Tumor necrosis factor-α triggers a cytokine cascade
yielding postoperative cognitive decline
Niccolò Terrandoa,b, Claudia Monacoc, Daqing Mab, Brian M. J. Foxwellc,1, Marc Feldmannc,2, and Mervyn Mazea,b,2
aDepartment of Anesthesia and Perioperative Care, University of California, San Francisco, CA 94143-0648;bDepartment of Anesthetics, Pain Medicine
and Intensive Care, Imperial College London, Chelsea and Westminster Hospital, London SW10 9NH, United Kingdom; andcKennedy Institute of
Rheumatology, Faculty of Medicine, Imperial College London, London W6 8LH, United Kingdom
Contributed by Marc Feldmann, September 30, 2010 (sent for review August 2, 2010)
Cognitive decline following surgery in older individuals is a major
clinical problem of uncertain mechanism; a similar cognitive decline
also follows severe infection, chemotherapy, or trauma and is
currently without effective therapy. A variety of mechanisms have
beenproposed,andexploring the role ofinflammation,werecently
reported the role of IL-1β in the hippocampus after surgery in mice
is upstream of IL-1 and provokes its production in the brain. Periph-
eral blockade of TNF-α is able to limit the release of IL-1 and pre-
vent neuroinflammation and cognitive decline in a mouse model of
surgery-induced cognitive decline. TNF-α appears to synergize with
gest a unique therapeutic potential for preemptive treatment with
anti-TNF antibody to prevent surgery-induced cognitive decline.
innate immunity|surgical complications|delirium|dementia
disease states, and determine whether the disease resolves or
the function of the CNS, including memory and cognition (5–7).
Critical illness and postoperative recovery are often associated with
cognitive decline, including memory dysfunction, especially in the
elderly, a segment of the population that is increasing rapidly (8).
Fluctuations in cognitive function and early-onset impairment in
memory abilities are frequently observed in the intensive care unit.
With an incidence ranging between 28 and 92% in hospitalized
medical patients, depending on age, patient comorbidity, and the
type of surgery, postoperative delirium is frequently diagnosed with
bedside clinical testing (confusion assessment method) (9, 10). Al-
though thisacute confusionstate istypically limited induration and
potentially reversible, postoperative delirium associates with signif-
cognitive decline that is detected through a battery of neuro-
psychological tests (14). After major noncardiac surgery, POCD
occurs in 7 to 26% of patients, and isindependentlyassociated with
poor short-term and long-term outcomes, including an increased
risk of mortality, inability to cope independently, premature un-
employment, and possible permanent dementia (15–17). In this
of surgery in the elderly) to extend our earlier laboratory studies of
postoperative cognitive decline, documenting the innate immune
nflammation triggered by innate mechanisms plays a pivotal
Early Increase in Systemic TNF and High-Mobility-Group Box Chro-
mosomal Protein-1. Following surgery under general anesthesia,
TNF-α wasthe firstcytokine tobe releasedandpeaked(P<0.05)at
30 min after surgery (Fig. 1A); in contrast, other proinflammatory
cytokines, such as IL-1β and IL-6, were not detected until 6 h post-
surgery (18). To assess whether the inflammatory response was pre-
ceded by systemic release of damage-associated molecular-pattern
(DAMPs) molecule, we measured high-mobility group-box chro-
mosomal protein 1 (HMGB-1), which isreleased after cell necrosis
(19). HMGB-1 blood levels were increased at 1 h (P < 0.05) and
peaked at 6 h (P < 0.001) before returning to baseline (Fig. 1B).
Significant differences in cytokine kinetics could be observed fol-
lowing infection (endotoxemia) (20) (Fig. S1). No cytokine or
DAMP changes were observed following general anesthesia only
(Fig. 1 C and D).
Anti-TNF Prophylaxis Prevents Neuroinflammation and Cognitive
Decline. In a number of systems TNF triggers the production of
proinflammatory cytokines, reported first in mouse “sepsis” (21)
and human rheumatoid arthritis (22). We next investigated the
effects of TNF-α blockade on systemic cytokine levels, neuro-
inflammation, and cognitive dysfunction. Preoperative adminis-
tration of anti-TNF effectively reduced the amount of systemic
IL-1β both at 6 and 24 h following surgery (P < 0.01, P < 0.001)
(Fig. 2 A and B). To corroborate the findings and ascertain the
kinetics and specificity of TNF-α blockade, we delayed the in-
jection of the antibody until 1 h after surgery and then measured
levels of IL-1β and IL-6; administration at this time with TNF-α
blockade had no effect (Fig. 2 A and B).
Although IL-1β is pivotal for hippocampal learning and mem-
ory, high levels can interfere with long-term potentiation and
synaptic plasticity (23). Prophylaxis with anti-TNF antibody at-
(Fig. 2C) (P < 0.01). Microglia, the innate immune cells of the
CNS, usually reside in the quiescent state; these cells are tightly
regulated and, upon activation, release cytotoxic compounds that
disrupt homeostatic processes and neuronal functions (24–26).
bodies and thin, long, ramified pseudopodia into an amoeboid
morphology,with enlargement ofthecell body(features described
To relate the inflammatory response to cognitive behavior, we
usedtrace fearconditioning inwhich miceare trainedtoassociate
a tone with a noxious foot-shock stimulation (27). Contextual fear
response shows reduced immobility (freezing) at postoperative
day 3, revealing hippocampal-dependent memory impairment
(Fig. 2E) (P < 0.05). Pretreatment with anti-TNF significantly
ameliorated this cognitive dysfunction (P < 0.05).
Author contributions: N.T., C.M., D.M., B.M.J.F., M.F., and M.M. designed research; N.T.
performed research; N.T., C.M., D.M., M.F., and M.M. analyzed data; and N.T., C.M., D.M.,
M.F., and M.M. wrote the paper.
The authors declare no conflict of interest.
1Deceased December 16, 2008.
2To whom correspondence may be addressed. E-mail: firstname.lastname@example.org or
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| November 23, 2010
| vol. 107
| no. 47www.pnas.org/cgi/doi/10.1073/pnas.1014557107
MyD88 Mediates Cognitive Decline Following Surgery. We had pre-
viously shown that IL-1 perpetuates the inflammatory response
in our model, and is prevented by blockade of IL-1 by IL-1 re-
ceptor antagonist pretreatment (18). To assess the relationship
between the TNF- and the IL-1-dependent pathways, we studied
the effects of surgery in mice lacking MyD88, the key signaling
by ELISA were increased following tib-
ial surgery. (A) Plasma levels of TNF-α
were significantly increased after 30
up-regulated after 1 h, peaking at 6 h,
and returning to baseline thereafter.
Plasma levels of both TNF-α (C) and
HMGB-1 (D) remained at baseline fol-
lowing exposure to general anesthe-
sia (2.1% isoflurane) and analgesia
(buprenorphine, 1 mg/kg s.c.). Results
are expressed as mean ± SEM (n = 6).
*, P < 0.05; **, P < 0.001 versus na-
ive by one-way ANOVA followed by
Student-Newman-Keuls test. ND, not
TNF-α and HMGB-1 measured
phylaxis on systemic cytokines, neu-
roinflammation and cognitive be-
havior. (A) Systemic IL-1β and (B) IL-6
levels following anti-TNF administra-
tion 18 h preoperatively. Delayed ad-
ministration (1 h) of the antibody did
not provide any reduction in plasma
cytokines. (C) Hippocampi were ex-
preemptive administration of anti-
TNF resulted in no changes in expres-
sion of IL-1β compared with naive
mice. (D) Densitometry of microglial
microgliosis compared withnaive and
surgical mice treated with anti-TNF.
(E) Mice subjected to surgery exhi-
bited reduced freezing to context
when compared with naive mice;
preoperative administration (18 h) of
anti-TNF mitigated the contextual
fear memory impairment. Error bars
represent the means ± SEM (n = 6, n =
10 for acute behavior). *, P < 0.05; **,
P < 0.01; ***, P < 0.001 by repeated
measures ANOVA followed by Stu-
Wallis followed by the Dunn’s multi-
ple comparison test was used for
categorical data. Ab, antibody; D, 1 h
delayed administration of antibody;
N + Ab, naive + Ab; S, surgery.
2. Effectsof anti-TNFpro-
Terrando et al.PNAS
| November 23, 2010
| vol. 107
| no. 47
adaptor of the IL-1 and TLR superfamily, but not involved in
TNF signaling (28). Compared with wild-type, the systemic cy-
tokine response to surgery was reduced, but not obliterated, in
MyD88−/−both at 6 and 24 h (Fig. 3 A and B) (P < 0.01, P <
0.01). However, when MyD88−/−were administered with anti-
TNF antibody, surgery-induced incremental change in IL-1β and
IL-6 were abrogated, although TNF-α levels were unaffected
(Fig. 3 A and B). Neither signs of neuroinflammation (Fig. 3 C
and D and Fig S2) nor a reduction in freezing behavior was
observed in MyD88−/−following surgery (Fig. 3E).
Anti-TNF Prevents Surgery-Induced Inflammation in Tlr4−/−Mice.
Toll-like receptor 4 (TLR4) is involved in a large proportion of
known danger signals, and more are emerging, such as tenascin-C
in inflamed joints (29). Having established pivotal roles for TNF-α
and MyD88 signaling in surgery-induced inflammation and cogni-
tive dysfunction, we targeted TLR4 to understand whether this
receptor could account for a MyD88-dependent signal in POCD
(30). Surgery in Tlr4−/−induced similar systemic and central in-
indicating that TLR-4 signaling is not required for establishment
of cognitive decline (Fig. 4A) (P < 0.05). When Tlr4−/−mice were
response, hippocampal inflammation and memory impairment
were abrogated (Fig. 4 B–E).
Postoperative cognitive decline is a poorly understood disorder,
very common even after noncardiac surgery, which can be mim-
ways and these results, taken together, demonstrate that TNF-α is
very important in this process. TNF-α acts upstream of IL-1β
and initiates the peripheral cytokine cascade leading to cognitive
decline (Fig. 2). Prophylaxis with anti-TNF antibody prevents
postoperative cognitive decline. TNF-α blockade successfully
reduces further elaboration of IL-1, which has been extensively
reported to modulate symptoms of “sickness behavior,” including
memory dysfunction (20, 31, 32). With a single dose of anti-TNF
monoclonal antibody, we have also effectively interfered with the
IL-1-dependent amplification mechanism and cognitive function
(freezing) was normalized.
Cytokines are central mediators of inflammatory events fol-
lowing all types of events, including peripheral trauma or infec-
tion, and have broad physiological effects both on the periphery
and in the CNS (33). TNF-α has a pivotal role in the initiation
and amplification of the inflammatory cascade; it is involved in
regulating chemokines and cytokines release, oxidative stress,
recruitment of immune cells and adhesion molecules, apoptosis,
healing, and tissue-specific repair mechanism. This function has
been extensively documented during the analysis of the patho-
genesis of rheumatoid arthritis (34). TNF-α also exerts neuro-
flammation in MyD88−/−. (A and B)
Surgery resulted in reduced plasma
levels of IL-1β and IL-6 both at 6 and
24 h, as measured by ELISA. Anti-TNF
prophylaxis in MyD88−/−reduced the
amount of systemic cytokines to base-
line levels. (C) Hippocampal collection
was carried out 6 h after treatment.
Surgery resulted in no changes in the
expression of IL-1β compared with na-
ive MyD88−/−(KO). (D) Densitometry
of microglial immunostaining showed
no significant changes in microgliosis.
(E) Contextual fear response is the
same in MyD88−/−animals receiving
surgery compared with naive and
means ± SEM (n = 6, n = 10 for acute
behavior). *, P < 0.05; **, P < 0.01 by
repeated measures ANOVA followed
by Student-Newman-Keuls test. Krus-
kal-Wallis followed by the Dunn’s
multiple comparison test was used for
categorical data. Ab, antibody;
MyD88−/−;+/+, MyD88+/+; ND, not de-
termined; S, surgery.
Effects of surgery-induced in-
| www.pnas.org/cgi/doi/10.1073/pnas.1014557107Terrando et al.
modulatory functions, in particular in regulating microglia and
astrocytes activation in the brain (35).
Systemic cytokines exert effects in the CNS both via direct and
indirect pathways. TNF-α has been shown to enter the brain
through relatively permeable areas in the blood–brain barrier
(BBB) (36). In our model, it is possible that transient changes in
BBB permeability, either caused by the systemic inflammatory
response or other factors including anesthesia (37), could enable
direct access to the brain. Neural afferents may also account for
signs of CNS inflammation (38). Because of the changes in the
peripheral inflammatory effects that were noted in anti-TNF
antibody-treated surgical animals, and with the lack of evidence
of acute changes in the BBB postsurgery, we consider it likely
that the antibody exerts its ameliorative action in the periphery.
This finding suggests a pivotal role for systemic inflammation in
producing neuroinflammation and cognitive decline. Various
indirect routes for cytokine effects need to be excluded, and the
key site of anti-inflammatory effects needs further elucidation.
In the brain, microglia can actively secrete cytokines and neu-
rotoxins upon activation (39). Using another model of peripheral
organ inflammation, peripheral TNF-α was necessary to stimulate
microglia activation and to allow recruitment of circulating mon-
yet unable to define the precise origin of this neuroinflammatory
response, in particular whether it is mediated by peripheral mon-
ocytes or resident microglia. Resulting neuroinflammation ac-
counts for behavioral changes, in particular in vulnerable areas,
such as the hippocampus (41). Local inflammation directly inter-
feres with the processes of memory consolidation, long-term po-
tentiation, synaptic plasticity, and neurogenesis, resulting in what
has been termed “sickness behavior”; inflammatory blockade has
been reported to restore hippocampal neurogenesis (42).
Having defined a role for TNF-α following surgery, we in-
vestigated whether MyD88 signaling, important for the TLR and
IL-1 families, was involved in surgery-induced cognitive dysfunc-
tion. There was no surgery-induced freezing impairment, as
a marker of cognitive decline in MyD88−/−(Fig. 3E). This finding is
in keeping with our recently reported role of IL-1β in the same
from surgery-induced inflammation or POCD, it appears that
TLR4 has redundant role in this disease pathway. DAMPs, such as
HMGB-1, are recognized by multiple receptors (e.g., RAGE, as
well as Tlr4 and Tlr2) and one of the other receptors might be able
toinducesignaling inthe absence oftheother(43).Administration
of anti-TNF to MyD88−/−or Tlr4−/−completely abrogated the
surgery-induced IL-1β response and downstream IL-6 production,
suggesting a perpetuating role for both these cytokines in the in-
flammatory response and, possibly, synergism with one another.
Anti-TNF antibody was the first cytokine-selective therapy that
rheumatoidarthritis,beingbotheffective andacceptablysafe (44).
Based upon our data, prophylaxis with anti-TNF antibody is a fea-
sible therapeutic option that is ready tobe exploited in theelective
markedly with POCD, as therapy with anti-TNF, effective pro-
phylactically in septic mice, could not be given early in humans
because of the clinical reality of sepsis. This finding is in marked
contrast to the preemptive manner in which this biologic can be
used before surgery. Because other injured states, such as major
trauma and especially chemotherapy, lead to similar states of cog-
nitive decline as POCD, it is possible that similar up-regulation of
cytokines may underpin these clinical syndromes, and this is worth
investigating. It is possible that by administering anti-TNF antibody
surgical patients may be at higher risk for opportunistic infections
and other postoperative complications; however, most long-term
randomized clinical trials have not reported any differences in the
incidence of infections in patients treated with anti-TNF. Tubercu-
losis relapse is the chief exemption, with long-term treatment in-
creasing its prevalence (46). For POCD, anti-TNF would be very
short term: a single injection. Another possible target to prevent
postoperative cognitive decline may be HMGB-1, which was sig-
nificantly increased following surgical trauma (Fig. 1B) (47). Over-
the peripheral cytokine cascade that results in cognitive decline.
Prophylaxis with a single dose of anti-TNF antibody attenuates the
inflammatory pathways. Therapy with TNF-α inhibitors is clinically
conditions, such as rheumatoid arthritis, Crohn’s disease, and an-
kylosing spondylitis (22, 48), and may thus be useful for the pre-
Animal Experiments. Experiments were performed in accordance to the
UnitedKingdom HomeOffice-approved license.Werandomlygrouped12-to
14-wk-old male C57BL/6 mice and assigned them to a specific experiment.
Homozygous MyD88−/−on a C57BL/6 background were provided by the the
Tlr4−/−. (A) Contextual fear response
phylaxis with anti-TNF 18 h before
surgery prevented the cognitive ab-
normality. (B and C) Preemptive ad-
ministration of anti-TNF in Tlr4−/−
and IL-6 to baseline levels, as mea-
sured by ELISA. (D) Tlr4−/−showed
signs of neuroinflammation with in-
creased expression of hippocampal
IL-1β, which was reduced by anti-
TNF. (E) Densitometry of microglial
immunostaining with CD11b. One
day after surgery, Tlr4−/−showed
with naive and surgical mice pre-
operatively treated with anti-TNF.
Error bars represent the means ±
categorical data. Ab, antibody; S, surgery;−/−, TLR4−/−;+/+, TLR4+/+.
Terrando et al.PNAS
| November 23, 2010
| vol. 107
| no. 47
Sanger Institute (49). Tlr4−/−mice were obtained from B&K Universal (50). Download full-text
Age-matched congenic inbred wild-type C57BL/6 mice were obtained from
Charles River. All animals were fed standard rodent chow and water ad
libitum, and were housed (< five mice per cage) in sawdust-lined cages in an
air-conditioned environment with 12-h light/dark cycles. All of the animals
were checked on a daily basis and if they evidenced poor grooming, hud-
dling, piloerection, weight loss, back arching, and abnormal activity, they
were eliminated from further consideration. Investigators who treated
animals knew the treatment groups and collected samples, which were then
analyzed by other investigators blinded to the specific treatment.
Surgery. We performed an open tibial fracture as previously described (51).
Briefly, we anesthetized mice with 2.1% isoflurane and analgesia with
buprenorphine (Buprenex, 0.1 mg/kg s.c). A middle incision was performed
on the left hind paw and a 0.38-mm pin was inserted in the intramedullary
canal, the periosteum stripped, and osteotomy performed. Aseptic con-
ditions were maintained throughout. We subjected mice to vehicle (saline)
or TNF-neutralizing antibody (52) (clone TN3, 100 μg per mouse; Sigma) 18 h
preoperatively. Blood was collected by cardiac puncture. Plasma cytokines
and hippocampal IL-1β were measured by ELISA according to the manu-
facturer’s instructions. Fixed brains were collected for immunohistochemical
DAB staining for microglia activation using CD11b (SI Methods).
Behavior. The behavioral study was conducted using a dedicated condi-
tioning chamber (Med Associates Inc.). Mice were trained and tested on
separate days. The fear-conditioning paradigm was used as previously
described (18). Freezing behavior was recorded 3 d after training. Mice
from each treatment group were randomly assigned for assessment of
either cytokine response or cognitive behavior to obviate possible con-
founding effects of behavioral testing on inflammatory markers (53)
Data Analysis. We used GraphPad v3.0 (GraphPad Software) to calculate the
mean, SD, andSEM, andperformstatistical tests.We analyzed multiple group
means by one-way analysis of variance, followed by Newman-Keuls post hoc
test wherever appropriate. The nonparametric test of Kruskal-Wallis fol-
lowed by the Dunn’s multiple comparison test was used for categorical data.
P values less than 0.05 were considered significant.
ACKNOWLEDGMENTS. This paper is dedicated to the memory of Professor
Brian Foxwell, an inspirational colleague who will be sadly missed. This work
was supported by the Westminster Medical School Research Trust, London,
United Kingdom, the Mathilda and Terence Kennedy Institute of Rheuma-
tology Trust, and Arthritis Research United Kingdom.
1. Bianchi ME, Manfredi AA (2009) Immunology. Dangers in and out. Science 323:
2. Wilson CJ, Finch CE, Cohen HJ (2002) Cytokines and cognition—The case for a head-
to-toe inflammatory paradigm. J Am Geriatr Soc 50:2041–2056.
3. Zhang Q, et al. (2010) Circulating mitochondrial DAMPs cause inflammatory responses
to injury. Nature 464(7285):104–107.
4. Medzhitov R (2008) Origin and physiological roles of inflammation. Nature 454:
5. Trompet S, et al.; PROSPER Group (2008) Genetic variation in the interleukin-1 beta-
converting enzyme associates with cognitive function. The PROSPER study. Brain 131:
6. Wan Y, et al. (2007) Postoperative impairmentof cognitive function in rats: A possible role
for cytokine-mediated inflammation in the hippocampus. Anesthesiology 106:436–443.
7. Rosczyk HA, Sparkman NL, Johnson RW (2008) Neuroinflammation and cognitive
function in aged mice following minor surgery. Exp Gerontol 43:840–846.
8. Ehlenbach WJ, et al. (2010) Association between acute care and critical illness
hospitalization and cognitive function in older adults. JAMA 303:763–770.
9. Girard TD, et al. (2010) Delirium as a predictor of long-term cognitive impairment in
survivors of critical illness. Crit Care Med 38:1513–1520.
10. Guenther U, et al. (2010) Validity and reliability of the CAM-ICU Flowsheet to
diagnose delirium in surgical ICU patients. J Crit Care 25:144–151.
11. Ely EW, et al. (2004) Delirium as a predictor of mortality in mechanically ventilated
patients in the intensive care unit. JAMA 291:1753–1762.
12. Ely EW, et al. (2001) The impact of delirium in the intensive care unit on hospital
length of stay. Intensive Care Med 27:1892–1900.
13. Milbrandt EB, et al. (2004) Costs associated with delirium in mechanically ventilated
patients. Crit Care Med 32:955–962.
14. Newman S, Stygall J, Hirani S, Shaefi S, Maze M (2007) Postoperative cognitive
dysfunction after noncardiac surgery: A systematic review. Anesthesiology 106:572–590.
15. Newman MF, et al.; Neurological Outcome Research Group and the Cardiothoracic
Anesthesiology Research Endeavors Investigators (2001) Longitudinal assessment of
neurocognitive function after coronary-artery bypass surgery. N Engl J Med 344:395–402.
16. Steinmetz J, Christensen KB, Lund T, Lohse N, Rasmussen LS; ISPOCD Group (2009) Long-
term consequences of postoperative cognitivedysfunction. Anesthesiology 110:548–555.
17. Moller JT, et al.; International Study of Post-Operative Cognitive Dysfunction (1998)
Long-term postoperative cognitive dysfunction in the elderly ISPOCD1 study. ISPOCD
investigators. Lancet 351:857–861.
18. Cibelli M, et al. (2010) Role of interleukin-1β in postoperative cognitive dysfunction.
Ann Neurol 68:360–368.
19. Scaffidi P, Misteli T, Bianchi ME (2002) Release of chromatin protein HMGB1 by
necrotic cells triggers inflammation. Nature 418:191–195.
20. Terrando N, et al. (2010) The impact of IL-1 modulation on the development of
lipopolysaccharide-induced cognitive dysfunction. Crit Care 14(6894):R88.
21. Fong Y, et al. (1989) Antibodies to cachectin/tumor necrosis factor reduce interleukin 1
beta and interleukin 6 appearance during lethal bacteremia. J Exp Med 170:1627–1633.
22. Brennan FM, Chantry D, Jackson A, Maini R, Feldmann M (1989) Inhibitory effect of
TNF alpha antibodies on synovial cell interleukin-1 production in rheumatoid arthritis.
23. Rachal Pugh C, Fleshner M, Watkins LR, Maier SF, Rudy JW (2001) The immune system and
memory consolidation: A role for the cytokine IL-1beta. Neurosci Biobehav Rev 25:29–41.
24. Aloisi F (2001) Immune function of microglia. Glia 36:165–179.
25. Rosi S, et al. (2009) Accuracy of hippocampal network activity is disrupted by
neuroinflammation: Rescue by memantine. Brain 132:2464–2477.
26. Ekdahl CT, Claasen JH, Bonde S, Kokaia Z, Lindvall O (2003) Inflammation is
detrimental for neurogenesis in adult brain. Proc Natl Acad Sci USA 100:13632–13637.
27. Chowdhury N, Quinn JJ, Fanselow MS (2005) Dorsal hippocampus involvement in
trace fear conditioning with long, but not short, trace intervals in mice. Behav
28. Liu ZG (2005) Molecular mechanism of TNF signaling and beyond. Cell Res 15:24–27.
29. Midwood K, et al. (2009) Tenascin-C is an endogenous activator of Toll-like receptor 4 that
is essential for maintaining inflammation in arthritic joint disease. Nat Med 15:774–780.
30. Akira S, Takeda K (2004) Toll-like receptor signalling. Nat Rev Immunol 4:499–511.
31. Kent S, et al. (1992) Different receptor mechanisms mediate the pyrogenic and
behavioral effects of interleukin 1. Proc Natl Acad Sci USA 89:9117–9120.
32. Dinarello CA (2009) Immunological and inflammatory functions of the interleukin-1
family. Annu Rev Immunol 27:519–550.
33. Feldmann M, et al. (2001) Cytokine Reference: A Compendium of Cytokines and
Other Mediators of Host Defense (Academic Press, London).
34. Feldmann M, Brennan FM, Maini RN (1996) Role of cytokines in rheumatoid arthritis.
Annu Rev Immunol 14:397–440.
35. Beutler B, Cerami A (1989) The biology of cachectin/TNF—A primary mediator of the
host response. Annu Rev Immunol 7:625–655.
36. Gutierrez EG, Banks WA, Kastin AJ (1993) Murine tumor necrosis factor alpha is
transported from blood to brain in the mouse. J Neuroimmunol 47:169–176.
37. Tétrault S, Chever O, Sik A, Amzica F (2008) Opening of the blood-brain barrier during
isoflurane anaesthesia. Eur J Neurosci 28:1330–1341.
38. Dantzer R (1994) How do cytokines say hello to the brain? Neural versus humoral
mediation. Eur Cytokine Netw 5:271–273.
39. Hanisch UK, Kettenmann H (2007) Microglia: Active sensor and versatile effector cells
in the normal and pathologic brain. Nat Neurosci 10:1387–1394.
40. D’Mello C, Le T, Swain MG (2009) Cerebral microglia recruit monocytes into the brain
in response to tumor necrosis factoralpha signaling during peripheral organ
inflammation. J Neurosci 29:2089–2102.
41. Gemma C, Fister M, Hudson C, Bickford PC (2005) Improvement of memory for
context by inhibition of caspase-1 in aged rats. Eur J Neurosci 22:1751–1756.
42. Monje ML, Toda H, Palmer TD (2003) Inflammatory blockade restores adult
hippocampal neurogenesis. Science 302:1760–1765.
43. van Zoelen MA, et al. (2009) Role of toll-like receptors 2 and 4, and the receptor for
advanced glycation end products in high-mobility group box 1-induced inflammation
in vivo. Shock 31:280–284.
44. Feldmann M, Maini RN (2010) Anti-TNF therapy, from rationale to standard of care:
What lessons has it taught us? J Immunol 185:791–794.
45. Tracey KJ, et al. (1987) Anti-cachectin/TNF monoclonal antibodies prevent septic shock
during lethal bacteraemia. Nature 330:662–664.
46. Keane J, et al. (2001) Tuberculosis associated with infliximab, a tumor necrosis factor
alpha-neutralizing agent. N Engl J Med 345:1098–1104.
47. Yang H, et al. (2004) Reversing established sepsis with antagonists of endogenous
high-mobility group box 1. Proc Natl Acad Sci USA 101:296–301.
48. Feldmann M, Maini RN (2003) Lasker Clinical Medical Research Award. TNF defined as
a therapeutic target for rheumatoid arthritis and other autoimmune diseases. Nat
49. Adachi O, et al. (1998) Targeted disruption of the MyD88 gene results in loss of IL-1-
and IL-18-mediated function. Immunity 9:143–150.
50. Hoshino K, et al. (1999) Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are
hyporesponsive to lipopolysaccharide: Evidence for TLR4 as the Lps gene product. J
51. Harry LE, et al. (2008) Comparison of the healing of open tibial fractures covered with
either muscle or fasciocutaneous tissue in a murine model. J Orthop Res 26:1238–1244.
52. Sheehan KC, Ruddle NH, Schreiber RD (1989) Generation and characterization of
hamster monoclonal antibodies that neutralize murine tumor necrosis factors. J
53. Nguyen KT, et al. (1998) Exposure to acute stress induces brain interleukin-1beta
protein in the rat. J Neurosci 18:2239–2246.
| www.pnas.org/cgi/doi/10.1073/pnas.1014557107Terrando et al.