Subcutaneous administration of β-hydroxybutyrate improves learning and memory of sepsis surviving mice

Abstract

Post-sepsis cognitive impairment is one of the major sequelae in sepsis survivors. Its prevention remains clinically challenging. Here we tested the effects and underlying mechanisms of exogenous β-hydroxybutyrate (BHB) on post-sepsis cognitive impairment. We found that subcutaneous administration of BHB increased survival and body weight recovery of sepsis mice and improved learning and memory of sepsis surviving mice in a cecal ligation and perforation-induced sepsis model. Additionally, the improvement of learning and memory of sepsis surviving mice was still detected even if BHB was administrated at the late stage of sepsis. In contrast, glucose solution did not show similar effects. Mechanistically, subcutaneous administration of BHB increased the BHB level of hippocampus, and limited neuroinflammation and neuroplasticity damage in sepsis mice. Intracerebroventricular administration of BHB also alleviated neuroinflammation and cognitive impairment of sepsis surviving mice. In the coculture of neurons, astrocytes, and BV2 cells (a microglial cell line), knocking down the expression of microglial HCA2 (BHB receptor) via a specific shRNA reduced the protection of BHB to lipopolysaccharide-induced inflammatory response and neuron damage more significantly than knocking down neuronal MCT2 (BHB transporter). These data showed that (1) BHB was a potential pharmacological adjunct treatment for prevention of post-sepsis cognitive impairment and (2) inhibiting neuroinflammation via HCA2 was an important mechanism.

Keywords: HCA2; Inflammation; MCT2; Sepsis; Sepsis associated encephalopathy; β-hydroxybutyrate.

Fig. 1

β-hydroxybutyrate(BHB) administration promotes survival and body weight recovery of CLP mice and improves cognitive dysfunction of CLP surviving mice (A) The survival of each group during the first 9 days after CLP surgery. (B) The body weight recovery of each group during the first 11 days after CLP surgery. *p =0.0003 vs CLP+NS group. (C) In the novel object recognition test 14−15 days after CLP, the time of the novel object of the CLP+NS and CLP+Glu groups were significantly less than that of the sham+NS group, but not of the CLP+BHB and CLP+BHB(L) groups (p>0.05). *p<0.05 vs sham + NS group. ns: p>0.05 vs sham + NS group. (D) In the Barnes maze test 16−19 days after CLP, compared to the CLP+NS group, the error number in CLP+BHB (p=0.002) and CLP+BHB(L) (p=0.045) groups decreased significantly, but the CLP+Glu (p=0.657) group did not. *p<0.05 vs LPS+NS group. N = 9-13/group. The data are expressed as the mean±SEM. sham+NS= sham + saline injection group. CLP+NS= CLP surgery+ saline injection group. CLP+BHB = CLP surgery+ BHB injection group. CLP+BHB (L)= CLP surgery + BHB injection of late stage group. CLP+Glu = CLP surgery+ glucose solution injection group. CLP= cecal ligation and perforation surgery

Fig. 2

Subcutaneous administration of BHB increased the BHB level of blood and brain of CLP mice (A) The BHB concentration in the hippocampus of CLP mice on the 2^nd and 4^th days after CLP surgery. *p< 0.05. (B) The BHB concentration of blood plasma in CLP mice on the 2^nd and 4^th days after CLP surgery. *p< 0.05. N = 3-5/time point/group. The data are expressed as the mean±SEM. sham+NS= sham + saline injection group. CLP+NS= CLP surgery+ saline injection group. CLP+BHB = CLP surgery+ BHB injection group. CLP+BHB-L= CLP surgery + BHB injection of late stage group. CLP+Glu = CLP surgery+ glucose solution injection group. CLP= cecal ligation and perforation surgery

Fig. 3

BHB administration improved neuroplasticity in CLP surviving mice (A) The representative images and quantitative analysis of doublecortin positive cells in the dentate gyrus of CLP mice 12 days after CLP surgery. *p< 0.05. N = 4/group. Bar=50 μm. (B) and (C) Representative Western blot analysis of synaptophysin and PSD-95 in the hippocampus 12 days after CLP surgery. * P< 0.05. N = 3/group. The data are expressed as the mean±SEM. sham+NS= sham + saline injection group. CLP+NS=CLP surgery+ saline injection group. CLP+BHB = CLP surgery+ BHB injection group. CLP+Glu = CLP surgery+ glucose solution injection group. CLP= cecal ligation and perforation surgery

Fig. 4

BHB administration limited neuroinflammationin and peripheral inflammation in CLP mice (A) Sepsis-induced changes of the leukocyte number and neutrophil percentage of peripheral blood were significantly limited by BHB. *p< 0.05. N = 4-5/time point/group. The data are expressed as the mean±SEM. (B) Levels of the inflammatory factors IL-1β and TNF-α in the hippocampus. *p< 0.05. N = 4-5/time point/group. (C) The representative images of Iba1 staining and a quantitative analysis of activated microglial cells in CA1 and the dentate gyrus of CLP mice 12 days after CLP surgery. *p < 0.05. Bar = 50 μm. sham+NS = sham + saline injection group. CLP+NS= CLP surgery+ saline injection group. CLP+BHB = CLP surgery+ BHB injection group. CLP+Glu = CLP surgery+ glucose solution injection group. CLP = cecal ligation and perforation surgery

Fig. 5

Intracerebroventricular administration of BHB improves cognitive dysfunction and limits neuroinflammation in CLP mice (A) and (B) Levels of the inflammatory factors IL-1β and TNF-α in the hippocampus 12 days after CLP . *p< 0.05. N = 4/group. The data are expressed as the mean±SEM. (C) and (D) In the novel object recognition test 11-12 days after CLP, the time of the novel object of the CLP+NS and CLP+Glu groups were significantly less than the CLP +BHB group during the test phase, although there was no intergroup difference during the familiarization phase. *p<0.05. N = 10−14 / group. The data are expressed as the mean±SEM. ns=no significance. sham+NS = sham + saline injection group. CLP+NS= CLP surgery+ saline injection group. CLP+BHB = CLP surgery+ BHB injection group. CLP+Glu = CLP surgery+ glucose solution injection group. CLP = cecal ligation and perforation surgery

Fig. 6

BHB administration limited LPS-induced neuron damage and inflammatory response via HCA2 and MCT2 in the coculture of neurons, astrocytes, and BV2 (A) A quantitative RT-PCR analysis showed that BHB inhibited the LPS-induced changes of the inflammatory markers Iba1 and TNF-α in microglial cells in a dose-dependent manner. *p< 0.05. N = 3/group. (B) Neuronal damage was marked by the levels of medium LDH and neuronal CCK-8. LPS induced obvious neuron damage, which was significantly limited by BHB. *p< 0.05. N =3-4/group. (C) BHB alleviated LPS-induced inflammatory response. Additionally, knocking down the levels of HCA2 of microglial cells and MCT2 of neurons both obviously blocked the protective effects of BHB on the inflammatory response. *p< 0.05. N =3-4/group. The data are expressed as the mean±SEM. NS= normal saline. LPS= lipopolysaccharides. HCA2 shRNA = lentivirus (LV) containing shRNA plasmid targeting HCA2. MCT2 shRNA= lentivirus (LV) containing shRNA plasmid targeting MCT2