Yehuda Shavit

Illness, Cytokines, and Depression

Abstract: Various medical conditions that involve activation of the immune system are associated with psychological and neuroendocrine changes that resemble the characteristics of depression. In this review we present our recent studies, designed to investigate the relationship between the behavioral effects of immune activation and depressive symptomatology. In the first set of experiments, we used a double‐blind prospective design to investigate the psychological consequences of illness in two models: (1) vaccination of teenage girls with live attenuated rubella virus, and (2) lipopolysaccharide (LPS) administration in healthy male volunteers. In the rubella study, we demonstrated that, compared to control group subjects and to their own baseline, a subgroup of vulnerable individuals (girls from low socioeconomic status) showed a significant virus‐induced increase in depressed mood up to 10 weeks after vaccination. In an ongoing study on the effects of LPS, we demonstrated significant LPS‐induced elevation in the levels of depression and anxiety as well as memory deficits. These psychological effects were highly correlated with the levels of LPS‐induced cytokine secretion. In parallel experiments, we demonstrated in rodents that immune activation with various acute and chronic immune challenges induces a depressive‐like syndrome, characterized by anhedonia, anorexia, body weight loss, and reduced locomotor, exploratory, and social behavior. Chronic treatment with antidepressants (imipramine or fluoxetine) attenuated many of the behavioral effects of LPS, as well as LPS‐induced changes in body temperature, adrenocortical activation, hypothalamic serotonin release, and the expression of splenic TNF‐α mRNA. Taken together, these findings suggest that cytokines are involved in the etiology and symptomatology of illness‐associated depression.

Cytokines, “Depression Due to A General Medical Condition,” and Antidepressant Drugs

Activation of the immune system during various medical conditions produces neural, neuroendocrine, and behavioral effects. The psychological and physiological effects of immune activation resemble many characteristics of depression. The essential features of depression are depressed mood and loss of interest or pleasure in all, or almost all activities (anhedonia). Several associated symptoms are also present, including, appetite disturbance, change in body weight, sleep disturbance, psychomotor disturbance, fatigue, loss of energy, and difficulty in thinking or concentrating (DSM-IV, 1994). Depression is also characterized by specific alterations in the functioning of neurochemical and neuroendocrine systems, including monoaminergic systems and the hypothalamic-pituitary-adrenal (HPA) axis (Brown, Steinberg, & van Praag, 1994; Holsboer, 1995). Most of these psychological and neuroendocrine symptoms appear both in humans and animals during diseases that involve immune activation. Based on these findings, and on several additional lines of evidence that will be presented below, we have recently argued that immune activation is involved in the etiology and symptomatology of depression associated with various medical conditions (Yirmiya, 1997).

Effects of anesthesia based on large versus small doses of fentanyl on natural killer cell cytotoxicity in the perioperative period.

Surgical stress and general anesthesia suppress immune functions, including natural killer cell cytotoxicity (NKCC).This suppression could be attributable, at least in part, to opiates. We have previously shown that large-dose fentanyl administration suppressed NKCC in rats. The present study sought to compare the effects of two anesthetic protocols, based on large-(LDFA) versus small (SDFA)-dose fentanyl anesthesia on NKCC in the perioperative period. Forty patients were included in this study; half were assigned to each protocol of anesthesia. In each anesthetic group, half the patients were undergoing surgery for malignant diseases, and half for benign conditions. Blood samples were collected during the perioperative period. NKCC was assessed using the chromium release assay. Initially, both types of anesthesia similarly suppressed NKCC, with a peak effect 24 h after surgery. The two types of anesthesia, however, differed in the rate of recovery of NKCC suppression. By the second postoperative day, NKCC returned to control values in the SDFA patients, whereas NKCC was still significantly suppressed after LDFA. These results indicate that LDFA causes prolonged suppression of NK cell function. Whether this suppression might have a long-term impact on the overall outcome, especially in cancer patients, remains to be determined. (Anesth Analg 1996;82:492-7)

The Effects of Postoperative Pain Management on Immune Response to Surgery

Surgery is associated with immune alterations, which are the combined result of tissue damage, anesthesia, postoperative pain, and psychological stress. In the present study, we compared the effects of several postoperative pain management techniques on postoperative immune function. Patients hospitalized for abdominal surgery were randomly assigned to one of three postoperative pain management techniques: opiates on demand (intermittent opiate regimen [IOR]), patient-controlled analgesia (PCA), and patient-controlled epidural analgesia (PCEA). Postoperative pain was assessed. Blood samples were collected before and 24, 48, and 72 h after surgery. Production of interleukin (IL)-1&bgr;, IL-2, and IL-6, natural killer cell cytotoxicity, and lymphocyte mitogenic responses were assessed. Patients of the PCEA group exhibited lower pain scores in the first 24 h after surgery compared with patients of the IOR and PCA groups. Mitogenic responses were suppressed in all groups in the first 24 h, returned to preoperative values by 72 h in the PCEA group, but remained suppressed in the PCA group. Production of IL-1&bgr; and IL-6 increased in the IOR and PCA groups, whereas it remained almost unchanged in the PCEA group. Patients receiving an epidural mixture of opiate and local anesthetics (PCEA group) exhibited reduced suppression of lymphocyte proliferation and attenuated proinflammatory cytokine response in the postoperative period.

Interleukin-1 antagonizes morphine analgesia and underlies morphine tolerance.

Pain sensitivity reflects a balance between pain facilitatory and inhibitory systems. To characterize the relationships between these systems we examined the interactions between the analgesic effects of morphine and the anti‐analgesic effects of the pro‐inflammatory cytokine interleukin‐1 (IL‐1). We report that administration of a neutral dose of IL‐1β abolished morphine analgesia in mice, whereas acute or chronic blockade of IL‐1 signaling by various IL‐1 blockers (IL‐1 receptor antagonist (IL‐1ra), α‐melanocyte‐stimulating hormone, or IL‐1 tri‐peptide antagonist) significantly prolonged and potentiated morphine analgesia. Morphine‐induced analgesia was also extended in strains of mice genetically impaired in IL‐1 signaling (mice with transgenic over‐expression of IL‐1 receptor antagonist, deletion of the IL‐1 receptor type I, or deletion of the IL‐1 receptor accessory protein). The finding that IL‐1 produces a marked anti‐analgesic effect, suggests that it may also be involved in the development of opiate tolerance. Indeed, genetic or pharmacological blockade of IL‐1 signaling prevented the development of tolerance following repeated morphine administration. Moreover, acute administration of IL‐1ra in wild type mice, either immediately following the cessation of acute morphine analgesia, or following the development of chronic morphine tolerance, re‐instated the analgesia, suggesting that blockade of the IL‐1 system unmasks the analgesic effect of morphine. These findings suggest that morphine produces an IL‐1‐mediated homeostatic response, which serves to limit the duration and extent of morphine analgesia and which underlies the development of tolerance.

Genetic impairment of interleukin-1 signaling attenuates neuropathic pain, autotomy, and spontaneous ectopic neuronal activity, following nerve injury in mice.

Abstract Peripheral nerve injury may lead to neuropathic pain, which is often associated with mechanical and thermal allodynia, ectopic discharge of from injured nerves and from the dorsal root ganglion neurons, and elevated levels of proinflammatory cytokines, particularly interleukin‐1 (IL‐1). In the present study, we tested the role of IL‐1 in neuropathic pain models using two mouse strains impaired in IL‐1 signaling: Deletion of the IL‐1 receptor type I (IL‐1rKO) and transgenic over‐expression of the IL‐1 receptor antagonist (IL‐1raTG). Neuropathy was induced by cutting the L5 spinal nerve on one side, following which mechanical and thermal pain sensitivity was measured. Wild‐type (WT) mice and the parent strains developed significant allodynia and hyperalgesia in the hind‐paw ipsilateral to the injury compared with the contralateral hind‐paw. The mutant strains, however, did not display decreased pain threshold in either hind‐paw. Pain behavior was also assessed by cutting the sciatic and saphenous nerves and measuring autotomy scores. WT mice developed progressive autotomy, beginning at 7 days post‐injury, whereas the mutant strains displayed delayed onset of autotomy and markedly reduced severity of the autotomy score. Electrophysiological assessment revealed that in WT mice a significant proportion of the dorsal root axons exhibited spontaneous ectopic activity at 1, 3, and 7 days following spinal nerve injury, whereas in IL‐1rKO and IL‐1raTG mice only a minimal number of axons exhibited such activity. Taken together, these results suggest that IL‐1 signaling plays an important role in neuropathic pain and in the altered neuronal activity that underlies its development.

Effects of Fentanyl on Natural Killer Cell Activity and on Resistance to Tumor Metastasis in Rats

Objectives: Opiates, which serve an integral role in anesthesia, suppress immune function, particularly natural killer cell cytotoxicity (NKCC). NK cells play an important role in tumor and metastasis surveillance. We reported that large-dose fentanyl anesthesia induced prolonged suppression of NKCC in patients undergoing abdominal surgery. The immune modulatory effects of opiates may depend on the interaction between dose and time of administration. The present study examined the effects of different doses of fentanyl, administered at different time points relative to tumor inoculation, on NKCC and on experimental tumor metastasis in rats. Methods: Fischer 344 rats were injected with low or high doses of fentanyl, 6 or 2 h before, simultaneously with or 1 h after being inoculated intravenously with MADB106 tumor cells. Lung tumor retention (LTR) was assessed 4 h after, and lung tumor metastases were counted 3 weeks after tumor inoculation. NKCC was assessed 1 h after the fentanyl injection. Results: At all time points, except 6 h before tumor inoculation, fentanyl (0.1–0.3 mg/kg) induced a dose-dependent increase in MADB106 LTR (2.3- to 74-fold). An intermediate dose of fentanyl (0.15 mg/kg) doubled the number of lung metastasis, and, within animal, suppressed NKCC and increased MADB106 LTR in a correlated manner. Conclusion: These findings indicate that fentanyl suppresses NKCC and increases the risk of tumor metastasis. Suppression of NK cells at a time when surgery may induce tumor dissemination can prove to be critical to the spread of metastases. It is suggested that the acute administration of a moderate dose of opiates during surgery should be applied cautiously, particularly in cancer patients.

Effects of Preemptive Analgesia on Pain and Cytokine Production in the Postoperative Period

Background The postoperative period is associated with increased production of proinflammatory cytokines, which are known to augment pain sensitivity, among other effects. In a previous study, the authors found that patients treated with patient-controlled epidural analgesia (PCEA) exhibited attenuated proinflammatory cytokine response in the postoperative period. In the present study, the authors examined whether preemptive analgesia continued with PCEA may further attenuate the proinflammatory cytokine response and reduce pain sensitivity in the postoperative period. They compared cytokine production in two groups of patients, one receiving PCEA, the other receiving preemptive epidural analgesia continued by PCEA. Methods Female patients hospitalized for transabdominal hysterectomy were randomly assigned to one of two pain management techniques: PCEA or preemptive epidural analgesia followed by PCEA (PA + PCEA). Postoperative pain was assessed using the visual analog scale. Blood samples were collected before, 24, 48, and 72 h following surgery. Production of the following cytokines was assessed ex vivo in stimulated peripheral blood mononuclear cells: interleukin (IL)-1&bgr;, tumor necrosis factor &agr;, IL-6, IL-1ra, IL-10, and IL-2. Results Patients of the PA + PCEA group exhibited lower pain scores throughout the 72 h postoperatively, compared with patients of the PCEA group. In patients of the PA + PCEA group in the postoperative period, production of IL-1&bgr;, IL-6, IL-1ra, and IL-10 was significantly less elevated, while IL-2 production was significantly less suppressed. Conclusions Proinflammatory cytokines are key mediators of illness symptoms, including hyperalgesia. The present results suggest that preemptive epidural analgesia is associated with reduced postoperative pain and attenuated production of proinflammatory cytokines.

Impairment of interleukin-1 (IL-1) signaling reduces basal pain sensitivity in mice: genetic, pharmacological and developmental aspects

The cytokine interleukin‐1 (IL‐1) has been implicated in modulation of pain perception under various inflammatory conditions. The present study examined the hypothesis that IL‐1 signaling is also involved in pain sensitivity under normal, non‐inflammatory states, using three mouse models of impaired IL‐1 signaling: targeted deletion of the IL‐1 receptor type I or the IL‐1 receptor accessory protein, and transgenic over‐expression of IL‐1 receptor antagonist within the brain and spinal cord. Thermal and mechanical pain sensitivity was assessed using the paw‐flick, hot‐plate, and von Frey tests. All mutant strains displayed significantly lower pain sensitivity, compared with their respective wild‐type control strains, and with their parent strains (C57BL/6, CBA and 129), in all tests. In contrast, mice with targeted deletion of the p55 or p75 TNF receptor, or of interleukin‐18, displayed normal or higher pain sensitivity compared to their respective controls. To differentiate between developmental vs. on‐going effects of IL‐1, mice were chronically treated with IL‐1 receptor antagonist (IL‐1ra) via osmotic micropumps, either in adulthood or prenatally (throughout the last 2 weeks of gestation). Adult mice that were treated with IL‐1ra either in adulthood or in utero, displayed lower pain sensitivity, similar to mice with impaired IL‐1 signaling. These findings suggest that basal pain sensitivity is genetically, developmentally and tonically influenced by IL‐1 signaling.

Stimulation of the hypothalamic paraventricular nucleus produces analgesia not mediated by vasopressin or endogenous opioids

The analgesic effect of electrical stimulation of the hypothalamic paraventricular nucleus (PVN) was studied. Additionally, the involvement of vasopressin and opioid peptides in this process was examined by comparing vasopressin-deficient (Brattleboro) and Long-Evans rats and by administering the opiate antagonist naloxone. Rats were chronically implanted with a stimulating electrode in the parvocellular (PVN-Pc) and magnocellular (PVN-Mg) divisions of the PVN. At least 10 days after surgery, the analgesic effects of PVN stimulation were examined in lightly anesthetized rats, using the tail-flick method, and in unanesthetized rats, using the hot-plate test. PVN stimulation produced marked analgesia in both tests. Current threshold for analgesia was lower from PVN-Pc than from PVN-Mg. Threshold did not differ significantly between Brattleboro and Long-Evans rats and was not affected by naloxone administration. The results indicate that the PVN is part of the brains pain inhibitory system, and show that the analgesia induced by PVN stimulation is not mediated by either vasopressin or opioid peptides.