Endotoxin-induced inflammation down-regulates L-type amino acid transporter 1 (LAT1) expression at the blood-brain barrier of male rats and mice

Abstract

Background: We recently reported that bacterial lipopolysaccharide (LPS)-induced inflammation decreases the expression of the primary thyroid hormone transporters at the blood-brain barrier, organic anion-transporting polypeptide 1c1 (OATP1c1) and monocarboxylate transporter 8 (MCT8). L-type amino acid transporters 1 and 2 (LAT1 & LAT2) are regarded as secondary thyroid hormone transporters, and are expressed in cells of the blood-brain or blood-cerebrospinal fluid barrier and by neurons. The purpose of this study was to examine the effect of LPS-induced inflammation on the expression of LAT1 and LAT2, as these may compensate for the downregulation of OATP1c1 and MCT8.

Methods: LPS (2.5 mg/kg body weight) was injected intraperitoneally to adult, male, Sprague-Dawley rats and C57Bl/6 mice, which were euthanized 2, 4, 9, 24 or 48 h later. LAT1 and LAT2 mRNA expression were studied on forebrain sections using semiquantitative radioactive in situ hybridization. LAT1 protein levels in brain vessels were studied using LAT1 immunofluorescence. Statistical comparisons were made by the non-parametric Kruskal-Wallis and Dunn's tests.

Results: In both species, LAT1 mRNA decreased in brain blood vessels as soon as 2 h after LPS injection and was virtually undetectable at 4 h and 9 h. During recovery from endotoxemia, 48 h after LPS injection, LAT1 mRNA in brain vessels increased above control levels. A modest but significant decrease in LAT1 protein levels was detected in the brain vessels of mice at 24 h following LPS injection. LPS did not affect LAT1 and LAT2 mRNA expression in neurons and choroid plexus epithelial cells.

Conclusions: The results demonstrate that LPS-induced inflammation rapidly decreases LAT1 mRNA expression at the blood-brain barrier in a very similar manner to primary thyroid hormone transporters, while changes in LAT1 protein level follow a slower kinetics. The data raise the possibility that inflammation may similarly down-regulate other blood-brain barrier transport systems at the transcriptional level. Future studies are required to examine this possibility and the potential pathophysiological consequences of inflammation-induced changes in blood-brain barrier transport functions.

Fig. 1

LAT1 mRNA is expressed in brain microvascular cells and neurons. A To confirm that LAT1 mRNA is expressed in multiple cell types, radioactive LAT1 in situ hybridization (A1 darkfield image, silver grain accumulation represents LAT1 mRNA expression) was combined with dual immunofluorescence (A2) for vimentin (red) and HuC/D (green) to visualize the pericyte coverage of vessels and neuronal cell bodies, respectively. The area shown is from the rostral part of the rat hypothalamic supraoptic nucleus (SO). In the merge image (A3), note that silver grains form dense clusters over blood vessels (arrows), while a lesser degree of silver grain accumulation is present over neuronal cell bodies (yellow arrowheads). oc optic chiasm, SO supraoptic nucleus. B, C LAT1 in situ hybridization was co-localized with LAT1 immunofluorescence (red) that labels the vascular endothelium. Note that silver grain clusters that represent high LAT1 mRNA expression are formed over capillaries denoted by LAT1 immunofluorescence (arrows in B3, C3). Images are from the rat hypothalamic ventromedial nucleus (B) and the mouse hypothalamic arcuate nucleus (C). Scale bar 50 µm for all images

Fig. 2

Effect of LPS on LAT1 mRNA expression in brain vessels of rats and mice. A X-ray film images of radioactive LAT1 in situ hybridization in rat forebrain sections. Note the reduced “graininess” of the labeling 9 h after LPS, which is especially visible in the thalamus. Neuronal labeling, well visible in the hippocampus and the hypothalamus, including the dorsomedial nucleus (DMH), is not altered. At 48 h after LPS, the “grainy” signal, which represents blood vessels, is more intense than in control brains over the entire section. Cp choroid plexus, DMH dorsomedial hypothalamic nucleus, Hip hippocampus, MHb medial habenular nucleus. Scale bar 2 mm. B Darkfield emulsion autoradiography images demonstrate the time course of LAT1 mRNA (silver grain accumulation, white) in the rat cortex. In the control cortex, intense hybridization signal is associated with major longitudinal vessels (arrows) and a multitude of small capillaries (numerous bright spots). This labeling vanishes by 4 and 9 h after LPS, only the much lower level parenchymal labeling (primarily neuronal) remains. At 48 h after LPS, hybridization signal labels a larger part of the vasculature than in controls, including major vessels (arrows) and capillaries. Scale bar 100 µm. C LAT1 mRNA in vessels of the mouse cortex. In the control brain, silver grains form clusters over major vessels (arrow) and capillaries, which disappear 9 h after LPS. 48 h after LPS, LAT1 hybridization signal in vessels is more intense than in controls. Arrows indicate major vessels. Scale bar 100 µm

Fig. 3

LAT1 mRNA decreased in vessels but not in neurons after LPS. A, B Darkfield emulsion autoradiography images show the distribution of LAT1 mRNA (silver grain accumulation, white) in the rat hypothalamus. Boxed areas from (A) are shown in higher magnification in (B). Low-level hybridization signal is distributed across the arcuate (Arc) and ventromedial nuclei (VMH), highly resembling a neuronal expression pattern; this signal does not change following LPS administration. Well-defined silver grains clusters over capillaries disappear 9 h after LPS, but become larger and more numerous at 48 h after LPS. 3v third ventricle, Arc, arcuate nucleus, ME median eminence, VMH ventromedial hypothalamic nucleus. Scale bar 200 µm on A; 100 µm on B. C LAT1 in situ hybridization signal in the hypothalamic paraventricular nucleus (PVH) of mice. While LAT1 signal in capillaries virtually disappears 9 h after LPS, the low-level neuronal LAT1 signal in the paraventricular nucleus remains visible. At 48 h after LPS, hybridization signal in vessels is markedly increased. 3v third ventricle, PVH hypothalamic paraventricular nucleus; Scale bar 100 µm

Fig. 4

Semiquantitative image analysis results of LAT1 in situ hybridization and LAT1 immunofluorescence. A LAT1 hybridization signal was quantified in the rat cortex. Values represent area covered by vascular LAT1 hybridization signal. B, C LAT1 immunofluorescence was quantified in the rat and mouse hypothalamus, ventro-lateral to the hypothalamic paraventricular nucleus. Values represent the average brightness value of the immunofluorescence. Groups were compared by the non-parametric Kruskal–Wallis test and then by Dunn’s multiple comparison test. *P < 0.05; **P < 0.01; vs control; Samples sizes are n = 4 or 5 rats or mice/group

Fig. 5

LAT1 protein is expressed in cerebral endothelial cells. A1–A3 Dual immunofluorescence for the endothelial marker CD31 (green) and LAT1 (red) in the mouse brain. LAT1 always colocalized with CD31 in vessels. Some vessels that were labeled for CD31 apparently lacked or had only little LAT1 protein expression (arrowhead). Dual immunofluorescence for the endothelial marker GLUT1 (green) and LAT1 (red) and in the mouse (B1–B3) and the rat (C1–C3) brain. LAT1 always colocalized with GLUT1 in vessels. A fraction of vessels labeled for GLUT1 were not labeled for LAT1 (arrowheads). Scale bar 50 µm

Fig. 6

LAT1 immunofluorescence in brain vessels of the rat and mouse hypothalamus. A LAT1 immunofluorescence in the rat hypothalamus at the level of the paraventricular nucleus (PVH). The intensity of labeling tended to be lower at 24 h after LPS administration. Scale bar 200 µm. B LAT1 immunofluorescence in hypothalamic vessels of mice; labeling intensity is modestly reduced at 24 and 48 h after LPS. Scale bar 100 µm

Fig. 7

LPS had no effect on LAT2 mRNA expression in the rat and mouse brain. A X-ray film images of radioactive LAT2 in situ hybridization from the rat forebrain. LAT2 is expressed in several neuronal populations and the choroid plexus, but its expression did not change after LPS administration. BLA basolateral amygdaloid nucleus, cp choroid plexus, PVH hypothalamic paraventricular nucleus. Scale bar 2 mm. B Darkfield emulsion autoradiography images from the rat cortex show no change in neuronal LAT2 mRNA expression after LPS. Scale bar 200 µm. C Similarly, no change in LAT2 hybridization signal was observed in neurons of the basolateral amygdaloid nucleus in mice. Scale bar 100 µm