Activity-dependent secretion of neuronal activity regulated pentraxin from vasopressin neurons into the systemic circulation

Abstract

Neuronal activity regulated pentraxin (Narp) is a secreted, synaptic protein that has been implicated in modulating synaptic transmission. However, it is unclear how Narp secretion is regulated. Since we noted prominent Narp immunostaining in vasopressin neurons of the hypothalamus and in the posterior pituitary, we assessed whether it, like vasopressin, is released into the systemic circulation in an activity-dependent fashion. Consistent with this hypothesis, electron microscopic studies of the posterior pituitary demonstrated that Narp is located in secretory vesicles containing vasopressin. Using affinity chromatography, we detected Narp in plasma and found that these levels are markedly decreased by hypophysectomy. In addition, we confirmed that injection of a viral Narp construct into the hypothalamus restores plasma Narp levels in Narp knockout mice. In checking for activity-dependent secretion of Narp from the posterior pituitary, we found that several stimuli known to trigger vasopressin release, i.e. hypovolemia, dehydration and endotoxin, elevate plasma Narp levels. Taken together, these findings provide compelling evidence that Narp is secreted from vasopressin neurons in an activity-dependent fashion.

Figure 1

Narp immunostaining in rat anterior hypothalamus and pituitary. Co-localization of Narp (red) and vasopressin (green) in the supraoptic (A,A′) and paraventricular nuclei (B,B′) of the rat anterior hypothalamus. (C) Intense staining of Narp terminals in the rat posterior pituitary. The intermediate lobe of the pituitary is essentially devoid of Narp. Scattered Narp cell body staining can be seen in the surrounding anterior pituitary. (D,D′) Narp (red) co-localizes with the alpha-subunit (green) in the anterior pituitary.

Figure 2

Specificity of Narp expression in pituitary. In mouse sections of the posterior pituitary processed for DAB immunohistochemistry, Narp staining is present in a wild-type mouse (A) but absent from a Narp KO (A′). (B) Western blot probed with Narp antibody showing hippocampal lysate and pituitary lysates from a wild type mouse and pituitary lysates from a Narp KO mouse. The native multimeric Narp complex is shown in (B). Addition of β-mercaptoethanol to the samples followed by boiling liberates Narp monomer (B′). Non-reduced bands in the anterior and posterior pituitary of wild-type mice co-migrate with the Narp band from hippocampal lysate. Note that equal amounts of protein were loaded into the anterior and posterior pituitary lanes and that the anterior pituitary Narp band is weaker than the posterior pituitary band indicating lower overall amounts of Narp in the anterior pituitary. In reduced lysates (B′), a duplex band is observed that migrates slightly slower than hippocampal Narp.

Figure 3

Composite of images showing cell processes in the posterior pituitary from a P35 rat double-labeled with 5 nm gold for Narp (short arrows) and 15 nm gold for vasopressin (long arrows). Note the concentration of Narp labeling in large, vasopressin-labeled secretory granules, while there is little or no labeling in small secretory vesicles (dashed circles). Scale bar is 100 nm.

Figure 4

Composite of images showing cells in the anterior pituitary from a P35 rat double-labeled with 5 nm gold for Narp (short arrows) and 10 nm gold for alpha subunit (long arrows). Note the concentration of Narp labeling in large, alpha subunit-labeled secretory granules. Scale bar is 100 nm.

Figure 5

Narp is present in plasma. In (A) Immunoblot of non-reduced lysates from rat plasma showing hippocampal lysate (Hip; 1^st lane); native plasma (2^nd lane); Narp isolated from plasma by taipoxin affinity chromatography (3^rd lane) which co-migrates with Narp from hippocampal lysates. A non-specific band is present in the taipoxin eluate that is also visible in plasma (2^nd lane) and is likely non-reduced IgG recognized by the secondary antibody. Preincubation of the antibody with the GST fusion protein blocks the appearance of the Narp band but not the non-specific band which migrates with a MW of about 220 (4^th lane). A Narp C-terminal antibody also detects Narp in plasma (5^th lane). (B) Detection of Narp in mouse plasma after isolation by taipoxin affinity chromatography. A slowly migrating, non-reduced Narp band is observed which is absent in the Narp KO. In reduced lysate from plasma, a cluster of bands is observed in samples from the wild-type mouse which migrates slightly slower than mouse hippocampal Narp and which is absent from the Narp KO. In the last two lanes we show Narp bands detected from reduced and non-reduced human plasma respectively which co-migrate with mouse Narp bands. The appearance and relative migration of the bands in the 4^th and 6^th lanes compared with the mouse hippocampal band is reminiscent of the Narp duplex band seen in reduced pituitary lysate adding further weight to the contention that Narp is being released into serum from pituitary. C. This panel shows non-reduced plasma lysate from mice aged 4, 8, 12 and 20 weeks processed in parallel.

Figure 6

Plasma Narp is released from pituitary. (A) Non-reduced plasma Narp levels are markedly diminished by hypophysectomy in 5 week old rats compared with age matched controls. Samples were harvested 9 –10 days post-surgery. (B) Plasma Narp is detectable in a Narp KO mouse after it has been injected with AAV-wild type Narp into the paraventricular nucleus of the hypothalamus. The Narp band co-migrates with wild type plasma and hippocampal Narp. The intensity of the IgG band varied across batches of taipoxin beads.

Figure 7

Regulation of plasma Narp. (A) Hypovolemic stress markedly increases levels of plasma Narp (non-reduced) in 5 week old rats compared with rats administered normal saline only. Increases are less robust in older rats (data not shown). (B) Five week old rats that were hypophysectomized do not show an increase in plasma Narp (non-reduced) levels after a hypovolemic stress compared with control hypophysectomized rats that were administered saline only. (C) Mice that have been dehydrated for 24 hours have increased levels of plasma Narp (reduced). (D) Mice treated with endotoxin also show increased levels of plasma Narp (non-reduced) (E) Restraint stress administered to 5 week old rats for either 2 or 3 hours failed to increase plasma Narp levels over control levels.