Paraventricular Hypothalamic Nucleus Upregulates Intraocular Pressure Via Glutamatergic Neurons

Abstract

Purpose: The neuroregulatory center of intraocular pressure (IOP) is located in the hypothalamus. An efferent neural pathway exists between the hypothalamic nuclei and the autonomic nerve endings in the anterior chamber of the eye. This study was designed to investigate whether the paraventricular hypothalamic nucleus (PVH) regulates IOP as the other nuclei do.

Methods: Optogenetic manipulation of PVH neurons was used in this study. Light stimulation was applied via an optical fiber embedded over the PVH to activate projection neurons after AAV2/9-CaMKIIα-hChR2-mCherry was injected into the right PVH of C57BL/6J mice. The same methods were used to inhibit projection neurons after AAV2/9-CaMKIIα-eNpHR3.0-mCherry was injected into the bilateral PVH of C57BL/6J mice. AAV2/9-EF1α-DIO-hChR2-mCherry was injected into the right PVH of Vglut2-Cre mice to elucidate the effect of glutamatergic neuron-specific activation. IOP was measured before and after light manipulation. Associated nuclei activation was clarified by c-Fos immunohistochemical staining. Only mice with accurate viral expression and fiber embedding were included in the statistical analysis.

Results: Activation of projection neurons in the right PVH induced significant bilateral IOP elevation (n = 11, P < 0.001); the ipsilateral IOP increased more noticeably (n = 11, P < 0.05); Bilateral inhibition of PVH projection neurons did not significantly influence IOP (n = 5, P > 0.05). Specific activation of glutamatergic neurons among PVH projection neurons also induced IOP elevation in both eyes (n = 5, P < 0.001). The dorsomedial hypothalamic nucleus, ventromedial hypothalamic nucleus, locus coeruleus and basolateral amygdaloid nucleus responded to light stimulation of PVH in AAV-ChR2 mice.

Conclusions: The PVH may play a role in IOP upregulation via glutamatergic neurons.

Glaucoma is the most common cause of irreversible blindness worldwide.¹ Current therapy aims to lower intraocular pressure (IOP), which is the primary risk factor for glaucoma.² Physiologically, normal IOP is characterized by binocular symmetry and diurnal fluctuation, suggesting that neuromodulation might be involved in the homeostatic regulation of IOP.³ Trigeminal nerve terminals distributed in the anterior chamber (AC) are proposed to serve as intraocular baroreceptors.^4–7 The trigeminal nerve may act as the primary afferent nerve to transmit signals to the brain.⁸ The hypothalamus is proposed to be the regulatory center for IOP.^9–11 The autonomic nerve endings are distributed in the AC, regulating the production and outflow of aqueous humor (AH) via the ciliary body, the trabecular meshwork, and Schlemm's canal.¹² In addition, IOP correlates strongly with ocular blood flow, which is also regulated by autonomic neurons in the brainstem that innervate the choroid.^13,14 

Recently, it was suggested that an efferent neural pathway exists from the hypothalamus to the autonomic nerve endings in the AC of rats.¹⁵ This provided anatomical evidence for the hypothalamic nuclei regulating IOP via the autonomic nerve system (ANS). The hypothalamic nuclei indirectly projecting to the autonomic nerves in the AC include the suprachiasmatic nucleus (SCN), the dorsomedial hypothalamic nucleus/perifornical region (DMH/PeF), the paraventricular hypothalamic nucleus (PVH), and the ventral tuberomammillary nucleus (VTM).¹⁵ Among the above nuclei, the SCN has been implicated in the circadian regulation of IOP,^16,17 and inhibition of GABA receptors in the DMH/PeF can considerably elevate IOP.¹⁰ However, it remains unclear whether the PVH and VTM participate in IOP neuromodulation as the others do. 

The PVH, a highly conserved region of the brain from zebrafish to humans, is located around the third ventricle.¹⁸ Magnocellular and parvocellular neurons are two types of neurons in the PVH, and some parvocellular neurons with long-range projections send axons to the brainstem and spinal cord, providing an important efferent route for the ANS.^19–23 Meanwhile, the PVH is also considered a critical central integration site for autonomic activity, which has been extensively studied in the regulation of cardiovascular system functions, such as systemic blood volume, circadian blood pressure, and cardiovascular response to pressure changes.^24–27 The PVH sends nerves to the kidneys and small arteries to control blood pressure as well.²⁸ Additionally, the ANS plays an important role in the neural regulation of IOP.¹² There are numerous connections between the PVH and the brain regions associated with IOP regulation. The SCN is primarily engaged in mammalian circadian regulation and circadian rhythm of IOP, and the SCN sends large numbers of projections to the PVH.²⁹ Stimulation of the DMH, ventromedial hypothalamic nucleus (VMH), supraoptic nucleus (SO) or amygdala neurons can induce increases in IOP,^10,30–32 and the PVH receives input from the above nuclei.^33–35 Therefore we hypothesize that the PVH may have a very important role in the neural modulation of IOP. 

Optogenetics is the combination of optical and genetic methods to modulate the activity of certain cells of a living tissue³⁶ and to excite or inhibit cellular activity when stimulated by the light of specific wavelengths.³⁷ This approach is capable of precisely modulating neural activity with cell type specificity in the sub-millisecond range.³⁸ To verify our hypothesis, we used optogenetics to evaluate IOP variation after PVH stimulation. 

Methods

Animals

All mice were kept in a standard laboratory animal environment of temperature and humidity with a 12/12-h light/dark cycle and free access to food and water. Male C57BL/6J mice (six to eight weeks of age) were obtained from the Animal Experimental Center of Tongji Medical College of Huazhong University of Science and Technology. Vglut2-Cre mice (Jackson no. 028863, gifts from Prof. Fuqiang Xu) were crossed with C57BL/6J mice. Adult transgenic mice (six to eight weeks) of both sexes were used. All surgical and experimental procedures followed the guidelines established by the Animal Care and Use Committees at the Huazhong University of Science and Technology and the Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences. 

Preparation of Viral Vectors

Adeno-associated virus (AAV) vectors for AAV-CaMKIIα-hChR2-mCherry, AAV-CaMKIIα-eNpHR3.0-mCherry, AAV-CaMKIIα-mCherry, AAV-EF1α-DIO-hChR2-mCherry, and AAV-EF1α-DIO-mCherry were packaged into 2/9 serotypes with final titers of approximately 5.00 × 10¹² genomic copies per milliliter. 

Virus Injection

The process for injecting the virus was identical to that used previously.³⁹ Briefly, mice were anesthetized with sodium pentobarbital (80 mg/kg) and mounted to a stereotaxic instrument (RWD, Shenzhen, China). Then 100 nl AAV-CaMKIIα-hChR2-mCherry, AAV-CaMKIIα-eNpHR3.0-mCherry, AAV-CaMKIIα-mCherry, AAV-EF1α-DIO-hChR2-mCherry, or AAV-EF1α-DIO-mCherry was injected into the PVH (anteroposterior [AP], −0.9 mm from the Bregma; mediolateral [ML], −0.2 mm from the midline; dorsoventral [DV], −4.8 mm from the Bregma surface in C57BL/6J mice and AP, −0.83 mm; ML, −0.2 mm; DV, −4.87 mm in Vglut2-Cre mice, respectively). For optogenetic inhibition, the coordinates are AP, −0.9 mm; ML, ±0.2 mm; DV, −4.8 mm. The glass micropipette was left in the brain for 10 min following injections to avoid virus leakage. 

Optogenetic Manipulation of PVH Neurons in Conscious Mice

A fiberoptic cannula (125 µm outer diameter, 0.37 numerical aperture; Newdoon, Shanghai) was stereotaxically implanted above the PVH (AP: −0.9 mm; ML, −0.2 mm; DV, −4.6 mm in C57BL/6J mice; AP: −0.83 mm; ML, −0.2 mm; DV, −4.67 mm in Vglut2-Cre mice). Dental cement was used to secure the fiber optics. 

After a three-week recovery from the fiber-optic cannula implantation, the mice were habituated to the head-fixed running machine and the optical fiber cables (0.8 m long, 200 µm diameter, RWD). For activation experiments, the cable was connected to the blue light source (470 nm), and a pulse generator (Master-8) was connected to the light stimulator (frequency, 20 Hz; duration, 20 ms). The stimulation was applied for 20 seconds with a three-minute interval and cycled five times. For inhibition experiments, the cable was attached to the yellow light source (589 nm), and the stimulation was conducted three minutes on/three minutes off and cycled five times. The same procedures were applied to all experimental and control animals, except the virus did not express ChR2 or eNpHR3.0 in control animals. 

Measurement of IOP in Conscious Mice

To measure IOP in conscious mice, a running wheel was used to keep the mice staying in place, and a head fixation device was used to fix the mouse head loosely. Before each official experiment, the mice were acclimated to the head fixation apparatus and IOP measurement procedure for 20 to 30 minutes over two consecutive days to minimize the presumed stress effects. IOP was measured by a handheld rebound tonometer (Icare TonoLab; Colonial Medical Supply, Franconia, NH, USA) according to the manufacturer's instructions in the same manner as in prior research.⁴⁰ The tonometer records six readings from the same eye, discards the highest and lowest numbers, and displays the average of the remaining four values automatically as a IOP value. Before light stimulation, five IOP values were recorded and the average value was taken as the baseline IOP. During the experiment, we measured IOP during a 20-second light-on period and one minute after the light was turned off. We measured the IOP as many as possible during the light-on and light-off periods, and the IOP values were collected for five cycles. Then the average IOPs of light on or off period were calculated respectively. Each measurement was performed first on the eye ipsilateral to the optogenetic fiber, and subsequently on the contralateral eye. IOP measurements were taken at similar time each day for the respective mice. 

Slice Preparation, Immunohistochemistry, and Imaging

Mice were overdosed with sodium pentobarbital (100 mg/kg, intraperitoneally) and transcardially perfused with phosphate-buffered saline (PBS, pH 7.4; Sinopharm, Shanghai, China), followed by PBS containing 4% paraformaldehyde (PFA, Sigma). After the brain tissues were carefully extracted from the skull for postfixation and cryoprotection, the brain tissues were sliced into 40 µm coronal sections using a cryostat microtome (Thermo Fisher Scientific) and stored at −20°C. 

Seventy-five minutes after photostimulation (15 min), mice were sacrificed for c-Fos immunostaining. The floating sections were washed in PBS and then incubated at 4°C in the following primary antibodies against rabbit c-Fos (1:1000 for 48 hours) in PBS containing 0.3% Triton X-100. The brain slices were rinsed three times with PBS before incubation with secondary antibodies at 37°C for 2 h (Alexa 488, 1:200; Abcam, Cambridge, MA, USA). Finally, the slices were stained with DAPI (1:4000; Beyotime Institute of Biotechnology, Jiangsu, China), mounted with 75% glycerol (Sinopharm) in PBS, and sealed with nail polish. The brain slices were imaged with a VS120 virtual microscopy slide scanning system (Olympus, Tokyo, Japan) and confocal (Leica, Wetzlar, Germany). Only mice with accurate viral expression and fiber embedding were subsequently included in the statistical analysis. 

Statistics

All data are expressed as the mean ± standard error of the mean (SEM). Graphs were created using GraphPad Prism v7 software and Adobe Illustrator. Statistical analysis was performed using SPSS 24.0 statistical software. For single-value comparisons, paired and unpaired t tests and the Kruskal‒Wallis test were utilized. To compare more than two groups, we used one-way ANOVA and the Kruskal‒Wallis test, followed multiple pairwise comparisons using the Bonferroni test. All statistical tests were two-tailed. P < 0.05 was considered statistically significant. 

Results

IOP Elevation in the Ipsilateral Eye is Higher Than That in the Contralateral Eye After Activation of Unilateral PVH Projection Neurons

CaMKIIα, the α-subunit of the serine/threonine kinase CaMKII, is expressed in the hippocampus, neocortex, thalamus, hypothalamus, cerebellum, and basal ganglia.^41,42 In subcortical structures, CaMKIIα is mainly localized in projection neurons.^43–45 Channelrhodopsin-2 (ChR2), a light-gated cation-selective channel that when activated depolarizes neural cells.⁴⁶ Therefore the projection neurons within the PVH could be activated by blue light after we injected AAV-CaMKIIα-hChR2-mCherry into the PVH. 

To identify the optimal light stimulation intensity, four light intensity levels (1, 2, 5, and 8 mW) were set up. The stimulation procedure was performed as shown in Figure 1A, and the ipsilateral IOP was measured. The percentage change of IOP under light stimulations with different intensity were compared. In AAV-ChR2 mice, the IOPs were increased after light stimulation (Fig. 1B, left), and the percentage change of IOP was 26% ± 8% (1 mW), 35% ± 12% (2 mW), 57% ± 10% (5 mW) and 37% ± 6% (8 mW), respectively (n = 4, Fig. 1B, right). The most significant changes in IOP were observed at 5 mW and 8 mW light stimulation. In subsequent studies, the light power was programmed to 5 mW to ensure optimal activation. 

Figure 1.

 

Optogenetic activation of PVH projection neurons increases ipsilateral IOP. (A) Schematic of the light administration process. (B) Robust IOP elevation at 1 mW, 2 mW, 5 mW, and 8 mW (n = 4 mice, P < 0.05). The left panel displays the trend of IOP after light stimulation. The right figure illustrates the degree of IOP variation under different light intensities. (C) Statistical analysis of baseline IOP in the two groups of mice. (D) Representative IOP changes in the two groups of mice with or without light stimulation of the right eye. (E) The IOP changes in the right eyes (ipsilateral to the optic fiber) before, during and after light stimulation in the AAV-ChR2 group (n = 11, repeated-measures ANOVA; P < 0.001). (F) The IOP of right eyes in the control group (n = 9, repeated-measures ANOVA; P > 0.05). (G) The IOP changes in the left eye in the ChR2 group before, during and after light stimulation (n = 11 mice, Kruskal-Wallis test; P < 0.001). (H) The IOP changes of left eyes in the control group (n = 9 mice, Kruskal-Wallis test; P > 0.05). (I) Comparison of bilateral IOP after blue light administration (n = 11 mice, paired t test, P < 0.05). (J) c-Fos cell counting in bilateral PVH in the ChR2 group and Control group. (K) Representative images showing virus expression and fiber position. Left: unilateral PVH expressing AAV-CaMKIIα-hChR2-mCherry; right: unilateral PVH expressing AAV-CaMKIIα-mCherry. Scale bar: 200 µm. The white dotted line indicates the PVH boundary, and the white arrow indicates optical fiber implantation. (L) Magnified view of the PVH. Representative images showing coexpression of PVH neurons. (a) Neurons expressing mCherry; (b) Neurons expressing c-Fos; (c) Nuclei stained with DAPI; (d) Arrows indicate neurons coexpressing mCherry and c-Fos. Scale bars: 30 µm. The panel represents the control group (a1–d1); the ipsilateral (a2–d2) and contralateral sides (a3–d3) of ChR2 group, respectively.

Figure 1.

 

Continued.

Before light stimulation, the bilateral baseline IOPs were measured, as shown in Figure 1C. Optogenetic activation of unilateral PVH projection neurons significantly increased the ipsilateral IOP in ChR2 mice, and the IOP fell back after the light was turned off (20.1 ± 0.8 mm Hg at baseline vs. 28.9 ± 0.6 mm Hg with the light on versus 24.3 ± 0.4 mm Hg with the light off, n = 11; Figs. 1D, E). In control mice injected with AAV-CaMKIIa-mCherry, there was no significant change on ipsilateral IOP (22.2 ± 0.7 mm Hg at baseline vs. 22.5 ± 0.3 mm Hg with the light on vs. 22.9 ± 0.4 mm Hg with the light off, n = 9; Figs. 1D, 1F). These findings demonstrate that optogenetic activation of unilateral PVH projection neurons potently increased ipsilateral IOP. 

Previous research has shown that the bilateral hypothalamus provides crossed and uncrossed signals to bilateral autonomic nerve endings in the AC.¹⁵ This suggested that the unilateral PVH might regulate bilateral IOP. Therefore we measured the contralateral IOP in both groups. We found that optogenetic activation of the PVH increased contralateral IOP in ChR2 mice as well, and the IOP fell after the light was stopped (20.5 ± 0.8 mm Hg at baseline vs. 26.4 ± 0.8 mm Hg when the light was turned on vs. 24.2 ± 0.5 mm Hg when the light was turned off, n = 11; Fig. 1G). In control mice, there was no significant change on contralateral IOP (21.5 ± 0.6 mm Hg at baseline vs. 22.4 ± 0.4 mm Hg with the light on vs. 22.6 ± 0.4 mm Hg with the light off, n = 9; Fig. 1H). However, compared with the ipsilateral eye, the content of IOP elevation of the contralateral eye was significantly lower (Fig. 1I). 

After IOP measurements, mice were perfused to determine viral expression as well as fiber position, as shown in Figure 1K. c-Fos protein is encoded by an immediate early gene and is commonly used as a marker of neuronal activity.⁴⁷ Therefore we observed the colabeling of the c-Fos and mCherry in the PVH. The immunofluorescence results showed a high level of c-Fos expression in the ipsilateral PVH, and a lower level of c-Fos expression in the contralateral PVH in the ChR2 group. At the same time, we counted c-Fos in the bilateral PVH of both groups. In the ChR2 group, the number of labeled neurons in the ipsilateral PVH was 231 ± 9, and the c-Fos+ neurons in the contralateral PVH was 98 ± 12. In the control group, the c-Fos+ neurons in bilateral PVH were 2 ± 1 and almost zero, respectively (Fig. 1J), whereas few c-Fos fluorescence signals could be observed in the PVH in the control group (Fig. 1L). 

Suppressing PVH Projection Neurons has no Impact on IOP

To evaluate whether suppressing PVH projection neurons will lower IOP, we injected AAV-CaMKIIα-eNpHR3.0-mCherry or AAV-CaMKIIα-mCherry into the bilateral PVH of C57BL/6J mice, respectively. In contrast to ChR2, activation of eNpHR3.0 photosensitive proteins under yellow light stimulation causes an influx of chloride ions and thereby inhibits neural activity.⁴⁸ The stimulation procedure was performed as shown in Figure 2A. The baseline IOP of mice in the eNpHR3.0 group was 24.9 ± 0.8 mm Hg, and the IOPs during and after yellow light exposure were 24.1 ± 0.8 mm Hg and 23.8 ± 1.1 mm Hg, respectively (Fig. 2B, n = 5). The baseline IOP of mice in the AAV-control group was 27.7 ± 0.4 mm Hg, and the IOP values with and without yellow light exposure were 28.0 ± 1.4 mm Hg and 27.7 ± 1.0 mm Hg, respectively (Fig. 2C, n = 3). Figure 2D shows the locations of viral expression and fiber position. Under yellow light stimulation, no significant IOP fluctuated in either the eNpHR3.0 group or the AAV-control group. These results suggested that suppressing the projection neurons within the PVH had no substantial effect on IOP under physiological conditions. 

Figure 2.

 

Optogenetic inhibition of PVH projection neurons produced no substantial decrease in IOP. (A) Schematic of the light administration process. (B) IOP changes of right eye in the eNpHR3.0 group (n = 5 mice, repeated-measures ANOVA; P > 0.05). (C) IOP changes of right eye in control group (n = 5 mice, repeated-measures ANOVA; P > 0.05). (D) Representative images showing virus expression and fiber position. Left: Bilateral PVH expressing AAV-CaMKIIα-eNpHR3.0-mCherry; right: Bilateral PVH expressing AAV-CaMKIIα-mCherry. Scale bar: 200 µm. The white arrow indicates the optical fiber position.

Activation of PVH Glutaminergic (PVH^vglut2+) Neurons Significantly Increases Bilateral IOP

Projection neurons are mainly divided into excitatory and inhibitory neurons, and the chief neurotransmitter released by excitatory neurons is glutamate.^49–51 Therefore we further investigated whether the IOP elevation effect following activation of PVH projection neurons was due to PVH^vglut2+ neurons excitation by injecting AAV-EF1α-DIO-hChR2-mCherry into the PVH of Vglut2-Cre mice. Figure 3B shows the bilateral baseline IOP of both groups. In ChR2 mice, the IOP was elevated during light stimulation, and the IOP fell back when the light stopped (ipsilateral: 16.2 ± 0.6 mm Hg at baseline vs. 29.7 ± 0.6 mm Hg with the light on vs. 19.7 ± 1.4 mm Hg with the light off, n = 5, Figure 3C; contralateral: 16.5 ± 0.3 mm Hg at baseline vs. 27.5 ± 2.3 mm Hg with the light on vs. 18.9 ± 1.3 mm Hg with the light off, n = 5; Fig. 3E). In control mice, there was no significant change on IOP (ipsilateral: 16.1 ± 0.5 mm Hg at baseline vs. 18.1 ± 1.0 mm Hg when the light was turned on vs. 18.0 ± 1.1 mm Hg when the light was turned off, n = 5; Figure 3D. Contralateral: 17.1 ± 1.4 mm Hg at baseline vs. 19.3 ± 1.1 mm Hg when the light was turned on vs. 19.4 ± 1.0 mmHg when the light was turned off, n = 5; Fig. 3F). There was no significant difference between bilateral IOPs after light stimulation (Fig. 3G). 

Figure 3.

 

Optogenetic activation of PVH^vglut2+ neurons induces a rapid increase in IOP. (A) Schematic of the light administration process. (B) Statistical analysis of the baseline IOP of the two groups. (C) IOP changes of the right eye in ChR2 group of Vglut2-Cre mice (n = 5 mice, repeated-measures ANOVA; P < 0.001). (D) IOP changes of the right eye in control group (n = 5 mice, Kruskal-Wallis test; P > 0.05). (E) IOP changes of the left eye in ChR2 group (n = 5 mice, repeated-measures ANOVA; P < 0.001). (F) IOP changes of the left eye in the control group (n = 5, Kruskal-Wallis test; P > 0.05). (G) Comparison of bilateral IOPs following blue light administration (n = 5 mice, paired t test, P < 0.05). (H) c-Fos cell counting in bilateral PVH in the ChR2 group and Control group. (I) Representative images showing virus expression and fiber position. Left: Bilateral PVH expressing AAV-EF1α-DIO-hChR2-mCherry; right: Bilateral PVH expressing AAV-EF1α-DIO- mCherry. Scale bar: 200 µm. The white arrow indicates optical fiber implantation. (J) Microscopy image of three channels in the PVH at high magnification. a: Neurons expressing mCherry; b: Neurons expressing c-Fos; c: Nuclei stained with DAPI; d: Arrow indicates neurons with coexpressing mCherry and c-Fos. Scale bars: 30 µm. The panel represents the control group (a1–d1), the ipsilateral (a2–d2), and contralateral sides (a3–d3) of the ChR2 group, respectively.

Figure 3.

 

Continued.

The mice were perfused to determine viral expression as well as fiber position after IOP measurements (Fig. 3I). Although we injected AAV-EF1α-DIO-hChR2-mCherry in the unilateral PVH, the c-Fos immunofluorescence results showed a large amount of c-Fos expression in the bilateral PVH of the AAV-ChR2 group, whereas little expression was observed in the PVH of the control group (Fig. 3J). At the same time, we counted c-Fos+ neurons in the bilateral PVH of both groups. In the AAV-ChR2 group, the number of labeled neurons was 113 ± 7 in ipsilateral PVH and 107 ± 14 in contralateral PVH. In the AAV-Control group, the c-Fos+ neurons were 21 ± 3 and 20 ± 4 in bilateral PVH, respectively (Fig. 3H). 

Several Nuclei Responded When PVH^vglut2+ Neurons Were Activated

To further determine the mechanism of IOP elevation after activation of PVH^vglut2+neurons, c-Fos immunofluorescence staining was performed on the brain slices of Vglut2-Cre mice with AAV-ChR2. Significant c-Fos⁺ signals were observed in ChR2 mice in the DMH, VMH, locus coeruleus (LC) and basolateral amygdaloid nucleus (BLA), which have been documented to be involved in IOP regulation (Figs. 4A–C). Few fluorescence signals could be observed in the above nuclei in the control mice. The numbers of labeled neurons in the DMH, VMH, LC, and BLA were significantly higher than those in the control mice (720 ± 46 vs. 89 ± 14, 341 ± 59 vs. 42 ± 9, 209 ± 14 vs. 8 ± 4, and 100 ± 8 vs. 40 ± 11, respectively) (P < 0.05, n = 10, Fig. 4D). These results indicated that at least those four IOP-associated nuclei responded when PVH^vglut2+neurons within the PVH were activated. 

Figure 4.

 

C-Fos staining of nuclei after activation of PVH ^vglut2+ neurons. (A–C) c-Fos labeling in the DMH, VMH, LC and BLA. Scale bars: 500 µm. The top panel corresponds to the AAV-ChR2 group, and the bottom row corresponds to the AAV-Control group. (D) The numbers of c-Fos⁺ neurons in the DMH, VMH, LC and BLA (n = 10 mice, independent t test, P < 0.05).

Discussion

Our findings suggest that specific activation of the PVH^vglut2+neurons induce an IOP elevation effect similar to nonspecific activation of PVH projection neurons. The PVH, which is a critical integration site that regulates autonomic and endocrine functions, is composed of magnocellular and parvocellular neurons.¹⁹ Magnocellular neurons are primarily engaged in neuroendocrine regulation^22,52; parvocellular neurons are capable of projecting to autonomic regions such as the brainstem and spinal cord, taking part in the regulation of sympathetic activity.^53,54 Our results showed that activation of unilateral PVH projection neurons significantly increased bilateral IOP, and the extent of IOP elevation of ipsilateral eye was significantly higher than that of contralateral eye after activation of unilateral PVH projection neurons. Recently, we reported that both crossed and uncrossed hypothalamus-pre-parasympathetic and hypothalamus-pre-sympathetic tracts exist in the efferent pathways between the bilateral hypothalamic nuclei (including the PVH, SCN, DMH/PeF, and VTM) and the autonomic innervation of the bilateral AC.¹⁵ Based on the role of autonomic innervation in the AC structures in IOP regulation,¹² our previous results provide anatomical evidence for the possible role of the PVH in IOP regulation via the ANS. These present results provide functional evidence for above hypothesis. The existence of crossed and uncrossed pathways between the bilateral hypothalamic nuclei and the autonomic innervation of the bilateral AC might provide a reasonable explanation for the bilateral effects of IOP elevation following unilateral PVH projection neurons activation. In addition, a lower level of c-Fos expression was observed in the contralateral PVH in the ChR2 group, suggesting that a few contralateral PVH projection neurons were activated concurrently, and it is possible that this also contributed to the increase in contralateral IOP. 

Projection neurons are mainly divided into excitatory and inhibitory neurons, and the chief neurotransmitter released by excitatory neurons is glutamate.^49–51 In the PVH, a large number of neurons are glutamatergic neurons expressing Vglut2.^55,56 As revealed by pharmacological, anatomical and electrophysiological investigations, glutamate is the predominant excitatory neurotransmitter in the PVH, and the excitability of neurons in the sympathetic nervous system is elevated after glutamate binds to their NMDA receptors.^57,58 We further investigated the role of glutaminergic neurons in the PVH on IOP using Vglut2-Cre mice, and found that stimulation of glutamatergic neurons within the PVH induced a considerable increase in IOP. Together with parasympathetic nerve endings from the ciliary ganglion or pterygopalatine ganglion, the sympathetic nerve endings from the superior cervical ganglion innervate the ciliary muscles, blood vessels of the ciliary body, and ciliary processes.^12,59,60 Stimulation of sympathetic nerves can cause the release of neuropeptide Y and adrenergic substances, which could lead to vasoconstriction of the ciliary body and a decrease in the production of aqueous humor.^61–63 The diameter of Schlemm's canal is reduced when the superior cervical ganglion is stimulated, followed by a rise in IOP, suggesting that local sympathetic stimulation can increase IOP.^64–66 Nocturnal IOP is lower, which correlates with a decreased sympathetic tone.⁶⁷ Clinically, drugs of sympathetic inhibition are used to treat glaucoma.⁶⁸ Base on the above evidence, we speculate that the IOP elevation induced by activation of glutamatergic neurons in the PVH may be attributable to elevated sympathetic activity. 

Different from unilateral activation of PVH projection neurons in C57BL/6J mice, there was no significant difference in IOP elevation between the bilateral eyes after activation of PVH^vglut2± neurons in Vglut2-Cre mice. As Vglut2-Cre mice brains are smaller in size, viral infection and fiber optic irradiation have a broader range and are more difficult to restrict to the unilateral PVH. The c-Fos results showed uniform activation in the PVH of the ChR2 group. This may explain why the ipsilateral advantage in IOP elevation does not occur after the activation of PVH^vglut2± neurons in Vglut2-Cre mice. 

To date, the BLA, LC and hypothalamic nuclei including the SCN, DMH/PeF, VMH, SO, and arcuate nucleus have been documented to be involved in IOP regulation.^{16,30–32,69–71} There are abundant fiber projections between the PVH and the above nuclei. The SCN, VMH, SO, and amygdala send inputs to the PVH^33–35,72; the PVH can also send outputs to the LC.⁷³ In this study, we performed whole-brain immunohistochemical staining for c-Fos expression after activation of PVH^vglut2+ neurons. The c-Fos⁺ signals were found in the PVH, DMH, VMH, LC and BLA, but not in the VTM, another hypothalamic nucleus that projects to the AC. It still unclear whether the VTM participates in IOP neuromodulation in the same way as the PVH. This result suggests that the DMH, VMH, LC and BLA may participate in IOP regulation. 

The above results demonstrate that PVH may positively regulate IOP via glutamatergic neurons. However, the IOP elevation effect in C57 mice did not disappear completely as quickly as in Vglut2-Cre mice after the opto-stimulation ended. This suggests that it may be not just glutamatergic neurons involved in PVH IOP regulation activity. 

In the inhibition experiment, the IOP baseline seems higher than others. Actually, the baseline IOPs in the eNpHR3.0 group was similar to that in the AAV-control group. Furthermore, we observed that yellow light stimulation elicited no significant IOP fluctuations in either the eNpHR3.0 group or the AAV-control group. Therefore inhibition of projection neurons in the PVH failed to influence IOP. We considered the following potential possibilities: (1) The PVH contains various types of neurons,⁷⁴ including Gad2 (glutamate decarboxylase)-expressing neurons, which are predominantly GABAergic; GABA is the primary inhibitory neurotransmitter in the central nervous system.^75,76 Therefore it is possible that non-glutamatergic neurons in the PVH, such as GABAergic neurons, negatively regulate IOP. (2) Negative regulation of IOP may be realized via other hypothalamic nuclei, such as the arcuate nucleus, a hypothalamic nucleus that has been documented to decrease bilateral IOP.⁷¹ (3) The PVH is not tonically active, so turning it off optogenetically would not alter IOP. 

It has been reported that activation of glutamate neurons in the PVH can elevate blood pressure (BP).⁷⁷ Although the results of experimental and clinical studies about IOP and BP are often contradictory, positive correlates between IOP and BP are still postulated.^78,79 One limitation of the present study is that our results cannot directly exclude the possibility that the IOP elevation is secondary to BP elevation. However, a secondary IOP elevation should be uniform in bilateral eyes, and the ipsilateral advantage in IOP elevation after unilateral photostimulation of the PVH projection neurons provides the indirect negative evidence to above possibility. 

In conclusion, our work demonstrates that the PVH may positively regulate IOP via glutamatergic neurons mostly. It is new functional evidence for the hypothalamic control of the IOP via the ANS. It has suggested a new direction for the study of the pathogenesis of glaucoma and the development of treatment. 

Acknowledgments

The authors thank AJE (American Journal Experts) for linguistic assistance and presubmission expert review. 

Supported by the National Natural Science Foundation of China (No. 81070727, No. 81670849; Beijing, China). 

Disclosure: L. Ma, None; Q. Liu, None; X. Liu, None; H. Chang, None; S. Jin, None; W. Ma, None; F. Xu, None; H. Liu, None