Molecular correlates of laminar differences in the macaque dorsal lateral geniculate nucleus

Abstract

In anthropoid primates, cells in the magnocellular and parvocellular layers of the dorsal lateral geniculate nucleus (dLGN) are distinguished by unique retinal inputs, receptive field properties, and laminar terminations of their axons in visual cortex. To identify genes underlying these phenotypic differences, we screened RNA from magnocellular and parvocellular layers of adult macaque dLGN for layer-specific differences in gene expression. Real-time quantitative reverse transcription-PCR and in situ hybridization were used to confirm gene expression in adult and fetal macaque. Cellular localization of gene expression revealed 11 new layer-specific markers, of which 10 were enriched in magnocellular layers (BRD4, CAV1, EEF1A2, FAM108A1, INalpha, KCNA1, NEFH, NEFL, PPP2R2C, and SFRP2) and one was enriched in parvocellular and koniocellular layers (TCF7L2). These markers relate to functions involved in development, transcription, and cell signaling, with Wnt/beta-catenin and neurofilament pathways figuring prominently. A subset of markers was differentially expressed in the fetal dLGN during a developmental epoch critical for magnocellular and parvocellular pathway formation. These results provide new evidence for the molecular differentiation of magnocellular and parvocellular streams through the primate dLGN.

Figure 1.

A schematic diagram identifying the target dLGN layers harvested for microarray analysis (left). Contralateral and ipsilateral retinal ganglion cell axons originate from nasal and temporal retinal ganglion cells, respectively, and terminate in layers 1, 4, and 6 (contralateral) or 2, 3, and 5 (ipsilateral) of the dLGN, visualized here by Nissl staining. Axons terminating in outer (1 and 2) and inner (3–6) layers form the magnocellular and parvocellular visual streams, respectively. The S layers are indicated (arrow). Magnocellular and parvocellular layers after microdissection with tissue punches are shown by post hoc Nissl staining (right).

Figure 2.

Distribution of normalized microarray data. Box plots representing robust multichip analysis and median polished gene expression data show a lack of variability in distribution between individual GeneChips. Samples from magnocellular (ML) (red) or parvocellular (PL) (blue) layers showed similar distributions in biological replicates (monkey RM123-7 versus monkey RM136-9) and between technical replicates within the same monkey.

Figure 3.

Genes differentially expressed in dLGN layers identified by microarray analysis. A, Volcano plot of robust multichip analysis and median polished Affymetrix GeneChip data. Fold expression levels between magnocellular and parvocellular layers are plotted on the x-axis (log₂ transformed) and t statistic p values are plotted on the y-axis (−log₁₀ transformed). Genes that displayed at least a 1.5-fold difference and a value of p < 0.05 are indicated in the top left and right quadrants of the plot. B, Genes identified in the volcano plot as having matched the cutoff criteria are hierarchically listed here by fold increase in parvocellular layers (red) compared with magnocellular layers (blue).

Figure 4.

RT-PCR confirmation of microarray measures. Twelve of the top genes identified by microarray were analyzed independently by RT-PCR. Each gene was measured in triplicate and plotted as fold increase in parvocellular (dark gray) expression levels over magnocellular (light gray) expression. Genes with fold differences opposite those observed by microarray are indicated (*). The bars indicate mean values ± SEM.

Figure 5.

In situ hybridization confirmation of magnocellular enriched gene expression. A, Coronal Nissl-stained section through monkey dLGN illustrating the cytoarchitecture, including the six principal cell layers and the S layers (S). The red dashed line indicates the separation between magnocellular (1 and 2) and parvocellular (3–6) layers. B, Coronal section through monkey dLGN immunostained with monoclonal antibody Cat-301, which recognizes an epitope on the protein aggrecan and is known to be elevated in magnocellular dLGN layers (Hendry et al., 1984). Some staining in S layers is also present (arrow). C, E, G, I, K, M, O, Images of film autoradiograms illustrating hybridization of radioactive cRNA probes specific to the genes BRD4 (C), CAV1 (E), EEF1A2 (G), FAM108A1 (I), KCNA1 (K), PPP2R2C (M), and SFRP2 (O) on coronal sections of monkey dLGN. Consistent with RT-PCR results, each gene displayed elevated expression in one or both magnocellular layers. D, F, H, J, L, N, P, Laminar differences in mRNA expression were quantified from film autoradiograms of sections processed for in situ hybridization for BRD4 (D), CAV1 (F), EEF1A2 (H), FAM108A1 (J), KCNA1 (L), PPP2R2C (N), and SFRP2 (P). Calibrated densitometric measures were obtained from the principal layers (1–6) of the dLGN. Measures are means ± SEM. *p < 0.01 (n = 5), ANOVA, post hoc Student's t test. Scale bar: (in O) C, E, G, I, K, M, 1 mm.

Figure 6.

TCF7L2 mRNA expression is overrepresented in parvocellular layers of monkey dLGN. A, Bright-field image of a Nissl-stained coronal section of monkey brain at the level of mid-dLGN. B, Image from a film autoradiogram illustrating in situ hybridization for TCF7L2 mRNA in a coronal section adjacent to that in A. TCF7L2 mRNA is mostly restricted to dorsal thalamus (DT) and shows prominent expression in dLGN. C, Higher power image of Nissl-stained dLGN showing the principal relay layers (1–6) and the S layers. D, Higher power image of a section adjacent to that in C showing TCF7L2 in situ hybridization. TCF7L2 mRNA expression is greater in the parvocellular layers (3–6) compared with magnocellular layers (1, 2). Hybridization to TCF7L2 mRNA is also observed in the S layers. E, Calibrated densitometric measures of film autoradiographs illustrate the enrichment for TCF7L2 hybridization in parvocellular layers compared with magnocellular layers. Measures are means ± SEM. *p < 0.01 (n = 5), ANOVA, post hoc Student's t test. Scale bars: (in B) A, B, 2 mm; (in D) C, D, 1 mm.

Figure 7.

TCF7L2 protein is enriched in parvocellular, S, and intercalated layers of the monkey dLGN. A, Nissl-stained coronal section of monkey dLGN illustrating the principal relay layers (1–6) and the intercalated layers that separate them. The S layers are not readily visible in this section. B, Immunohistochemical labeling of TCF7L2 protein in a coronal section of monkey dLGN adjacent to that in A. C, E, Higher power images of Nissl-stained cells in layer 1 (C) and 6 (E) of dLGN. D, F, Higher power images from sections adjacent to those in C and E show immunoreactivity for TCF7L2 protein in layers 1 (D) and 6 (F) of dLGN. More immunoreactive cells are observed in layer 6 compared with layer 1. G–I, Double immunofluorescent labeling of TCF7L2 (red) and type II α calcium/calmodulin-dependent protein kinase (CAMK2A) (green) in dLGN. Merged images (I) show colocalization of TCF7L2 and CAMK2A in the intercalated layers of the dLGN and in the S layers (insets). Scale bars: (in A) A, B, 1 mm; (in D) C–F, 250 μm; (in I) G–I, 200 μm.

Figure 8.

IPA was performed using a gene set combined of the novel cell-specific dLGN layer markers identified and those previously reported. A, Thirteen of 15 molecules queried were mapped to a single IPA network. Manual expansion of the network (nodes and edges in orange) identified a connection between the phosphatase subunit, PPP2R2C, and the neurofilament protein NEFM via the neurofilament molecules NEFH, NEFL, and INα. Nodes and edge symbols are explained in legends at right. Open symbols, Network molecules identified by IPA. Filled symbols, Confirmed dLGN layer-specific markers (focus molecules). B, Two networks of proteins associated through direct and indirect interactions at a level greater than chance were identified and scored based on the number of input genes (focus molecules) linked with the network. The top functions and/or diseases significantly overrepresented by molecules in these networks are listed.

Figure 9.

Phosphatase activity and neurofilament expression is greater in magnocellular layers of monkey dLGN. A, Immunohistochemical staining of a coronal dLGN section with SMI-32 monoclonal antibody showing increased staining of the magnocellular layers (1, 2) of the dLGN compared with parvocellular layers (3–6). B, C, Higher power images of SMI-32 immunostaining shows denser reaction product over layer 1 (B) compared with layer 6 (C) and illustrate positive immunoreactivity within cell bodies (arrows). D–H, Autoradiographic images of coronal dLGN sections processed for in situ hybridization showing increased hybridization of riboprobes specific to PPP2R2C (D), NEFL (E), NEFM (F), NEFH (G), and INα (H) mRNA in magnocellular layers of the dLGN. Scale bar: (in H) A–H, 1 mm.

Figure 10.

Markers of dLGN laminae in adult brain are expressed in E55 thalamus. A, Coronal Nissl-stained section through an E55 monkey thalamus illustrating cytoarchitectural differentiation of thalamic nuclei. The location of the dLGN (*) in relation to the anlagen of dorsal thalamus (DT) and epithalamus (ET) is indicated by the dashed bounding box. M, P, or K laminae cannot be reliably identified at this age (Huberman et al., 2005). An arrow indicates the location of the third ventricle. B–D, Autoradiographic images of coronal sections adjacent to that in A processed for in situ hybridization with riboprobes specific to the genes NEFL (B), INα (C), and TCF7L2 (D). NEFL and INα are both heavily expressed in the developing dLGN, whereas TCF7L2 displays relatively low levels of expression. Scale bar: (in D) A–D, 1 mm. Abbreviations: CTX, Neocortex; DT, dorsal thalamus; ET, epithalamus; GE, ganglionic eminence.