Johan Ruud

Identification of rat brainstem neuronal structures activated during cancer‐induced anorexia

In cancer‐related anorexia, body weight loss is paradoxically associated with reduced appetite, which is contrary to the situation during starvation, implying that the normal coupling of food intake to energy expenditure is disarranged. Here we examined brainstem mechanisms that may underlie suppression of food intake in a rat model of cancer anorexia. Cultured Morris 7777 hepatoma cells were injected subcutaneously in Buffalo rats, resulting in slowly growing tumor and reduced food intake and body weight loss after about 10 days. The brainstem was examined for induced expression of the transcription factors Fos and FosB as signs of neuronal activation. The results showed that anorexia and retarded body weight growth were associated with Fos protein expression in the area postrema, the general visceral region of the nucleus of the solitary tract, and the external lateral parabrachial nucleus, structures that also display Fos after peripheral administration of satiating or anorexigenic stimuli. The magnitude of the Fos expression was specifically related to the size of induced tumor, and not associated with weight loss per se, because it was not present in pair‐fed or food‐deprived rats. It also appeared to be independent of proinflammatory cytokines, as determined by the absence of increased cytokine levels in plasma and induced cytokine and cyclooxygenase expression in the brain. The findings thus provide evidence that cancer‐associated anorexia and weight loss in this model is associated with activation of brainstem circuits involved in the suppression of food intake, and suggest that this occurs by inflammatory‐independent mechanisms. J. Comp. Neurol. 504:275–286, 2007.

Cyclooxygenase-1 mediates the immediate corticosterone response to peripheral immune challenge induced by lipopolysaccharide.

Immune-induced activation of the hypothalamus-pituitary-adrenal axis is mediated by cyclooxygenase derived prostaglandins. Here we examined the role of cyclooxygenase-1 in this response, by using genetically modified mice as well as pharmacological inhibition. We found that mice with a deletion of the gene encoding cyclooxygenase-1, in contrast to wild type mice, did not show increased plasma corticosterone at 1h after immune challenge by peripheral injection of bacterial wall lipopolysaccharide, whereas the corticosterone levels were similarly elevated in both genotypes at 6h post-injection. Pretreatment of mice with the selective cyclooxygenase-1 inhibitor SC-560, given orally, likewise inhibited the rapid corticosterone response. These findings, taken together with our recent demonstration that the delayed stress hormone response to immune challenge is dependent on cyclooxygenase-2, show that the two cyclooxygenase isoforms play distinct, but temporally supplementary roles for the stress hormone response to inflammation.

Cancer-induced anorexia in tumor-bearing mice is dependent on cyclooxygenase-1.

It is well-established that prostaglandins (PGs) affect tumorigenesis, and evidence indicates that PGs also are important for the reduced food intake and body weight loss, the anorexia-cachexia syndrome, in malignant cancer. However, the identity of the PGs and the PG producing cyclooxygenase (COX) species responsible for cancer anorexia-cachexia is unknown. Here, we addressed this issue by transplanting mice with a tumor that elicits anorexia. Meal pattern analysis revealed that the anorexia in the tumor-bearing mice was due to decreased meal frequency. Treatment with a non-selective COX inhibitor attenuated the anorexia, and also tumor growth. When given at manifest anorexia, non-selective COX-inhibitors restored appetite and prevented body weight loss without affecting tumor size. Despite COX-2 induction in the cerebral blood vessels of tumor-bearing mice, a selective COX-2 inhibitor had no effect on the anorexia, whereas selective COX-1 inhibition delayed its onset. Tumor growth was associated with robust increase of PGE(2) levels in plasma - a response blocked both by non-selective COX-inhibition and by selective COX-1 inhibition, but not by COX-2 inhibition. However, there was no increase in PGE(2)-levels in the cerebrospinal fluid. Neutralization of plasma PGE(2) with specific antibodies did not ameliorate the anorexia, and genetic deletion of microsomal PGE synthase-1 (mPGES-1) affected neither anorexia nor tumor growth. Furthermore, tumor-bearing mice lacking EP(4) receptors selectively in the nervous system developed anorexia. These observations suggest that COX-enzymes, most likely COX-1, are involved in cancer-elicited anorexia and weight loss, but that these phenomena occur independently of host mPGES-1, PGE(2) and neuronal EP(4) signaling.

Inflammation- and tumor-induced anorexia and weight loss require MyD88 in hematopoietic/myeloid cells but not in brain endothelial or neural cells

Loss of appetite is a hallmark of inflammatory diseases. The underlying mechanisms remain undefined, but it is known that myeloid differentiation primary response gene 88 (MyD88), an adaptor protein critical for Toll‐like and IL‐1 receptor family signaling, is involved. Here we addressed the question of determining in which cells the MyD88 signaling that results in anorexia development occurs by using chimeric mice and animals with cell‐specific deletions. We found that MyD88‐knockout mice, which are resistant to bacterial lipopolysaccharide (LPS)‐induced anorexia, displayed anorexia when transplanted with wild‐type bone marrow cells. Furthermore, mice with a targeted deletion of MyD88 in hematopoietic or myeloid cells were largely protected against LPS‐induced anorexia and displayed attenuated weight loss, whereas mice with MyD88 deletion in hepatocytes or in neural cells or the cerebrovascular endothelium developed anorexia and weight loss of similar magnitude as wild‐type mice. Furthermore, in a model for cancer‐induced anorexia‐cachexia, deletion of MyD88 in hematopoietic cells attenuated the anorexia and protected against body weight loss. These findings demonstrate that MyD88‐dependent signaling within the brain is not required for eliciting inflammation‐induced anorexia. Instead, we identify MyD88 signaling in hematopoietic/myeloid cells as a critical component for acute inflammatory‐driven anorexia, as well as for chronic anorexia and weight loss associated with malignant disease.—Ruud, J., Wilhelms, D. B., Nilsson, A., Eskilsson, A., Tang, Y.‐J., Ströhle, P., Caesar, R., Schwaninger, M., Wunderlich, T., Bäckhed, F., Engblom, D., Blomqvist, A. Inflammation‐ and tumor‐induced anorexia and weight loss require MyD88 in hematopoietic/myeloid cells but not in brain endothelial or neural cells. FASEB J. 27, 1973–1980 (2013). www.fasebj.org

Interleukin-6 primarily produced by non-hematopoietic cells mediates the lipopolysaccharide-induced febrile response

Interleukin-6 (IL-6) is critical for the lipopolysaccharide (LPS)-induced febrile response. However, the exact source(s) of IL-6 involved in regulating the LPS-elicited fever is still to be identified. One known source of IL-6 is hematopoietic cells, such as monocytes. To clarify the contribution of hematopoietically derived IL-6 to fever, we created chimeric mice expressing IL-6 selectively either in cells of hematopoietic or, conversely, in cells of non-hematopoietic origin. This was performed by extinguishing hematopoietic cells in wild-type (WT) or IL-6 knockout (IL-6 KO) mice by whole-body irradiation and transplanting them with new stem cells. Mice on a WT background but lacking IL-6 in hematopoietic cells displayed normal fever to LPS and were found to have similar levels of IL-6 protein in the cerebrospinal fluid (CSF) and in plasma and of IL-6 mRNA in the brain as WT mice. In contrast, mice on an IL-6 KO background, but with intact IL-6 production in cells of hematopoietic origin, only showed a minor elevation of the body temperature after peripheral LPS injection. While they displayed significantly elevated levels of IL-6 both in plasma and CSF compared with control mice, the increase was modest compared with that seen in LPS injected mice on a WT background, the latter being approximately 20 times larger in magnitude. These results suggest that IL-6 of non-hematopoietic origin is the main source of IL-6 in LPS-induced fever, and that IL-6 produced by hematopoietic cells only plays a minor role.

Anorexia-cachexia syndrome in hepatoma tumour-bearing rats requires the area postrema but not vagal afferents and is paralleled by increased MIC-1/GDF15

The cancer‐anorexia‐cachexia syndrome (CACS) negatively affects survival and therapy success in cancer patients. Inflammatory mediators and tumour‐derived factors are thought to play an important role in the aetiology of CACS. However, the central and peripheral mechanisms contributing to CACS are insufficiently understood. The area postrema (AP) and the nucleus tractus solitarii are two important brainstem centres for the control of eating during acute sickness conditions. Recently, the tumour‐derived macrophage inhibitory cytokine‐1 (MIC‐1) emerged as a possible mediator of cancer anorexia because lesions of these brainstem areas attenuated the anorectic effect of exogenous MIC‐1 in mice.

Long-term treatment with antidepressants, but not environmental stimulation, induces expression of NP2 mRNA in hippocampus and medial habenula

In experimental settings, antidepressant treatment as well as a stimulating environment has a positive influence on cognition and hippocampal plasticity. One putative mediator of this process is Neuronal Pentraxin 2 (NP2, Narp), known to mediate clustering of glutamatergic AMPA receptors at synapses, and demonstrated to play a role in activity-dependent synaptogenesis and synaptic plasticity. This study demonstrates that NP2 mRNA is robustly expressed in all hippocampal subregions and the medial habenula (MHb), both regions implicated in cognitive functions. Furthermore, NP2 mRNA expression is upregulated in the hippocampal subregions as well as in the MHb after long-term treatment with different antidepressant drugs regardless of monoaminergic profile, suggesting NP2 as a common mode of action of different antidepressant drugs. This effect occurs at the time frame where clinical response is normally achieved. In contrast, neither environmental enrichment nor deprivation has any influence on long-term NP2 mRNA expression. These findings support an involvement of NP2 in the pathway of antidepressant-induced plasticity, but not EE-induced plasticity; that NP2 might constitute a common link for the action of different types of antidepressant drugs and that the MHb could be a putative region for further studies of NP2.

Deletion of the gene encoding MyD88 protects from anorexia in a mouse tumor model

The anorexia-cachexia syndrome, characterized by a rise in energy expenditure and loss of body weight that paradoxically are associated with loss of appetite and decreased food intake, contributes significantly to the morbidity and mortality in cancer. While the pathophysiology of cancer anorexia-cachexia is poorly understood, evidence indicates that pro-inflammatory cytokines are key mediators of this response. Although inflammation hence is recognized as an important component of cancer anorexia-cachexia, the molecular pathways involved are largely unknown. We addressed this issue in mice carrying a deletion of the gene encoding MyD88, the key intracellular adaptor molecule in Toll-like and interleukin-1 family receptor signaling. Wild-type and MyD88-deficient mice were transplanted subcutaneously with a syngenic methylcholanthrene-induced tumor (MCG 101) and daily food intake and body weight were recorded. Wild-type mice showed progressively reduced food intake from about 5days after tumor transplantation and displayed a slight body weight loss after 10days when the experiment was terminated. In contrast, MyD88-deficient mice did not develop anorexia, and displayed a positive body weight development during the observation period. While the MyD88-deficient mice on average developed somewhat smaller tumors than wild-type mice, this did not explain the absence of anorexia, because anorexia was seen in wild-type mice with similar tumor mass as non-anorexic knock-out mice. These data suggest that MyD88-dependent mechanisms are involved in the metabolic derangement during cancer anorexia-cachexia and that innate immune signaling is important for the development of this syndrome.