the present study, we sought to identify such target-derived signals. Using next generation sequencing, we discovered that relay cells in dLGN (but not principal neurons in other retinorecipient nuclei) express Leucine Rich Repeat Transmembrane Neuronal 1 (LRRTM1), a known inducer of excitatory synaptogenesis (Linhoff et al., 2009; de Wit et al., 2009). Genetic deletion of LRRTM1 led to a loss of complex RG synapses and thus reduced retinal convergence in visual thalamus. While mutants lacking LRRTM1 and complex RG synapse exhibit normal visual acuity and contrast sensitivity, they display impaired performance in a set of more complex visual tasks that require processing multiple distinct elements of the visual field. Taken together, these results not only identify a novel mechanism underlying the establishment of retinal convergence in visual thalamus, but also importantly provide the first insight into the functional significance of complex RG synapses (and, possibly, retinal convergence) in vision.

Results

Unique transformation of retinal terminals in dLGN coincides with eye-opening

To examine the emergence of the unique morphology of retinal terminals in developing mouse dLGN, two approaches were applied: retinal terminals were either immunolabeled with antibodies against vesicular glutamate transporter 2 (VGluT2, a synaptic vesicular component only present in retinal terminals in visual thalamus) (Hammer et al., 2014; Land et al., 2004), or were anterogradely labeled by intraocular injection of fluorescent-conjugated Cholera Toxin B (CTB) (Muscat et al., 2003) (Figure 1B–H; Figure 1—figure supplement 1). Shortly after their initial formation (P3-P8), VGluT2- or CTB-labeled terminals appeared similar in size and morphology in dLGN and the adjacent retinorecipient ventral lateral geniculate nucleus (vLGN). However, by eye-opening (P12-P14), terminals in dLGN underwent significant enlargement compared to those in vLGN and other (Hammer et al., 2014) retinorecipient nuclei (Figure 1B–H; Figure 1—figure supplement 1). The unique developmental transformation of retinal terminals in dLGN at eye-opening (rather than at their initial formation), suggested that this was not the result of purely cell intrinsic mechanisms in dLGN-projecting classes of RGCs.

To test this hypothesis, we assessed retinal terminals generated by a single class of ON-OFF direction-selective RGCs whose axons branch to innervate both dLGN and superior colliculus (SC) (Dhande et al., 2015; Kim et al., 2010). This class of RGC is specifically labeled in Mnx1-GFP mice (also called Hb9-GFP)(Trenholm et al., 2011). Despite originating from branches of individual retinal axons, those terminals present in dLGN were dramatically larger than those in SC (0.83 ± 0.1 µm2 in dLGN vs 0.36 ± 0.01 µm in SC, p<0.01 by t-test, n = 3, Figure 1I,J). These data suggest that target-derived cues are generated in dLGN around the time of eye-opening to pattern the transformation of retinal terminals.

The approaches described above do not provide the resolution required to differentiate simple and complex RG synapses, therefore, we used serial block-face scanning electron microscopy (SBFSEM) to identify whether both simple and complex RG synapses emerged at eye-opening. While SBFSEM ultrastructural analysis revealed the presence of both simple and complex RG synapses shortly after eye-opening (P14) we were only able to identify simple-like RG synapses prior to eye-opening (at P8) (Figure 1K). Moreover, we delivered brainbow AAVs (Figure 1L) intraocularly in newborn Calb2-Cre mice (in which a large proportion of RGCs express Cre recombinase) to generate multi-colored RGCs and assess the development of complex RG synapses (Hammer et al., 2015). Similar to SBFSEM analysis, brainbow AAV-labeling failed to detect clusters of retinal terminals at P8 (Figure 1M), but clearly revealed clusters of retinal terminals originating from distinct RGCs as early as P10 and P14 (Figure 1N,O). Thus, around eye-opening, dLGN-specific molecular mechanisms must emerge to induce the unique transformation of both simple and complex RG synapses.

Identification of target-derived synaptic organizing molecules in dLGN

To identify target-derived synaptic organizers present at eye-opening in dLGN (but not other retinorecipient regions), we performed next-generation transcriptome analysis of developing mouse visual thalamus (Figure 2A; Figure 2—source data 1). We assessed four different developmental time points, two before eye-opening (P3 and P8), and two at (P12) or after (P25) eye-opening (Figure 1). Comparing gene expression profiles in both dLGN and vLGN revealed hundreds of differentially and developmentally expressed mRNAs (Figure 2B). We focused our attention on a small subset of genes that were significantly enriched in dLGN (compared to vLGN) and whose highest expression coincided with eye-opening and the emergence of simple and complex RG synapses. Two genes with well-established roles in inducing excitatory synapses fit those criteria: Lrrtm1 and neuritin 1 (Nrn1) (Figure 2C–E) (Linhoff et al., 2009; Javaherian and Cline, 2005). We confirmed the enrichment of these genes at eye-opening in dLGN (but not vLGN) by qPCR, in situ hybridization and western blot (Figures 2F–I and 3A–D; Figure 3—figure supplement 1A–C). While the increase in expression of these genes coincides with eye-opening and the onset of experience-dependent transformation of retinal nerve terminals, we did not find a decrease in their expression in the absence of visual inputs (Figure 3—figure supplement 1F). In addition to their low expression level in vLGN, it is important to point out that Lrrtm1 and Nrn1 mRNAs were either absent or only weakly expressed in other retinorecipient nuclei, such as the SC and suprachiasmatic nucleus (SCN) (Figure 3E; Figure 3—figure supplement 1D,E). There were, however, significant differences in the distribution of Lrrtm1 and Nrn1 mRNAs in other regions of the visual system. Lrrtm1 was not generated by RGCs (although it was expressed in the INL) or by many cells in primary visual cortex (vCTX) (Figure 3E,F), whereas Nrn1 was robustly expressed by both RGCs and by cells in vCTX (Figure 3E,F) (see also Fujino et al., 2008; Nedivi et al., 1998).

Figure 2 with 1 supplement see all
Identification of Lrrtm1 and Nrn1 as candidate synaptic organizing cues in dLGN.

(A) Next generation RNAseq was performed on RNA isolated from dLGN and vLGN at P3, P8, P12 and P25. (B) Volcano scatter plots show differentially expressed mRNAs in the developing dLGN. (C) Relative Lrrtm1 and Nrn1 mRNA levels in dLGN at P8, P12 and P25 compared to P3 by RNAseq. Data are relative values comparing different ages, *p<0.0001 by Wald Chi-Squared Test (DESeq2). (D and E) Enrichment of Lrrtm1 (D) and Nrn1 (E) mRNAs in dLGN compared to vLGN at four ages in wild type mice. Data are relative values comparing dLGN and vLGN, *p<0.0001 by Wald Chi-Squared Test (DESeq2). (F and G) Developmental expression of Lrrtm1 (F) and Nrn1 (G) mRNAs in wild type dLGN by qPCR. Data are shown as Mean ±SEM; ***p<0.0001, **p<0.01, *p<0.05 by ANOVA. (H and I) ISH for Lrrtm1 (F) and Nrn1 (G) mRNAs in coronal sections of wild type P25 mouse brains. Scale bar, 1 mm (H and I).

https://doi.org/10.7554/eLife.33498.004
Figure 3 with 1 supplement see all
Developmental and region-specific expression of Lrrtm1 and Nrn1.

(A) ISH for Lrrtm1 and Nrn1 mRNAs in the developing visual thalamus. dLGN encircled by red dots. (B–D) Western blots show LRRTM1 protein level is higher in dLGN than vLGN (B) and increases in the dLGN postnatally (C and D). Data are shown as Mean ±SEM; ***p<0.0001, **p<0.01, *p<0.05 by ANOVA. (E) Expression of Lrrtm1 and Nrn1 mRNAs in coronal sections of P25 mouse brains. Boxes in visual cortex (vCTX) and superior colliculus (SC) are shown in higher magnifications on the right. (F) Expression of Lrrtm1 and Nrn1 mRNAs in P25 retina. Yellow arrowheads denote mRNA expression. ONL, outer nuclear layer; INL, inter nuclear layer; GCL, ganglion cell layer. Scale bars, 200 µm (A), 500 µm (E), 50 µm (insets of vCTX and SC and F).

https://doi.org/10.7554/eLife.33498.007

Finally, we sought to address the cell-specific expression of Lrrtm1 and Nrn1 in visual thalamus. Since Lrrtm1 encodes a transmembrane cell adhesion molecule and Nrn1 encodes a GPI-linked membrane associated extracellular molecule (Naeve et al., 1997; Linhoff et al., 2009), we hypothesized that dLGN relay cells must generate these molecules for them to act post-synaptically at RG synapses. We combined in situ hybridization analysis using riboprobes against these two genes with molecular and genetic approaches to label different cell types in dLGN. First, we demonstrated that both Lrrtm1 and Nrn1 mRNAs are generated by neurons and not glia since they were co-expressed by Syt1-expressing neurons (93% of Syt1+ neurons co-expressed Lrrtm1; 98% of Syt1+ neurons co-expressed Nrn1; Figure 4A,B; Figure 3—figure supplement 1G,H) but not by IBA1-expressing microglia or GFP-labeled astrocytes in Aldh1l1-GFP mice (Figure 4C–F). Next, to differentiate which types of neurons generate these synaptogenic cues, we assessed Lrrtm1 and Nrn1 mRNA expression in glutamate decarboxylase (GAD67)-expressing inhibitory interneurons and in Crh-Cre::tdt transgenic mice in which excitatory thalamocortical relay cells are fluorescently labeled (Taniguchi et al., 2011). Results revealed Lrrtm1 and Nrn1 are exclusively produced by dLGN relay cells (100% of Crh-Cre::tdt+ cells co-expressed Lrrtm1 and Nrn1; 0% and 2% of Gad1+ cells co-expressed Lrrtm1 and Nrn1, respectively; Figure 4G–J; Figure 3—figure supplement 1I,J). Based on their developmental and cell-specific expression, these molecules therefore represented prime candidates to influence the development of simple and complex RG synapses.

dLGN relay cells generate Lrrtm1 and Nrn1.

(A and B) Double in situ hybridization (ISH) for Syt1 and either Lrrtm1 (A) or Nrn1 (B) in P14 wild type dLGN. (C and D) ISH for either Lrrtm1 (C) or Nrn1 (D) in dLGN of P14 Aldh1l1-GFP mice revealed no astrocytic expression of these mRNAs. (E and F) ISH for Lrrtm1 (E) or Nrn1 (F) and immunostaining (IHC) for the microglia marker IBA1 in P14 wild type dLGN. (G and H) ISH for Lrrtm1 (G) or Nrn1 (H) and IHC for GAD67 in P14 (Lrrtm1) and P25 (Nrn1) wild type mice revealed no mRNA expression by inhibitory interneurons. Red arrows depict GAD67+ interneurons. (I and J) Double ISH for either Lrrtm1 (I) or Nrn1 (J) and Gad1 in P25 Crh-Cre::tdt dLGN revealed mRNA expression by relay cells. Scale bar, 20 µm (A–J).

https://doi.org/10.7554/eLife.33498.009

LRRTM1 is required for the development of complex RG synapses

Previous studies have reported roles for both LRRTM1 and NRN1 (also called Candidate Plasticity Gene 15, CPG15) in inducing the formation and maturation of excitatory synapses (Cantallops et al., 2000; Ko et al., 2011; Linhoff et al., 2009; Nedivi et al., 1998; Soler-Llavina et al., 2011). In addition, NRN1 contributes to the development and maturation of retinal arbors (Cantallops et al., 2000). To test whether these molecules are necessary for the development of retinal terminals, we assessed the morphology of retinal terminals in dLGN of mice lacking LRRTM1 (Linhoff et al., 2009) or NRN1 (Fujino et al., 2011) using VGluT2 immunostaining and CTB anterograde labeling. These studies revealed a significant decrease in the number of large VGluT2+ and CTB+ puncta in dLGN of Lrrtm1−/− mice at and after eye-opening (Figure 5A–C; Figure 5—figure supplement 1A,B,E), suggesting a role for this molecule in the maturation and/or refinement of RG circuitry. Retinal terminals in neonatal dLGN (i.e. before eye-opening) or in vLGN were not affected by the loss of LRRTM1 (Figure 5A–F; Figure 5—figure supplement 1A–F). Since retinal projections account for only a small proportion (5–10%) of all projections innervating relay cells residing in dLGN (Monavarfeshani et al., 2017), we also assessed whether the loss of LRRTM1 altered other types of terminals in dLGN. None of the non-retinal inputs examined appeared affected in Lrrtm1−/− mutant mice (Figure 5—figure supplement 1G–I). Similar analysis in Nrn1−/− mutants failed to identify developmental deficits in the density, size or distribution of retinal terminal in dLGN (Figure 5—figure supplement 2A,B).

Figure 5 with 2 supplements see all
Loss of Lrrtm1 results in smaller vglut2-positive puncta in dLGN but not vLGN.

(A and D) Immunostaining of VGluT2+ retinal terminals in dLGN (A) and vLGN (D) of littermate control and Lrrtm1−/− mice at P3, P8, P14 and P25 and adult mice (P60-85). Red boxes in P25 panels are shown in higher magnification on the right. (B and E) Cumulative (cum.) distribution of VGluT2+ puncta size in control and Lrrtm1−/− mice dLGN (B) and vLGN (E). (C and F) Average VGluT2+ puncta size in control and Lrrtm1−/− mice dLGN (C) and vLGN (F). Data represent Mean ±SEM; **p=0.0005, *p<0.05 by ANOVA. Scale bar, 20 µm (A and C), 10 µm (insets).

https://doi.org/10.7554/eLife.33498.010

As described earlier, an important limitation of these techniques is that they cannot differentiate simple or complex RG synapses. It was therefore unclear whether individual retinal terminals were smaller in Lrrtm1−/− mice, or clusters of retinal terminals were absent in these mutants. To overcome this technical limitation, we employed both SBFSEM and multicolor brainbow-AAV labeling of retinal axons in dLGN of Lrrtm1−/− mutant and control mice. In SBFSEM, retinal terminals were distinguished from all other synaptic terminals by their round vesicles and pale mitochondria (Rafols and Valverde, 1973; Hammer et al., 2014; Bickford et al., 2010) and were traced throughout the entire volume of the imaged tissues. In total, 534 retinal terminals were analyzed in Lrrtm1−/− mice and 646 in controls (n = 3 mice per genotypes). While the majority (90%) of RG synapses were classified as complex in controls, only 37% of retinal terminals fell into this category in Lrrtm1−/− mutant dLGN (Figure 6A–C; Figure 6—figure supplement 1A,B). Similarly, analysis of multicolor-labeled retinal terminals by brainbow AAVs showed fewer and smaller clustered retinal terminals in dLGN of Lrrtm1−/− mice (Figure 6G–I). The majority of retinal terminal clusters in mutants (82.6%) contained less than four distinct inputs (identified by their unique colors). In contrast, the majority (80%) of clusters in controls contained more than four distinct retinal terminals.

Figure 6 with 1 supplement see all
Loss of complex RG synapses in Lrrtm1−/− mice.

(A and B) SBFSEM images of retinal terminals in P42 control (A, labeled in green) and Lrrtm1−/− (B, labeled in red) dLGN. RG synapses are depicted in insets a1-a4 (Ctl) and b1-b4 (Lrrtm1−/−). In insets, each retinal terminal is depicted in a unique color. Similar colors in different insets do not represent axonal branches of the same RGC. 3D reconstruction of retinal terminals and relay cell dendrite are depicted on the right. The black arrows denote the position of dendrites stemming from relay cells somas. (C) Percentage of retinal terminals contributing to forming RG synapses with 1, 2, 3 or ≥4 distinct retinal terminals in P42 Lrrtm1−/− and control dLGN. Data represent Mean ±SEM; ***p<0.0001, by ANOVA. (D–F) Quantification of terminal size (D), actual number of active zones per terminal (E) and the active zones: terminal area ratio for simple and complex RG synapses in dLGN of Lrrtm1−/− and control mice (NT = nerve terminal). Data represent Mean ±SEM; ***p<0.0001, by ANOVA. (G and H) Retinal terminals were multicolor-labeled by injecting 1–2 µl brainbow AAVs into the vitreous humor of Lrrtm1−/−::Calb2-Cre and control mice. (I) Color analysis of clustered retinal terminals in wild type and Lrrtm1−/− mutants revealed a lower level of complex RG synapses in mutants. Scale bar, 10 µm (A and the 3D image), 20 µm (G), 1 µm (insets).

https://doi.org/10.7554/eLife.33498.013

Thus, there was a significant loss of complex RG synapse in the absence of LRRTM1. In fact, these numbers underrepresent the loss of retinal convergence in mutants, since our criteria for defining a complex RG synapse requires the presence of just two distinct retinal inputs. Not only was there a significant loss of complex RG synapses in mutants, but those complex synapses that remained contained significantly fewer retinal terminals. In control dLGN about 86% of complex RG synapses contained between 4–14 retinal terminals, whereas the majority of complex RG synapses in Lrrtm1−/− mutants contained only 2 or 3 inputs (Figure 6C). While the reduced number of complex RG synapses (and retinal inputs at the few persisting complex RG synapses) might be caused by fewer retinal axons in mutants, we failed to observe a significant loss of RGC axons in the optic nerves of Lrrtm1−/− mice (Figure 6—figure supplement 1).

Surprisingly, we also observed an increase in individual retinal terminal size in both simple and complex RG synapses in Lrrtm1−/− mice (Figure 6D). This increase in terminal size in Lrrtm1−/− mutant dLGN was accompanied by a significant increase in active zone number compared to control RG synapses (Figure 6E). However, when we normalized active zone number to terminal size there was no difference in the density of active zones in control and mutant RG synapses (Figure 6F).

Impaired visual behaviors in mice lacking LRRTM1

The functional consequence of LRRTM1 deletion and the loss of complex RG synapses was assessed by a two-alternative forced swim test (Prusky et al., 2000; Wong and Brown, 2006; Huberman and Niell, 2011). In this test, mice learn to associate a visual cue with a hidden platform that allows them to escape the water (Figure 7A). In order to confirm the necessity of vision for performing this task we asked whether Atoh7−/− mice (also called Math5−/−), which are genetically blind (Wang et al., 2001), can detect the positive visual cue and find the hidden platform. Atoh7−/− mice were unable to perform this task, demonstrating the importance of vision in this assay (Figure 7—figure supplement 1).

Figure 7 with 2 supplements see all
Complex RG synapses are required for visual behaviors.

(A) Schematic diagrams depicting the two-alternative forced swim behavior task. A’ depicts a mouse’s view of the visual displays (e.g. vertical gratings). (B and C) Training Lrrtm1−/− (red) and control (green) mice to detect a vertical grating display versus a gray screen (B, n = 20 for Ctl; and 21 for Lrrtm1−/−) or horizontal grating display (C, n = 19 for Ctl; and 18 for Lrrtm1−/−). Examples of visual displays are depicted in B’ and C’. (D and E) Percentage of correct choices made by Lrrtm1−/− (red) and control (green) mice for detection of vertical gratings with increasing spatial frequency (D; cpd, cycle per degree, n = 20 for Ctl; and 21 for Lrrtm1−/−) or decreasing contrast versus a gray screen (E, n = 10 for Ctl; and nine for Lrrtm1−/−). Examples of visual displays are depicted in D’ and E’. (F and G) Correct choices made by Lrrtm1−/− (red) and control (green) mice for discriminating vertical grating from horizontal grating either with increasing spatial frequency (F, n = 10 for Ctl; and 10 for Lrrtm1−/−) or with decreasing contrast (G, n = 10 for Ctl; and nine for Lrrtm1−/−). D1-D7 are the consecutive days of the test phase. Examples of visual displays are depicted in F’ and G’. For B-G, dashed line represents the 70% correct threshold for successful completion of task. All data are shown as Mean ±SEM; ***p<0.0001, **p<0.01, *p<0.05 by ANOVA.

https://doi.org/10.7554/eLife.33498.015

To explore the role of LRRTM1 (and complex RG synapses) in vision, mice were trained for 8 days to detect a vertical grating (0.17 cycle per degree, cpd) on S+ monitor positioned above the hidden platform, compared with a gray screen or a horizontal grating display on the S- monitor (Figure 7A). Mice that exceeded 70% accuracy in locating the hidden platform were considered capable of detecting and discriminating the visual cues (Prusky et al., 2000). Lrrtm1−/− mutants and controls displayed equal abilities to distinguish the vertical gratings (i.e. the positive visual cue) from a gray screen or horizontal gratings at the end of training, although the initial learning phase of Lrrtm1−/− mice was modestly impaired (Figure 7B,C), which is in agreement with previous findings showing a delayed response of Lrrtm1−/− mutant mice to new environment (Takashima et al., 2011). Although we used vertical gratings as the positive cue in all subsequent experiments, we tested whether controls or mutants could also learn these tasks if horizontal gratings were the positive cue. We found no difference in mutants or controls learning to associate the hidden platform under a screen with horizontal gratings (Figure 7—figure supplement 1B). Moreover, in order to demonstrate that control or mutant mice were not capable of detecting the hidden platform itself (instead of associating it to the visual cue), we trained control and mutant mice (for 8 days) to associate the positive visual cue (S+) (with near 100% accuracy) with the platform. We then moved the platform below the negative visual cue (S-) and tested each mouse for 10 trials in day 9. Mutants and controls swam toward the positive visual cue that lacked the rescue platform, confirming they could not visually detect the hidden platform (Figure 7—figure supplement 1C).

By changing the frequency of the vertical bars, we next tested visual acuity in Lrrtm1−/− mutants. Results indicate that acuity was similar between mutants and controls, both falling below the 70% correct criteria at spatial frequencies above 0.57 cpd (Figure 7D). There was a statistically significant difference between Lrrtm1−/− and control mice at a single spatial frequency (0.62 cpd), however, at this frequency both performed poorly in the task (Figure 7D). Next, we altered the contrast of the vertical grating bars rather than the spatial frequency or orientation. Similarly, Lrrtm1−/− mutants failed to show significant differences compared to wild type mice (Figure 7E). Taken together these results suggest Lrrtm1−/− mutants do not exhibit deficits in visual acuity, simple pattern recognition or contrast sensitivity.

Next, we exposed mice to more complex visual tasks in which multiple features of the visual scene were altered at once. For example, we challenged mice to differentiate vertical and horizontal gratings while increasing the spatial frequency. In this more complex task, we found Lrrtm1−/− mutants performed significantly worse than controls (Figure 7—figure supplement 1D). Since there are conflicting data indicating spatial memory deficits in mice lacking LRRTM1 (Voikar et al., 2013; Takashima et al., 2011), we next adjusted our experimental design to rule out the influence of spatial memory impairment on performing the visual tasks. After 8 days of training with a new cohort of mutant and control mice, pattern discrimination was again tested while also increasing the spatial frequency or decreasing the contrast of both vertical and horizontal gratings. After each day of testing, we checked the ability of mice to perform the initial, standard discrimination task (i.e. to discriminating vertical and horizontal grating with 0.17 cpd and 100% contrast). Throughout the experiments, Lrrtm1−/− mice failed to show any signs of memory deficits in this task. Interestingly, while control mice were able to discriminate vertical and horizontal bars at a spatial frequency of 0.32 cpd, Lrrtm1−/− mice performance dropped significantly under the 70% threshold during these more complex tasks (Figure 7F). Similarly, at lower contrast (i.e. 25% and 10%) mutant mice lacked the sensitivity to discriminate vertical and horizontal grating patterns (Figure 7G). As another set of controls, we repeated these behavioral tasks with Nrn1−/− mutants and found no deficit in their performance in either the simple or complex visual tasks (Figure 7—figure supplement 1E–J). Taken together, these results indicate that mice lacking LRRTM1 and complex RG synapses exhibit abnormalities in performing more complex visual tasks.

Disscussion

The recent discovery by three independent groups (Hammer et al., 2015; Rompani et al., 2017; Morgan et al., 2016) that a shockingly high level of retinal convergence exists onto thalamic relay cells in rodents has raised a number of questions for the field. Are these converging inputs functional? Is retinal convergence in visual thalamus important for vision? What are the developmental mechanisms that underlie the establishment of retinal convergence in dLGN? We are only beginning to scratch the surface in answering these questions. For example, Litvina and Chen (2017) only recently applied an optogenetic approach to demonstrate a higher level of functional retinal convergence onto dLGN relay cells. Here, we sought to address the last question mentioned above, what mechanisms underlie the establishment of retinal convergence in visual thalamus. Using an unbiased screen, we identified LRRTM1 as a target-derived cue necessary for the formation of retinal convergence onto dLGN relay cells. Analysis in LRRTM1-deficient mice revealed that the lack of this synaptic adhesion molecule led to impaired visual function. We interpret these results to suggest that complex RG synapses are necessary for visual processing. It is important to point out that we also observed an increase in the retinal never terminal size (and active zone number) in the absence of LRRTM1 and this may also contribute to deficits in visual behaviors.

LRRTM1 as a target-derived synaptic organizer in visual thalamus

LRRTMs are transmembrane proteins that act as transsynaptic signals to trigger excitatory synaptogenesis (Linhoff et al., 2009; de Wit et al., 2009; de Wit and Ghosh, 2014; Um et al., 2016). When present in the postsynaptic membrane, LRRTM1 binds to the extracellular domain of neurexins to induce presynaptic differentiation in contacting axons (Linhoff et al., 2009; Siddiqui et al., 2010). In visual thalamus, LRRTM1 is specifically expressed by relay cells (and not other cells) and its transsynaptic partners, neurexins, are generated by RGCs (Figure 2—figure supplement 1) (Sajgo et al., 2017; Shigeoka et al., 2016). Therefore, based on our results, we hypothesized that LRRTM1-neurexin interactions are critical for the formation of complex RG synapses. Although the necessity of neurexins in retinogeniculate connectivity has yet to be thoroughly examined, the loss of CASK, a MAGUK protein necessary for trafficking neurexins to the presynaptic membrane, leads to abnormal retinogeniculate connectivity and optic nerve hypoplasia (LaConte et al., 2016; Srivastava et al., 2016; Moog et al., 2011; Liang et al., 2017).

It is important to point out that neurexins have other postsynaptic partners expressed in visual thalamus, including neuroligins and other LRRTMs (Figure 2—figure supplement 1B–G) (Laurén et al., 2003; Varoqueaux et al., 2006), each capable of inducing excitatory synaptogenesis elsewhere in the brain or in vitro (Fox and Umemori, 2006; Craig et al., 2006; Ko et al., 2009; Siddiqui et al., 2010). The presence of LRRTM1, other LRRTMs, and neuroligins in dLGN raises an interesting possibility that simple and complex RG synapses may be assembled through different postsynaptic interactions with neurexins. As such, astrocytes may also contribute to the signals that regulate simple or complex RG synapses. Retinal terminals in simple RG synapse are ensheathed by astrocytic processes (Hammer et al., 2014; Bickford et al., 2010), and these astrocytes are known to produce extracellular factors capable of bridging neurexin–neuroligin interactions to facilitate excitatory synaptogenesis (Kucukdereli et al., 2011; Singh et al., 2016).

The presence of multiple postsynaptic neurexins partners in dLGN also raises the possibility that their abundance (or overabundance) may prevent some level of activity-dependent refinement in dLGN. Overexpression of different combinations of neurexin-binding partners in postsynaptic neurons has been shown to impair synapse elimination in vitro (Ko et al., 2011). The emergence of complex RG synapses at eye-opening may therefore represent synapses with an overabundance of neurexin-binding receptors, in which strong trans-synaptic adhesion prevent complete activity-dependent RG refinement. While certainly possible, we see this as unlikely given the dramatic refinement of retinal arbors around eye-opening in mice, and since this RG refinement itself gives rise to retinal bouton clustering (Dhande et al., 2011; Hong et al., 2014).

Retinal convergence: artifact or by design?

The discovery of an extraordinary level of retinal convergence on mouse relay cells has left the field pondering whether such convergence is an artifact of impaired refinement (as described above) or whether there is functional significance to such ‘fuzzy’ connectivity (as one group has termed this retinogeniculate convergence) (Morgan et al., 2016). It is easy to discount the importance of retinal convergence onto relay cells and the role that complex RG synapses may play in vision, since many groups (including our own [Hammer et al., 2014]) have demonstrated that relay cells receive a very small number of strong, functional inputs from the retina (Chen and Regehr, 2000; Jaubert-Miazza et al., 2005; Hooks and Chen, 2006; Litvina and Chen, 2017). Many of the techniques used to identify high levels of retinogeniculate convergence in mice have been anatomical in nature (e.g. ultrastructural analysis, anterograde multicolor labeling of RGCs and retrograde trans-synaptic tracing) (Hammer et al., 2015; Morgan et al., 2016; Rompani et al., 2017), leading to the possibility that ‘form’ does not fit ‘function’ in mouse visual thalamus. Recent optogenetic analysis of the RG circuit in mice has revealed a substantially higher level of functional retinal convergence on relay cells, however the strength of these inputs widely varies (Litvina and Chen, 2017). Functional roles for weak RG synapses remain unclear.

In the present study, we took advantage of the loss of complex RG synapses in Lrrtm1−/− mice to begin to shed light on the functional significance of retinal convergence on thalamic relay cells. While the ability of Lrrtm1−/− mice to perform tasks with simple visual cues appeared unaltered compared with controls, they performed poorly on tasks where more than one feature of the visual scenes was altered at once. Although these mice lack LRRTM1 globally, such deficits are likely the direct result of impaired RG circuitry for several reasons. First, LRRTM1 is largely absent from retina and visual cortex (Figure 3E–F), sites whose function are required for the performance of these visual tasks. Second, global deletion of LRRTM1 failed to result in synaptic or cytoarchitectural changes in other brain regions that process visual information (Figure 7—figure supplement 2A–F). For these reasons, we believe that results presented here provide the first clues that complex RG synapses (and retinal convergence) are not functionally insignificant artifacts of impaired or incomplete activity-dependent refinement, but rather are an important component of processing and relaying visual information from the retina to visual cortex.

Materials and methods

Key resources table
Reagent type (species)
or resource
DesignationSource or referenceIdentifiers
Virus, AAV9AAV9.hEF1a.lox.TagBFP.lox.eYFP.lox.
WPRE.hGH-InvBYF
Penn Vector CoreAV-9-PV2453
Virus, AAV9AAV9.hEF1a.lox.mCherry.lox.mTFP1.
lox.WPRE.hGH-InvCheTF
Penn Vector CoreAV-9-PV2454
Mouse (C57/Bl)Lrrtm1−/− (Lrrtm1tm1Lex)MMRRCRRID:MMRRC_031619-UCD
Mouse (C57/Bl)Atoh7tm1Gla (Atoh7−/−)MMRRCRRID:MMRRC_042298-UCD
Mouse (C57/Bl)Aldh1l1-EGFPMMRRCRRID:MMRRC_011015-UCD
Mouse (C57/Bl)Nrn1−/− (Nrn1tm1.2Ndiv)The Jackson LaboratoryRRID:IMSR_JAX:018402
Mouse (C57/Bl)Rosa-stop-tdTThe Jackson LaboratoryRRID:IMSR_JAX:007905
Mouse (C57/Bl)Crh-CreMMRRCRRID:MMRRC_030850-UCD
Mouse (C57/Bl)Calb2-CreThe Jackson Laboratory10774
antibodymouse anti-GAD67EMD MilliporeMAB5406, RRID:AB_2278725
antibodymouse anti-actinEMD MilliporeMAB1501, RRID:AB_2223041
antibodyrabbit anti-IBA1Wako019–19741, RRID:AB_839504
antibodyrabbit anti-mGluR1aFrontier Institute co.,ltdRRID:AB_2571799
antibodyrabbit anti-VGluT1Synaptic Systems135402, RRID:AB_2187539
antibodyrabbit anti-VGluT2Synaptic Systems135511
antibodysheep anti-LRRTM1Synaptic SystemsAF4897, RRID:AB_10643427
antibodysheep (POD)-conjugated anti-DIGRoche11426346910
antibodysheep (POD)-conjugated anti-FLRoche11207733910
commercial assay or kitPrepX PolyA mRNA Isolation KitWafergen400047
commercial assay or kitPrepX RNA-Seq for Illumina
Library Kit, 48 samples
Wafergen400046
commercial assay or kitQuant-iT dsDNA HS KitInvitrogenQ33120
commercial assay or kitSuperscript II Reverse Transcriptase
First Strand cDNA Synthesis kit
Invitrogen18064014
commercial assay or kitAurum Total RNA Fatty and
Fibrous Tissue kit
BioRad7326870
commercial assay or kitpGEM-T Easy Vector SystemsPromegaA1360
commercial assay or kitAmbion MAXIscript T7 In Vitro
Transcription Kit
Thermo Fisher ScientificAM1312
commercial assay or kit
commercial assay or kitiTaq SYBRGreen SupermixBioRad1725124
commercial assay or kitTyramide Signal Amplification
(TSA) systems
PerkinElmerNEL75300 1KT
commercial assay or kitAmersham ECL Prime Western
Blotting Detection Reagent
GE Healthcare Life SciencesRPN2236
commercial assay or kitTruSeq PE Cluster
Kit v3-cBOT-HS
illuminaPE-401–3001
commercial assay or kitTruSeq SBS Kit v3-HS
(200-cycles)
illuminaFC-401–3001
chemical compound, drugAlexa-conjugated cholera
toxin beta subunit
Thermo Fisher ScientificC22841
chemical compound, drugFluorescein RNA
Labeling Mix
Roche11685619910
chemical compound, drugDIG RNA Labeling MixRoche11277073910
chemical compound, drugProteinase KThermo Fisher ScientificEO0491
chemical compound, drugParaformaldehyde,
EM grade
EMS19202
chemical compound, drugSodium cadodylateEMS12300
chemical compound, drugTissue Freezing MediumEMS72592
chemical compound, drugGlutaraldehydeEMS16220
chemical compound, drugPrehybridization SolutionSigmaP1415
chemical compound, drugHeparin SodiumFisher ScientificBP2425
chemical compound, drugYeast RNARoche10109223001
chemical compound, drugBlocking reagent (ISH)Roche11096176001
software, algorithmTrakEM2ImageJ pluginRRID:SCR_008954
software, algorithmFijiFijiRRID:SCR_002285
software, algorithmImageJNIHRRID: SCR_003070
software, algorithmPrismGraphPadRRID: SCR_002798
software, algorithmGabor-patch generatorhttps://www.cogsci.nl/N/A
sequence-based reagentCloning primers for Gad1:
F:TGTGCCCAAACTGGTCCT;
R:TGGCCGATGATTCTGGTT
Integrated DNA TechnologiesN/A
sequence-based reagentqPCR primer for Nrn1:
F:TTCCCCCGCGTTCTCTAAAC;
R:GCCTGCACCAGGTAAGCTAT
Integrated DNA TechnologiesN/A
sequence-based reagentqPCR primer for Lrrtm1:
F:AGCAGCTGAATGGAGGTTGTC;
R:AGTGTAGACAGAGGCCGAGTAG
Integrated DNA TechnologiesN/A
sequence-based reagentqPCR primer for Gapdh:
F:CGTCCCGTAGACAAA ATG GT;
R:TTGATG GCAACAATC TCCAC
Integrated DNA TechnologiesN/A
recombinant DNA reagentNrn1 (5367281)DharmaconMMM1013-202769896
recombinant DNA reagentLrrtm1 (5321979)DharmaconMMM1013-202769075
recombinant DNA reagentSyt1 (5363062)DharmaconMMM1013-202709704

Animals

CD1 and C57/BL6 mice were obtained from Charles River (Wilmington, MA) or Harlan (Indianapolis, IN). Lrrtm1−/− mice were obtained from MMRRC (stock # 031619-UCD), Nrn1−/− (stock # 018402), Calb2-Cre (stock # 010774) and Rosa-stop-tdT mice (stock # 007905) were all obtained from Jackson Laboratory. Crh-Cre (stock # 030850-UCD) and Aldh1l1-EGFP (stock # 011015-UCD) mice were obtained from W. Guido (University of Louisville) and S. Robel (Virginia Tech), respectively. Atoh7−/− (stock# 042298-UCD) were obtained from S. W. Wang and were described previously (Wang et al., 2001). Mice were housed in a 12 hr dark/light cycle and had ad libitum access to food and water. Late dark-reared (LDR) mice were placed in a light-tight room from P20-P31. Dissections of LDR mice were performed at P31 in a dark room under red light. All experiments were performed in compliance with National Institutes of Health (NIH) guidelines and protocols and were approved by the Institutional Animal Care and Use Committee (IACUC# 15-137VTCRI, 15-167VTCR and 15-174VTCRI) and Institutional Biosafety Committee (IBC# 15–038) at Virginia Tech.

Immunohistochemistry (IHC)

Anesthetized mice were transcardially perfused with phosphate-buffered saline (PBS; pH 7.4) and 4% paraformaldehyde in PBS (PFA; pH 7.4). Dissected brains and eyes were post-fixed in 4% PFA for 12–16 hr at 4°C. Tissues were cryopreserved in 30% sucrose solution for 2–3 days, embedded in Tissue Freezing Medium (Electron Microscopy Sciences, Hatfield, PA), and cryosectioned (16 μm sections). Sections were air-dried onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA) and frozen at −80°C until further processing. For IHC, slides were incubated in blocking buffer (2.5% bovine serum albumin, 5% Normal Goat Serum, 0.1% Triton-X in PBS) for 1 hr. Primary antibodies were diluted in blocking buffer as follows: GAD67 (Millipore MAB5406) 1:700; IBA1 (Wako 019–19741) 1:1000; VGluT2 (Synaptic Systems 135511) 1:500; VGluT1 (Synaptic Systems 135402) 1:700; mGluR1a (Frontier Institute co. AB_2571799) 1:250 and incubated on tissue sections for >12 hr at 4°C. After washing three times in PBS, fluorescently conjugated secondary antibodies (1:1000 in blocking buffer) were incubated on sections for 1 hr at room temperature. After five washes with PBS, sections were stained with DAPI (1:5000 in water) and were mounted with Vectashield (Vector Laboratories, Burlingame, CA). Images were acquired on a Zeiss LSM 700 confocal microscope. When comparing sections from different age groups or genotypes, images were acquired with identical parameters. A minimum of three animals (per genotype and per age) were compared in all IHC experiments.

Riboprobe production

pCMV-SPORT6 Plasmids carrying Syt1 (cat # 5363062), Nrn1 (cat # 5367281), and Lrrtm1 (cat # 5321979) were obtained from GE Dharmacon. Gad1 1 Kb cDNA (corresponding to nucleotides 1099–2081) was generated using Superscript II Reverse Transcriptase First Strand cDNA Synthesis kit (cat # 18064014, Invitrogen, La Jolla, CA) according to the manufacturer manual, amplified by PCR using primers mentioned in the primers list, gel purified, and then cloned into a pGEM-T Easy Vector using pGEM-T Easy Vector kit, (cat # A1360, Promega, Madison, WI) according to the kit manual. Sense and anti-sense riboprobes against Gad1, Syt1, Nrn1, and Lrrtm1 were synthesized from 5 µg linearized plasmids using digoxigenin-(DIG) or fluorescein-labeled uridylyltransferase (UTP) (cat # 11685619910, cat # 11277073910, Roche, Mannheim, Germany) and the MAXIscript in vitro Transcription Kit (cat # AM1312, Ambion, Austin, TX) according to the kit manual. 5 µg of Riboprobes (20 µl) were hydrolyzed into ~0.5 kb fragments by adding 80 µl of water, 4 µl of NaHCO3 (1 M), 6 µl Na2CO3 (1 M) and incubating the mixture in 60°C for specific amounts of time determined for each probe by the following formula: Time=(Xkb-0.5)/(Xkb*0.055), where X is the full length of the RNA probe. RNA fragments were finally precipitated in 250 µl 100% ethanol containing 5 µl Acetic acid, 10 µl NaCl (5 M) and 1 µl glycogen (5 mg/ml). Finally, the pellet dissolved in 50 µl of RNAase-free water.

In situ hybridization (ISH)

ISH was performed on 16 μm sections prepared as described above. Sections were fixed in 4% PFA for 10 min, washed with PBS for 15 min, incubated in proteinase K solution (1 µg/ml in 50 mM Tris PH 7.5, 5 mM EDTA) for 10 min, washed with PBS for 5 min, incubated in 4% PFA for 5 min, washed with PBS for 15 min, incubated in acetylation solution (196.6 ml water, 2.6 ml triethanolamin, 0.35 ml HCl, 0.5 ml acetic acid) for 10 min, washed with PBS for 10 min, incubated in 0.1% triton (in PBS) for 30 min, washed with PBS for 40 min, incubated in 0.3% H2O2 (in water) for 30 min, washed with PBS for 10 min, pre-hybridized with hybridization solution (50 ml of Sigma 2X prehyb solution, 25 mg Roche yeast RNA and 8 mg heparin) for 1 hr, hybridized with 50 µl of heat-denatured diluted riboprobes (1–2 µl of riboprobe in 50 µl hybridization solution heated for 10 min in 70°C), mounted with cover slips and kept at 60°C overnight. On day 2, coverslips were gently removed in 60°C preheated 2X saline-sodium citrate (SSC) buffer, and slides were washed 5 times in 60°C preheated 0.2X SSC buffer for 2–3 hr at 60°C. Slides were washed three times with Tris-buffered saline (TBS) and blocked for 1 hr with blocking buffer (0.2% Roche blocking reagent, 10% lamb serum in TBS) prior to overnight 4°C incubation with horseradish peroxidase (POD)-conjugated anti-DIG or anti-fluorescent antibodies (cat # 11426346910 and cat # 11207733910, Roche). On day 3, bound riboprobes were detected by staining with Tyramide Signal Amplification (TSA) system (cat # NEL75300 1KT, PerkinElmer, Shelton, CT). For double ISH, sections were washed in TBS after the TSA reaction, then incubated in 0.3% H2O2 for 30 min, washed with TBS for 10 min, incubated with the second POD-conjugated antibody in blocking buffer and detected with TSA system as described above. Images were obtained on a Zeiss LSM700 confocal microscope. A minimum of three animals per genotype and age were compared in ISH experiments.

Quantitative real time PCR (qPCR)

Pooled tissues (5–7 animals per sample) were isolated from P3, P8, P12 and P25 mice, and RNA was purified using the Aurum Total RNA Fatty and Fibrous Tissue kit (cat # 7326870, BioRad) according to the kit manual. cDNAs were generated with Superscript II RT (Invitrogen). qPCR was performed on a CFX Connect real time system (BioRad) using iTaq SYBRGreen Supermix (cat # 1725124, BioRad) according to the kit protocol. The following cycling conditions were used with 12.5 ng of cDNA: 95°C for 30 s and 42 cycles of amplification (95°C for 10 s, 60°C for 30 s) followed by a melting curve analysis. Relative quantities of RNA were determined using the ΔΔ-CT method (Schmittgen and Livak, 2008). A minimum of n = 3 biological replicates (each in triplicate) was run for each gene. Each individual run included separate Glyceraldehyde-3-Phosphate Dehydrogenase (gapdh), Actin, or 18 s rRNA control reactions. qPCR primers can be found in the primer list.

Western blot

Mice were perfused with PBS, brains removed, and d- and vLGN were dissected separately in ice-cold PBS. Tissues were pooled from >5 littermates per group and subsequenctly lysed in modified loading buffer containing 50 mM Tris–HCl (pH 6.8), 2% sodium dodecyl sulfate (SDS), 10% glycerol, and protease inhibitors (1 mM PMSF). Samples were homogenized, boiled for 10 min, and insoluble material was removed. Protein concentrations were determined by Micro BCA Protein Assay Kit (cat # 23235, Pierce, Rockford, IL). Equal amounts of protein were loaded and separated by SDS-PAGE and transferred to a PVDF membrane as described previously (Fox et al., 2007). After blocking in 5% non-fat milk in PBS (containing 0.05% Tween), PVDF membranes were incubated with primary antibodies (LRRTM1 [Synaptic Systems AF4897], Actin [EMD Millipore MAB1501]), followed by HRP-conjugated secondary antibodies. Immunoblotted proteins were detected with Amersham ECL Prime Western Blotting Detection Reagent (cat # RPN2236).

Intraocular injection of anterograde tracers and AAVs

For intraocular injections, mice were anesthetized with isoflurane or hypothermia, and 1–2 μl of 1 mg/ml CTB was injected into the eye intravitreally with a fine glass pipette attached to a picospritzer. After 2 days, perfused and PFA fixed brains were sectioned (90 μm) using a Vibratome (HM650v, ThermoFisher). Sections were stained with DAPI and mounted with Vectashield (Vector Laboratories, Burlingame, CA). Images were acquired on a Zeiss LSM 700 confocal microscope. A similar approach was used to inject 1–2 μl of a 1:1 mixture of the following AAVs into the eyes: AAV9.hEF1a.lox.TagBFP.lox.eYFP.lox.WPRE.hGH-InvBYF (AV-9-PV2453, 3.47e13 gc/ml) and AAV9.hEF1a.lox.mCherry.lox.mTFP1.lox.WPRE.hGH-InvCheTF (AV-9-PV2454, 1.04e13 gc/ml). AAVs were injected into the eyes of P0 or P12 mice and 1–2 weeks after the injection, mice were anesthetized, perfused, and their brains were fixed in 4% PFA overnight. Brains were then sectioned (90 μm) using a Vibratome and sections were mounted with Vectashield. Images were acquired on a Zeiss LSM 700 confocal microscope.

Serial block-face scanning electron microscopy

Mice were perfused with 0.1M sodium cacodylate buffer containing 4% PFA and 2.5% glutaraldehyde. Brains were immediately vibratomed (300 μm coronal sections), and dLGN tissues were dissected and shipped to Renovo Neural (Cleveland, OH). Processing and image acquisition were performed as described in detail previously (Mukherjee et al., 2016; Hammer et al., 2014). Serial image stacks were analyzed using TrakEM2 in Fiji (Cardona et al., 2012). Presence of synaptic vesicles and pale mitochondria have been used as features to distinguished retinal terminals from non-retinal terminals in dLGN (Hammer et al., 2014; Bickford et al., 2010). Analysis of data sets were performed independently by four researchers to ensure unbiased results.

RNA sequencing

RNA was isolated from vLGN and dLGN at four different ages (P3, P8, P12 and P25) and was shipped to the Genomics Research Laboratory at Virginia Tech’s Biocomplexity Institute for RNAseq analysis. Quality of total RNA was checked on Agilent BioAnalyzer 2100 (Agilent Technologies, Santa Clara CA). Libraries were generated using Apollo 324 Robot (Wafergen, CA). 500 ng of total RNA (with RIN ≥9.0) was enriched for polyA RNA using PrepX PolyA mRNA Isolation Kit (cat # 400047, Wafergen, Fremont, CA) and was then onverted into a library of template molecules using the PrepX RNA-Seq for Illumina Library Kit (cat # 400046, Wafergen, Fremont, CA). Validation of the 280–300 bp libraries (160–180 bp insert) was completed using an Agilent 2100 Bioanalyzer and quantitated using Quant-iT dsDNA HS Kit (cat # Q33120, Invitrogen). Eight individually indexed cDNA libraries were pooled and sequenced on an Illumina HiSeq, resulting in a minimum of 40–50 million reads. Libraries were clustered onto a flow cell using Illumina’s TruSeq PE Cluster Kit v3-cBOT-HS (cat # PE-401–3001), and sequenced 2 × 100 PE using TruSeq SBS Kit v3-HS (200-cycles) (cat # FC-401–3001). Low-quality base calls, sequences with low-complexity tails, and adaptor sequences were removed using a combination of Btrim and EA-utils. Sequencing reads were then aligned to the mouse genome (Tophat2/Bowtie) and expression determined via HTSeq counting. DESeq2 has been used to determine fold change and statistical significance of changes between samples.

Visual behavior tasks

Two alternative forced swim tasks were performed in a trapezoid shaped pool (sides a = 25 cm, b = 80 cm, c and d = 143 cm) with two side-by-side monitors (19 inches, V196L, Acer) placed at the wide end (b) of the tank and separated by a black divider (42 cm). Detailed instructions for the apparatus were described previously (Prusky et al., 2000). A rescue platform (37 cm ×13 cm × 14 cm) was hidden under water below the monitor with the positive visual cue (termed the S+ side). Visual cues (i.e. different grating pattern) were generated in the Gabor-patch generator (https://www.cogsci.nl/gabor-generator). The visual cue and hidden platform were moved to the right or left screens in a pseudorandom manner with the following orders: LRLLRLRRLR, RLRRLRLLRL, RRLRLLRLRL and LLRLRRLRLR. During the behavioral tasks the room was dark, but a 60 W bulb was positioned above the holding cages. During the visual tasks, mice were held in separate cages which were placed on heating pads and lined with paper towels. A day before starting experiments, mice were acclimated to the experimenter and the pool through handling, a 1–2 min period of direct contact with the hidden platform at either arms, and submersion into the water at gradually-increasing distances from the hidden platform. The ability of mice to detect and associate the S+ monitor displaying vertical gratings with the rescue platform (in contrast to the lack of a platform beneath the S- screen that displayed either a gray or horizontal gratings) was assessed. Behavioral tasks included a training phase (8 days) and a testing phase (10–12 days). For training phases, mice were placed at the release chute and given one minute to find the platform for 8–10 trials per day. A trial was recorded as a correct choice if a mouse passed the choice line on the S+ side, while passing the choice line on the S- side was recorded as an incorrect choice. After arriving at the rescue platform, mice were placed back into their individual cages only if they made the correct choice. When a mouse made an incorrect choice, it was placed back at the release chute to perform another trial immediately before going back to its home cage. After 8 days of training, mice learned to find the positive visual cue (i.e. vertical gratings for most tasks, however horizontal gratings in Figure 7—figure supplement 1B) with a >80% accuracy. To test visual acuity and contrast sensitivity, we increased the spatial frequency and decreased the contrast of vertical gratings (i.e. the S+ cue), respectively. In the testing phase of the detection tasks, 10 trials of a given task (e.g. detection of vertical gratings with spatial frequency of 0.32 cpd versus a gray screen) were performed in 10 consecutive days (one per day). For the testing phase of the discrimination tasks two different approaches were used with different cohort of Lrrtm1−/−, Nrn1−/− and control mice. First, a similar approach as the one mentioned above for the detection task (Figure 7—figure supplement 1J). Second, 10 trials of a task (e.g. discrimination of vertical gratings with spatial frequency of 0.32 cpd versus horizontal gratings with spatial frequency of 0.32 cpd) were tested in a single day (Figure 7F,G; Figure 7—figure supplement 1I,J). No more than six animals were tested in a given session. Each mouse (P56-90) performed no more than 10 trials per day.

Quantifications and statistics

For quantifying the size of retinal terminals labeled with fluorescent CTB or VGluT2 immunostaining, the area of isolated puncta (which may contain one or more RGC terminals) were measured in 20X or 40X confocal images of dLGN and vLGN sections by semi-manual selection of the puncta in the ImageJ. 3–7 animals (three sections per animals) were analyzed per age and genotype and the cumulative frequencies of different terminal sizes were obtained. Two-way ANOVA analysis was used to determine any significant change in the distribution of retinal terminal sizes between groups. Intensity and density of the signals in immunostained images of dLGN, vLGN and vCTX were measured in ImageJ. 3–7 animals (three sections per animals) were analyzed per genotype and age and the mean values were compared between groups. T-test or ANOVA were used to determine any significant difference of the mean values between groups.

For SBFSEM, retinal terminals were identified by their unique ultrastructural features including the presence of round synaptic vesicles and pale mitochondria (Hammer et al., 2014; Bickford et al., 2010). Retinal terminals clustering onto the same portion of a dendritic branch were classified as complex RG synapses if the membranes of terminals touched each other and were not isolated from each other by glial processes. On the other hand, a retinal terminal isolated from other retinal terminals was classified as a simple RG synapse. In each mouse, retinal terminals were identified regardless of their simple or complex designation and were then assigned to one of these two classes. The proportion of retinal terminals participating in each class was averaged from data sets obtained from the dLGN of three mice per genotype (2–3 data sets were obtained per mouse). T-test or ANOVA analysis were used to determine any significant difference of the mean values between groups.

The performance of a mouse in the training sessions was reported as the percentage of correct choices the mouse made out of 8 or 10 trials per day (e.g. day 1), and then an average of daily performances was calculated for each group of mice. The performance of a mouse in the test phase of both detection and discrimination tasks were reported as the percentage of correct choices the mouse made out of 10 trials per given task (e.g. for 10% contrast) and these values were used to calculate the mean for a group of mice (e.g. control group). T-test or ANOVA analysis were used to determine any significant difference of the mean values between groups.

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
    Aspects of synaptic and laminar organization of the mammalian lateral geniculate body
    1. RD Lund
    2. TJ Cunningham
    (1972)
    Investigative Ophthalmology 11:291–302.
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
    Stimulus contrast and retinogeniculate signal processing
    1. DL Rathbun
    2. HJ Alitto
    3. DK Warland
    4. WM Usrey
    (2016)
    Frontiers in Neural Circuits, 10, 10.3389/fncir.2016.00008, 26924964.
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
    Synaptic organization of the dorsal lateral geniculate nucleus in the adult hamster. An electron microscope study using degeneration and horseradish peroxidase tracing techniques
    1. KF So
    2. G Campbell
    3. AR Lieberman
    (1985)
    Anatomy and Embryology 171:223–234.
  70. 70
  71. 71
  72. 72
  73. 73
  74. 74
  75. 75
  76. 76
  77. 77
  78. 78
  79. 79
  80. 80
  81. 81
  82. 82
  83. 83

Decision letter

  1. Jeremy Nathans
    Reviewing Editor; Johns Hopkins University School of Medicine, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your manuscript "LRRTM1 underlies synaptic convergence in visual thalamus" to eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor, Jeremy Nathans, and a Senior Editor, Gary Westbrook. As you will see, all of the reviewers were impressed with the importance and novelty of your work.

I am including the three reviews at the end of this letter, as there are a variety of specific and useful suggestions in them. We appreciate that the reviewers' comments cover a range of suggestions for improving the manuscript. Please use your best judgment in deciding which of these can be accommodated in a reasonable period of time. In particular, the first suggestion from reviewer #1 ("It would be interesting to perform eye suturing experiments…") is an excellent idea but, is in our collective opinion beyond the scope of the present work. Most of the other reviewer comments represent relatively minor points that we believe can be accommodated.

Reviewer #1:

In this study, Monavarfeshani et al. have uncovered a molecular mechanism by which retinothalamic convergence occurs during development. It has recently been shown that a high degree of retinal convergence onto thalamic relay cells occurs in the dLGN of both rodents and higher mammals. Here, the authors describe a role for Leucine Rich Repeat Transmembrane Neuronal 1 (LRRTM1) expressed by dLGN relay cells in mediating this process. The expression is quite selective for dLGN relay cells, and this is a major strength of this study. The authors used multiple methods, including serial block-face scanning EM, to support their conclusions and also showed complex visual behavior deficits in lrrtm1 mutant mice. The authors demonstrate a role for lrrtm1 in retinothalamic convergence, thus providing an interesting addition to our understanding of the molecular underpinnings of this developmental phenomenon. Pending response to the issues raised below, this study appears suitable for publication in this journal.

1) What is the role of visual input on the development of retinothalamic convergence? It would be interesting to perform eye suturing experiments to determine if this phenomena still occurs at P14 even without visual input, strengthening the claim that lrrtm1 is required for convergence, especially since some clustering still occurs in lrrtm1 knockouts (Figure 5). This would not entail Serial Blockface EM, but one of the simpler approaches presented in this study and would really strengthen this present communication.

2) Do the authors have any later timepoints for the vGLUT2 staining and CTB experiments to demonstrate that the clustering is not simply delayed in the lrrtm1 mutants? (Related to Figure 5 and Figure 5—figure supplement 1)

3) At what timepoint are the brainbow AAVs being delivered for the tissue harvested at P8 and P10? I would expect that these need to be in utero injections, but it isn't clear in the Materials and methods section.

4) What developmental age are the retina ISH data in Figure 3F?

5) Is there any explanation for the difference in GAD67 staining between Figure 4G and H? Is the red arrow in Figure 4H pointing to a GAD67 positive cell? Please make this clear in the figure legend.

6) Can the authors provide quantification for the percent colocalization of lrrtm1 and nrn1 in relay cells (Figures 4I and J)?

7) In Figure 6—figure supplement 1A and B, it is not clear what conclusion one is supposed to draw about the connection between SBFSEM and CTB/vGLuT2 labeling. It is hard to make out a clear difference between controls and lrrtm1 mutants, as there is no quantification. Also, the word "Age" is written in the bottom left figure of Panel A, but it is not clear that this is supposed to appear there.

8) In Figure 5—figure supplement 1, what age is the retinal tissue that was used for staining?

9) Subsection “Identification of target-derived synaptic organizing molecules in dLGN”: molecules should be molecule.

10) Subsection “Impaired visual behaviors in mice lacking LRRTM1: performing task performance sounds awkward.

11) Subsection “Impaired visual behaviors in mice lacking LRRTM1”: behaviors task should be behavioral tasks.

12) Discussion section, clusters should be cluster.

13) Discussion section, cell should be cells.

14) Discussion section, ontogenetic should be optogenetic.

15) Subsection “LRRTM1 as a target-derived synaptic organizer in visual thalamus”: synapse should be synapses.

16) Subsection “Retinal convergence: artifact or by design?”: rely should be relay.

17) The first portion of the Discussion section is redundant with Introduction section of this paper - this can be edited extensively.

Reviewer #2:

In their submission entitled "LRRTM1 underlies synaptic convergence in visual thalamus", Monavarfeshani and colleagues first demonstrate that the formation of complex RG synapses in the dLGN begins between days P8 and P14 using a thorough battery of independent techniques. To investigate what signaling may underlie development of these synapses, they performed RNAseq experiments comparing the dLGN to the vLGN (where complex synapses do not form) over the course of development and identify lrrtm1 and nrn1 as differentially expressed candidate genes that have previously been implicated in excitatory synapse formation. Using multiple methods, they demonstrate that lrrtm1 mRNA and protein levels are increased at the appropriate developmental timepoints specifically in the dLGN and that this expression is specific to CRH positive relay neurons. To demonstrate a role of lrrtm1 in complex synapse formation, they use a knockout mouse to demonstrate that the formation of complex synapses is decreased when lrrtm1 is lost. Finally, to examine what visual functions may be mediated by complex synapses in the dLGN, visual function in lrrtm1 knockout mice is investigated using a forced swim task where mice discriminate between a visual stimulus of vertical bars and a gray screen, or horizontal bars. Knockout mice perform similarly to controls in regards to frequency and contrast detection thresholds when compared to a gray screen, but are significantly different from controls when tested against horizontal bars.

Overall this work is well done and is of appropriate interest to the community. Especially the use of multiple techniques to examine the formation and disruption of complex synapses in the dLGN. The recent understanding that there is significant retinal convergence in the LGN opens many questions as to the purpose of such circuit architecture. These experiments begin to elucidate the mechanisms of their formation in addition to what role they might play in visual processing.

I do believe that the behavioral aspects of this submission still require further clarification and experiments. To start, from my reading of the Materials and methods section, it is not very clear to me how the testing schedule in Figure 7D and E differ from Figure 7F and G. It appears as though mice in all experiments are trained using stimuli as in B and/or C for 8 days. Then in panels D and E they are tested on 1 of the 8 different conditions, once per day for 10 to 12 days. In panels F and G they are tested on the same stimulus 10-12 times for the indicated days. I think it would be best to further clarify the exact nature of these experiments since the effects on the "complex" visual task mediated by complex synapse loss is a crux of this papers findings. In addition, I am curious why the discrimination task (F and G) is tested on a divergent schedule from the detection task (D and E). This discrepancy makes it difficult to directly compare the two systems. While I appreciate the schedule in F and G to rule out known learning deficits in the lrrtm1 knockout, it would make a stronger case if the behaviors were tested similarly. Finally, the other difference between the detection and discrimination tasks is the availability of horizontal bars. It is plausible that the lrrtm1 knockout doesn't lead to an inability to discriminate, but instead effects detection of vertical bars differentially relative to horizontal bars. Thus, if experiments in panels 7D and E could be repeated putting horizontal bars against a gray background, or conversely testing panels F and G against a gray background, this would indicate a more clear role for discrimination itself as the trait lacking in lrrtm1 knockout mice. Conversely, if a difference is still seen moving panels F and G to a detection task, this might indicate the difference is due to the schedule of behavioral testing.

On the Cumulative Function graphs the y axis appears to pass beyond 100%. This makes it difficult to determine at what point on the x-axis the maximum has been reached. Perhaps rescale the y-axis appropriately or add a landmark on the graph to indicate 100%.

Would it be possible/helpful to expand the brainbow images into a Supplemental figure? It is quite hard to discern the discrete axons presented in some of the figures.

Viral titers used for the brainbow experiments should be indicated in the Materials and methods section.

Reviewer #3:

This is an interesting study that demonstrates a role for LRRTM1 in the maturation of retinal synapses in the dorsal lateral geniculate nucleus (dLGN). The interest arises from two sets of recent studies. One showed that retinal terminals differ in a target dependent manner, even when they likely arise from the same retinal cell types. Thus, LRRTM1 could be a factor that regulates branch-specific synaptogenesis. The second (reported in papers from Fox's group and two others) showed a greater degree of retinal convergence on dLGN neurons than had been supposed. LRRTM1 seems to selectively effect the complex synapses that mediate this convergence. Thus, although mechanistic analysis is lacking here, the result is quite important.

The results are for the most part fairly straightforward, and they are uniformly convincing and well documented. One confound is that the behavioral phenotype may not result exclusively from the LGN defect, but the authors have done all the controls they can, and are quite candid about the limitation. The only better approach would be a conditional mutant, but this would not straightforward because it would require an LGN-specific Cre line (the LRRTM seems to act postsynaptically) which may not be available.

The obviously valuable addition would be physiological tests of decreased convergence in the mutant. My own view is that there are few labs capable of this sort of recording and it is likely beyond the capabilities of the Fox lab, so there is no point in asking for it. I think the paper stands on its own, and likely others will take up the physiological challenge. I look forward to reading the other reviewers' opinion on this point.

Figure 1. Puncta size is nicely quantified from VGlut2 staining (panels C-H) but the authors also go on to document interareal differences using transgenic (I), EM (J) and viral (K-M) methods. It would be nice to see rudimentary quantification of at least one of these – perhaps Hb9-GFP would be most straightforward – to test whether the effects detected by the various methods are of similar magnitude.

Figure 1H and Figure 1—figure supplement 1G look remarkable similar. Please check to be sure there was not an inadvertent duplication.

Subsection “LRRTM1 is required for the development of complex RG synapses”. Having found that 10 and 63% of terminals are big in the two genotypes it is unnecessary to report that 90 and 37% are not big.

Subsection “LRRTM1 is required for the development of complex RG synapses”. Lrrtm1 loss includes both a decrease in the number of complex synapses as well as a decrease in the complexity of the remaining complex synapses. Is it possible to provide a single metric that includes both effects? (This percentage decrease could be included in the text; an additional figure is not necessary.)

Subsection “LRRTM1 is required for the development of complex RG synapses”. Although the focus is on synapse complexity, there is also an effect on synapse size (Figure 6D). This is potentially important. Is there more to say about it? And shouldn't it be noted as a possible confound in the interpretation of the behavioral results?

Subsection “Impaired visual behaviors in mice lacking LRRTM1”. The description of the complex task in which the major effect was observed is somewhat cursory. More is needed. Do the panels in Figure 7F and 7G represent different sets of animals?

Discussion section. The first paragraph of the Discussion section on the recent discovery of convergence in LGN, is largely repetitive of the Introduction. One or the other could be reduced.

Figure 5—figure supplement 1 and Figure 7—figure supplement 1. The title claims more than is delivered in this figure.

https://doi.org/10.7554/eLife.33498.019

Author response

Reviewer #1:

1) What is the role of visual input on the development of retinothalamic convergence? It would be interesting to perform eye suturing experiments to determine if this phenomena still occurs at P14 even without visual input, strengthening the claim that lrrtm1 is required for convergence, especially since some clustering still occurs in lrrtm1 knockouts (Figure 5). This would not entail Serial Blockface EM, but one of the simpler approaches presented in this study and would really strengthen this present communication.

We agree that these are important questions, but also believe that eye-suturing experiments are beyond the scope of the current manuscript. However, we did test whether lrrtm1 or nrn1 mRNA expression was influenced by visual input. We measured mRNA levels in the dLGN of both light-deprived mice (using a late dark rearing model in which deprivation after P20 leads to a weakening of the strength of retinal inputs and an increase in the number of functional retinal inputs onto relay cells; Hooks and Chen, 2008) and transgenic mice which lack retinal projections to the brain (math5-/- mutants; Wang et al., 2001; Hammer et al., 2014). Our data (which we have added in Figure 3—figure supplement 1) show that the expression of lrrtm1 and nrn1 mRNAs are not significantly influenced by altering visual inputs to dLGN. These results may seem somewhat surprising given the established role of Nrn1 (a.k.a. CPG15) as an activity-regulated molecule in the developing brain. However, previous studies have shown that thalamic expression of nrn1 is indeed insensitive to changes in neuronal activity, in contrast to levels in visual cortex (Corriveau et al., 1999).

2) Do the authors have any later timepoints for the vGLuT2 staining and CTB experiments to demonstrate that the clustering is not simply delayed in the lrrtm1 mutants? (Related to Figure 5 and Figure 5—figure supplement 1)

Yes, this is an important point. We have now added VGluT2 and CTB labeling of retinal terminal in adult control and mutant mice (60-85 days of age). Significant differences in the size of CTB- and VGluT2-labeled puncta exist in adult LRRTM1-deificient mice. These results have been added to Figures 5 and Figure 3—figure supplement 1. It is also important to point out that the SBFSEM analysis of mutant and control dLGN (in Figure 6) was performed at P42, an age in which retinogeniculate synapses appear adult-like anatomically and functionally.

3) At what timepoint are the brainbow AAVs being delivered for the tissue harvested at P8 and P10? I would expect that these need to be in utero injections, but it isn't clear in the Materials and methods section.

In order to label retinogeniculate terminals at P8, we performed AAV injections at P0. Surprisingly, the short period of infection (8 days) is sufficient in neonates for the expression and transport of fluorescent reporter proteins in RGCs. These details have been added to the Materials and methods section.

4) What developmental age are the retina ISH data in Figure 3F?

These experiments were performed on P25 retina. This age has now been added to figure legend.

5) Is there any explanation for the difference in GAD67 staining between Figure 4G and H? Is the red arrow in Figure 4H pointing to a GAD67 positive cell? Please make this clear in the figure legend.

This is a very astute observation. The difference is due to slight differences in the age of tissue imaged (P14 vs. P25). We have specified the difference in age in the image and have replaced the images with slightly better images. We have also added text in the figure legend that explains the red arrows indicate GAD67+ dLGN interneurons (and have added these arrows to both panels in Figure 4G and H).

6) Can the authors provide quantification for the percent colocalization of lrrtm1 and nrn1 in relay cells (Figures 4I and J)?

Quantifying the percent colocalization with all thalamic relay cells is quite difficult because tools to specifically label relay cells (and not other neurons) are not available. The Crh-Cre line labels only a subset of relay cells. For this reason, we quantified the total number of syt1, gad1, lrrtm1, and nrn1 expressing cells in dLGN (by comparing to DAPI labeling). We then quantified the percent of syt1+, gad1+, or Crh-Cre+ neurons that co-express either lrrtm1 or nrn1. The important part is that 100% of the Crh-Cre+ neurons express lrrtm1 or nrn1. This data has been added to the Figure 3—figure supplement 1.

7) In Figure 6—figure supplement 1, it is not clear what conclusion one is supposed to draw about the connection between SBFSEM and CTB/vGLuT2 labeling. It is hard to make out a clear difference between controls and lrrtm1 mutants, as there is no quantification. Also, the word "Age" is written in the bottom left figure of Panel A, but it is not clear that this is supposed to appear there.

We have revised the figure legend to clarify this point. The point was to use a visual element to try to explain why VGluT2- and CTB-punta size decreases in mutants (by confocal analysis), but that at the EM level terminal size is not decreased but rather clusters are missing. This supplement was meant to help clarify this point, but if the reviewers prefer we can remove it.

The word “Age” has also been removed.

8) In Figure 5—figure supplement 1, what age is the retinal tissue that was used for staining?

This retinal tissue was from P12. This has been added to the figure legend.

9) Subsection “Identification of target-derived synaptic organizing molecules in dLGN”: molecules should be molecule.

Thank you. This has been fixed.

10) Subsection “Impaired visual behaviors in mice lacking LRRTM1”: performing task performance sounds awkward.

Thank you. This has been fixed.

11) Subsection “Impaired visual behaviors in mice lacking LRRTM1”: behaviors task should be behavioral tasks.

Thank you. This has been fixed.

12) Discussion section, clusters should be cluster.

Thank you. This has been fixed.

13) Discussion section, cell should be cells.

Thank you. This has been fixed.

14) Discussion section: ontogenetic should be optogenetic

Thank you. This has been fixed.

15) Subsection “LRRTM1 as a target-derived synaptic organizer in visual thalamus”: synapse should be synapses.

Thank you. This has been fixed.

16) Subsection “Retinal convergence: artifact or by design?”: rely should be relay.

Thank you. This has been fixed.

17) The first portion of the Discussion section is redundant with Introduction section of this paper-this can be edited extensively.

This section has been revised extensively.

Reviewer #2:

I do believe that the behavioral aspects of this submission still require further clarification and experiments. To start, from my reading of the Materials and methods section, it is not very clear to me how the testing schedule in Figure 7 D and E differ from Figure 7F and G. It appears as though mice in all experiments are trained using stimuli as in B and/or C for 8 days. Then in panels D and E they are tested on 1 of the 8 different conditions, once per day for 10 to 12 days. In panels F and G they are tested on the same stimulus 10-12 times for the indicated days. I think it would be best to further clarify the exact nature of these experiments since the effects on the "complex" visual task mediated by complex synapse loss is a crux of this papers findings.

Yes, this is correct. We have amended the Materials and methods section and Results section to better describe this approach and the results.

In addition, I am curious why the discrimination task (F and G) is tested on a divergent schedule from the detection task (D and E). This discrepancy makes it difficult to directly compare the two systems. While I appreciate the schedule in F and G to rule out known learning deficits in the lrrtm1 knockout, it would make a stronger case if the behaviors were tested similarly.

In fact, we started by using a similar schedule for the discrimination tasks but when we observed difficulty of performing these tasks (particularly in the mutants) we worried about whether learning deficits were confounding our results. We understand the confusion in using different paradigms. We have now included new data where we tested the discrimination task without also assessing task memory. We have added this data into Figure 7—figure supplement 1. We have added it to the supplement and not the main figure, because of the lower number of mutants and controls (n=6 per genotype) we had available at the time we received these reviews. While this analysis was performed on a limited number of animals, it did show a significant decrease in mutant performance in these tasks, supporting that data in Figure 7.

Finally, the other difference between the detection and discrimination tasks is the availability of horizontal bars. It is plausible that the lrrtm1 knockout doesn't lead to an inability to discriminate, but instead effects detection of vertical bars differentially relative to horizontal bars. Thus, if experiments in panels 7D and E could be repeated putting horizontal bars against a gray background, or conversely testing panels F and G against a gray background, this would indicate a more clear role for discrimination itself as the trait lacking in lrrtm1 knockout mice. Conversely, if a difference is still seen moving panels F and G to a detection task, this might indicate the difference is due to the schedule of behavioral testing.

We initially tried to address this concern (that the orientation of the gratings could impact performance) by showing that mutants, and controls for that matter, could learn to differentiate vertical bars from either gray screens or horizontal gratings with similar ability (see Figure 7B,C). We now realize that what was missing from those analyses was the absence of vertical gratings, which could have biased our interpretation of those learning curves. To address this, we have now repeated those learning trials with a different cohort of mice (n=6 controls and 7 mutants), testing their ability to differentiate horizontal gratings (which were the positive cue) from a gray screen. Mutants and controls learned this task similar to how they learned to differentiate vertical gratings from gray screens or horizontal gratings. This new data is included in Figure 7—figure supplement 1 and is described in the text. Taken together, all of these results lead us to believe that LRRTM1 loss does not result in an altered ability to detect vertical versus horizontal bars.

The second part of this questions appears to suggest that we perform a contrast-sensitivity function for mutants and controls by challenging mice to a positive cue that changes both the spatial frequency and contrast of the gratings (while presenting a gray screen as a negative cue). We agree this is a good idea, but due to the number of animals required (especially to test both vertical and horizontal bars which would have to be done in 2 different cohorts of mice) and the time required, we believe these experiments are beyond the scope of the current manuscript.

On the Cumulative Function graphs the y axis appears to pass beyond 100%. This makes it difficult to determine at what point on the x-axis the maximum has been reached. Perhaps rescale the y-axis appropriately or add a landmark on the graph to indicate 100%.

These graphs have been rescaled as suggested.

Would it be possible/helpful to expand the brainbow images into a Supplemental figure? It is quite hard to discern the discrete axons presented in some of the figures.

We feel that these are critical for the manuscript and should remain in the main figures. However, we understand that some of the images are difficult to see the fine details so we have increased the size of the images in the revised version of the manuscript.

Viral titers used for the brainbow experiments should be indicated in the Materials and methods section.

These details have now been added to the methods of the revised manuscript.

Reviewer #3:

Figure 1. Puncta size is nicely quantified from VGlut2 staining (panels C-H) but the authors also go on to document interareal differences using transgenic (I), EM (J) and viral (K-M) methods. It would be nice to see rudimentary quantification of at least one of these – perhaps Hb9-GFP would be most straightforward – to test whether the effects detected by the various methods are of similar magnitude.

We agree. We previously documented ultrastructural differences between retinal terminals in dLGN and vLGN using SBFSEM (Hammer et al., 2014). Here, we went back, as suggested, and analyzed terminal sizes in the dLGN and SC of HB9-GFP transgenic mice. A cumulative frequency distribution of GFP+ puncta size in dLGN and SC has now been added to Figure 1. Additionally, we added the average size of GFP+ puncta in the SC and dLGN of HB9-GFP mice to the revised text.

Figure 1H and Figure 1—figure supplement 1G look remarkable similar. Please check to be sure there was not an inadvertent duplication.

Thank you! You are absolutely correct and we have now fixed this mistake.

Subsection “LRRTM1 is required for the development of complex RG synapses”. Having found that 10 and 63% of terminals are big in the two genotypes it is unnecessary to report that 90 and 37% are not big.

We have revised this sentence to avoid redundancy.

Subsection “LRRTM1 is required for the development of complex RG synapses”. Lrrtm1 loss includes both a decrease in the number of complex synapses as well as a decrease in the complexity of the remaining complex synapses. Is it possible to provide a single metric that includes both effects? (This percentage decrease could be included in the text; an additional figure is not necessary.)

While we agree conceptually with this (to simplify the phenotypes) we have had trouble doing this without omitting important data. For example, one possibility is that we could report a single metric like we did for the brainbow analysis (Figure 6I). However, this would not accurately demonstrate the decrease in complex RG synapse number in the mutants.

(Perhaps, it is important to note that for our AAV-brainbow analysis in mutants and controls did use only a single metric (by only looking at complex retinal terminals and classifying them into RG synapse with less or more than 4 terminals). The rationale for this was that we do not feel that it is scientifically valid to quantify a decrease in the number of complex RG synapses with a method (i.e. viral delivery of genes encoding fluorescent proteins) that does not label all retinal axons.)

Subsection “LRRTM1 is required for the development of complex RG synapses”. Although the focus is on synapse complexity, there is also an effect on synapse size (Figure 6D). This is potentially important. Is there more to say about it? And shouldn't it be noted as a possible confound in the interpretation of the behavioral results?

We agree that this is a really remarkable phenotype and could be very important. We have now added additional data on this point, showing that active zones are increased in both simple and complex RG boutons in LRRTM1-deficient mutants. It is important to note, however, that when we normalize for terminal size the density of axon zones appears no different than in control boutons. All of this has been added to Figure 6. We have also added more details about the increase in bouton size in LRRTM1-deficient mutants in the discussion (and how this could confound interpretation of our results).

Subsection “Impaired visual behaviors in mice lacking LRRTM1”. The description of the complex task in which the major effect was observed is somewhat cursory. More is needed. Do the panels in Figure 7F and 7G represent different sets of animals?

We have added more detail to the methods and Results sections, as described above for questions from reviewer 2. Yes, different sets of animals we used for those experiments (Figure 7F and G) and for the new data now added to Figure 6—figure supplement 1.

Discussion section. The first paragraph of the Discussion section on the recent discovery of convergence in LGN, is largely repetitive of the Introduction. One or the other could be reduced.

This has been revised.

Figure 5—figure supplement 1 and Figure 7—figure supplement 1. The title claims more than is delivered in this figure.

We agree and have not only changed the title but have split this figure into Figure 7—figure supplement 2 and Figure 2—figure supplement 1. This allowed us to provide additional, high magnification images of VGluT1+ and VGluT2+ nerve terminals in layer IV of visual cortex of mutants and controls.

https://doi.org/10.7554/eLife.33498.020

Article and author information

Author details

  1. Aboozar Monavarfeshani

    1. Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    2. Department of Biological Sciences, Virginia Tech, Blacksburg, United States
    Contribution
    Conceptualization, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon 0000-0001-8906-5115
  2. Gail Stanton

    1. Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    2. Virginia Tech Carilion School of Medicine, Roanoke, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  3. Jonathan Van Name

    Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  4. Kaiwen Su

    Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  5. William A Mills III

    1. Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    2. Translational Biology, Medicine, and Health Graduate Program, Virginia Tech, Blacksburg, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  6. Kenya Swilling

    Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  7. Alicia Kerr

    1. Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    2. Translational Biology, Medicine, and Health Graduate Program, Virginia Tech, Blacksburg, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  8. Natalie A Huebschman

    Roanoke Valley Governor School, Roanoke, United States
    Contribution
    Data curation, Formal analysis
    Competing interests
    No competing interests declared
  9. Jianmin Su

    Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  10. Michael A Fox

    1. Developmental and Translational Neurobiology Center, Virginia Tech Carilion Research Institute, Roanoke, United States
    2. Department of Biological Sciences, Virginia Tech, Blacksburg, United States
    3. Virginia Tech Carilion School of Medicine, Roanoke, United States
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Investigation, Writing—original draft, Project administration, Writing—review and editing
    For correspondence
    mafox1@vtc.vt.edu
    Competing interests
    No competing interests declared
    ORCID icon 0000-0002-1649-7782

Funding

National Eye Institute (EY021222)

  • Michael A Fox

Brain and Behavior Research Foundation

  • Michael A Fox

National Eye Institute (EY024712)

  • Michael A Fox

Virginia Tech Carilion Research Institute (Medical Research Scholars Program)

  • Aboozar Monavarfeshani

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

This work was supported in part by the National Institutes of Health – National Eye Institute grants EY021222 and EY024712 (MAF), an Independent Investigator grant from the Brain and Behavior Foundation (MAF), and a fellowship from the VTCRI Medical Research Scholars Program (AM). The authors are grateful to Drs. G Valdez, K Mukherjee and YA Pan for helpful comments and discussions. We also thank Drs. S Robel (VTCRI) and W Guido (U Louisville) for generously supplying Aldh1l1-GFP and Crh-Cre mice, respectively.

Ethics

Animal experimentation: All experiments were performed in compliance with National Institutes of Health (NIH) guidelines and protocols and were approved by the Institutional Animal Care and Use Committee (IACUC# 15-137VTCRI, 15-167VTCR and 15-174VTCRI) and Institutional Biosafety Committee (IBC# 15-038) at Virginia Tech.

Reviewing Editor

  1. Jeremy Nathans, Reviewing Editor, Johns Hopkins University School of Medicine, United States

Publication history

  1. Received: November 11, 2017
  2. Accepted: February 8, 2018
  3. Accepted Manuscript published: February 9, 2018 (version 1)
  4. Version of Record published: February 26, 2018 (version 2)

Copyright

© 2018, Monavarfeshani et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 938
    Page views
  • 138
    Downloads
  • 0
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

  1. Further reading

Further reading

    1. Neuroscience
    2. Structural Biology and Molecular Biophysics
    Riley Perszyk et al.
    Research Article
    1. Neuroscience
    Povilas Karvelis et al.
    Research Article Updated