Microglial Cell Activation Neurogenesis by Negative

TAM Receptors Affect Adult Brain Neurogenesis by Negative Regulation of Microglial Cell Activation

This information is current as Rui Ji, Shifu Tian, Helen J. Lu, Qingjun Lu, Yan Zheng, of September 24, 2021. Xiaomin Wang, Jixiang Ding, Qiutang Li and Qingxian Lu J Immunol 2013; 191:6165-6177; Prepublished online 15 November 2013; doi: 10.4049/jimmunol.1302229

http://www.jimmunol.org/content/191/12/6165 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2013/11/18/jimmunol.130222 Material 9.DC1 http://www.jimmunol.org/ References This article cites 100 articles, 32 of which you can access for free at: http://www.jimmunol.org/content/191/12/6165.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 24, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

TAM Receptors Affect Adult Brain Neurogenesis by Negative Regulation of Microglial Cell Activation

Rui Ji,*,† Shifu Tian,* Helen J. Lu,* Qingjun Lu,‡,x,{ Yan Zheng,{ Xiaomin Wang,{ Jixiang Ding,‖ Qiutang Li,*,# and Qingxian Lu*,†,#

TAM tyrosine kinases play multiple functional roles, including regulation of the target genes important in homeostatic regulation of cytokine receptors or TLR-mediated signal transduction pathways. In this study, we show that TAM receptors affect adult hippocampal neurogenesis and loss of TAM receptors impairs hippocampal neurogenesis, largely attributed to exaggerated inflammatory responses by microglia characterized by increased MAPK and NF-kB activation and elevated production of proinflammatory cytokines that are detrimental to neuron stem cell proliferation and neuronal differentiation. Injection of LPS causes even more severe inhibition of BrdU incorporation in the Tyro32/2Axl2/2Mertk2/2 triple-knockout (TKO) brains, consistent with the LPS- elicited enhanced expression of proinflammatory mediators, for example, IL-1b, IL-6, TNF-a, and inducible NO synthase, and this Downloaded from effect is antagonized by coinjection of the anti-inflammatory drug indomethacin in wild-type but not TKO brains. Conditioned medium from TKO microglia cultures inhibits neuron stem cell proliferation and neuronal differentiation. IL-6 knockout in Axl2/2 Mertk2/2 double-knockout mice overcomes the inflammatory inhibition of neurogenesis, suggesting that IL-6 is a major downstream neurotoxic mediator under homeostatic regulation by TAM receptors in microglia. Additionally, autonomous trophic function of the TAM receptors on the proliferating neuronal progenitors may also promote progenitor differentiation into immature neurons. The Journal of Immunology, 2013, 191: 6165–6177. http://www.jimmunol.org/

icroglial cells, a diverse set of innate immune cells brain injury and chronic degenerative diseases (4–9). However, distributed throughout the entire CNS, actively scan the chronic inflammation and uncontrolled activation of microglia are M CNS microenvironment (1) and provide trophic or detrimental to neuronal functions and neurogenesis (10). maintenance support for normal neuron activity (2). Serving as Microglial cells express TLRs, which can be activated by en- a major immunosurveillance cell type in the CNS (3), microglia dogenous and exogenous ligands (11–13). Activation of TLRs express all necessary receptors and molecules for recognition of causes rapid activation of microglial cells and initiates multiple invading microbes, pathogenic stimuli, proinflammatory cytokines, downstream signaling pathways, with the most common being the by guest on September 24, 2021 and cellular debris (spent or damaged neuronal organelles). When Erk1/2 and p38 MAPK pathway and the IkB kinase (IKK)/NF-kB activated, they are able to mount rapid innate immune responses signal transduction pathway, which lead, respectively, to activation with increased production of proinflammatory cytokines and che- of AP-1 or NF-kB and their subsequent nuclear binding to AP-1 mokines not only in response to systemic infection, but also to and kB binding sites on the promoters of multiple proinflammatory genes (14–18). LPS from Gram-negative bacteria binds specifically to TLR4 on microglia and triggers intracellular sig- *Department of Ophthalmology and Visual Sciences, University of Louisville School of naling through the MAPK or IKK/NF-kB pathway, leading to rapid Medicine, Louisville, KY 40202; †Department of Anatomical Sciences and Neurobiology, transcriptional activation of innate immune-responsive genes, in- University of Louisville School of Medicine, Louisville, KY 40202; ‡Beijing Tongren Eye x cluding those coding for IL-1b,IL-6,andTNF-a. Although Institute, Beijing Capital Medical University, Beijing 100050, China; Beijing School of Ophthalmology, Beijing Capital Medical University, Beijing 100050, China; {Basic Col- microglia are vital in immune surveillance and in defending the ‖ lege of Medicine, Beijing Capital Medical University, Beijing 100050, China; Birth CNS from foreign or local danger, unrestrained and prolonged Defects Center, University of Louisville School of Dentistry, Louisville, KY 40202; and #James Graham Brown Cancer Center, University of Louisville School of Medicine, activation of brain-resident microglia is detrimental to normal Louisville, KY 40202 brain function and neuronal survival. There is evidence that sys- Received for publication August 26, 2013. Accepted for publication October 9, 2013. temic or local chronic inflammation in the CNS is detrimental not This work was supported by National Institutes of Health Grant R01 EY018830 and by only to normal neural function (19), but also to the neurogenesis the Research to Prevent Blindness special award (to Qingxian Lu). Q. Li was supported and differentiation of neuronal stem cells (NSCs) into immature by National Institutes of Health Grant R01 EY019891. Qingxian Lu and J.D. were sup- neurons (10, 20–24). LPS-elicited microglial inflammation indu- ported by an Institutional Development Award from the National Institute of General Medical Sciences of the National Institutes of Health under Grant P20GM103453. ces the release of proinflammatory cytokines affecting NSC pro- Address correspondence and reprint requests to Dr. Qingxian Lu, University of liferation in vitro and inhibiting hippocampal neurogenesis and Louisville School of Medicine, 301 E. Muhammad Ali Boulevard, Louisville, KY neuronal differentiation, and these negative effects are antagonized 40202. E-mail address: [email protected] by immunosuppressive drugs (21–28). The online version of this article contains supplemental material. Interestingly, microglia may have evolved to keep the brain Abbreviations used in this article: DC, dendritic cell; DCX, doublecortin; DG, dentate immune response in close check. To avoid exaggerated immune gyrus; DKO, double-knockout; GFAP, glial fibrillary acidic protein; IKK, IkB kinase; iNOS, inducible NO synthase; NSC, neural stem cell; poly(I:C), polyinosinic- responses to infection or pathogenic changes, innate immune cells, polycytidylic acid; qPCR, quantitative PCR; SGZ, subgranular zone; SOCS, suppres- including microglia, have developed several regulatory mecha- sor of cytokine signaling; SVZ, subventricular zone; TAM, Tyro3, Axl, and Mertk; nisms to terminate their own innate immune responses. The best TKO, triple-knockout; WT, wild-type. studied mechanisms for termination of proinflammatory cytokine Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 gene expression include 1) the rapid cytoplasmic re-expression of www.jimmunol.org/cgi/doi/10.4049/jimmunol.1302229 6166 TAM RECEPTORS AND ADULT NEUROGENESIS

IkB that inhibits NF-kB transcriptional activity, 2) the quick gain Reagents and materials of phosphatases causing dephosphorylation of MAPKs, and 3) the The NSC culture supplements B-27 (1003) and N-2 (503) were purchased efficient suppression and termination of multiple cytokine receptor from Invitrogen (San Diego, CA). Recombinant mouse Gas6 was pur- signaling by newly synthesized suppressor of cytokine signaling chased from R&D Systems (Minneapolis, MN). LPS (Escherichia coli (SOCS) protein (15) or transcriptional repressors for proinflammatory serotype 055:B5) and indomethacin were purchased from Sigma-Aldrich. cytokine genes (29). The Abs used for Western blotting were rabbit polyclonal anti-mouse p38 or anti-mouse pErk and mouse polyclonal anti–phospho-p38 and anti- In a search for upstream modulators that inhibit cytokine re- mouse Erk42/44 (all from Santa Cruz Biotechnology, Santa Cruz, CA), ceptor signaling, the Tyro3, Axl, and Mertk (TAM) receptor ty- HRP-conjugated donkey or sheep polyclonal anti-goat/rabbit/mouse IgG rosine kinases, which are expressed on dendritic cells (DCs) and (Amersham Biosciences, Piscataway, NJ), goat anti-mouse Axl, Tyro3 or macrophages, were found to function as important immunomod- Mertk (Santa Cruz Biotechnology), and monoclonal mouse anti-mouse b-actin (Sigma-Aldrich). The Abs used for immunocytochemistry were ulators (15, 30–33). This family of receptors on innate immune monoclonal mouse anti-mouse b-tubulin III (anti-TUJ-1; StemCell Tech- cells plays a pivotal inhibitory regulatory role by limiting pro- nologies, Vancouver, BC, Canada), anti-mouse NeuN (EMD Millipore, longed and unrestricted signaling initially triggered by cytokines Billerica, MA), anti-mouse glial fibrillary acidic protein (GFAP; EMD or pathogen-associated molecular patterns receptors by inhibition Millipore), anti-mouse doublecortin (DCX; Santa Cruz Biotechnology), of NF-kB signaling and upregulation of SOCS and Twist proteins, and anti-mouse Iba-1 (Wako Chemicals, Richmond, VA). which terminate cytokine signaling or block the binding of NF-kB RNA isolation, cDNA synthesis, and real-time quantitative to its target gene promoters (15, 29, 34). In TAM triple-knockout PCR (TKO) mice, the loss of TAM receptors on DCs and macrophages Total RNA was extracted from cultured NSCs, microglia, or the hippo- (35–37) or other nonprofessional phagocytes (32, 38) leads to campus using TRIzol reagent (Invitrogen) for tissues or RNeasy kits Downloaded from defective phagocytosis and overproduction of proinflammatory (Qiagen) for cultured cells. Total RNA (1 mg) from each sample was treated cytokines, resulting in chronic inflammation and systemic auto- with DNase I to remove traces of genomic DNA, then reverse transcribed immune disorders (15, 31). Development of autoimmunity in these into first-strand cDNA using qScript cDNA SuperMix kits (Quanta Bio- Sciences, Gaithersburg, MD) for real-time quantitative PCR (qPCR) mutant mice causes autoantibody deposition in, and autoreactive analysis. A sample of 50 ng cDNA was used to quantify gene expression by lymphocyte infiltration into, a variety of tissues, including the qPCR using a SYBR Green–based PCR reaction mixture on an MX3005p brain (30). system (Agilent Technologies, Santa Clara, CA). http://www.jimmunol.org/ The negative effects of TAM receptors on DCs and macrophages Isolation and primary culture of NSCs on the innate immune response prompted us to investigate their functional roles on microglia and NSCs and the impact on adult Newborn mice at postnatal days 0 or 1 were used for isolation of hippo- hippocampal neurogenesis. In this study, we explored how TAM campal stem cells. After trypsinization, the single-cell suspension at a concentration of 1 3 105 cells/ml was cultured in complete NSC medium receptors regulate microglia activation and examined their effect (DMEM/F12, 13 B-27, 13 N-2, 20 ng/ml basic fibroblast factor, 20 ng/ml on adult neurogenesis. We showed that microglia lacking TAM epidermal growth factor, 13 penicillin/streptomycin [Invitrogen]). The receptors were hyperactivated and produced increased amounts of proliferating neuronal stem cells formed NSC spheres in 4 d. proinflammatory cytokines, especially in response to activation by Isolation and primary culture of brain microglia and astrocytes LPS, polyinosinic-polycytidylic acid [poly(I:C)], or CpG. This hy- by guest on September 24, 2021 perreactivity of mutant microglia was shown to play a major role in A forebrain single-cell suspension was prepared as described above, and then 3 3 106 viable cells in 3 ml microglia culture medium (DMEM/F12 the impaired hippocampal neurogenesis observed in vivo. Condi- medium [Invitrogen] containing 10% FBS, 2 mM GlutaMAX, 13 tioned medium from TKO microglia cultures was found to be penicillin/streptomycin) were plated into each 3.5-cm cell culture dish and neurotoxic for cultured wild-type (WT) NSCs and, of the neurotoxic cultured for 20 d, with the culture medium being changed every 4–5 d. substrates in conditioned medium that might be detrimental to After reaching confluence, the cells were washed with PBS and trypsinized neurogenesis, IL-6 was found to be a major player, because addition with 0.06% trypsin at 37˚C for 40 min, and then the trypsin was quenched with DMEM containing 10% FBS and the detached astrocytes in the of neutralizing anti–IL-6 Ab to the conditioned medium or knockout culture medium were saved, transferred to a new tube, and triturated into of the IL-6 gene in TAM TKO mice restored hippocampal neuro- single cells before plating in a new six-well plate. The cells still attached to genesis to a level comparable to that in WT controls. Additionally, the culture dishes were mainly microglia and were further cultured for 1 loss of TAM receptors significantly impaired the differentiation more day before being used for functional studies. To assess cell purity, microglial cultures were immunostained with anti– process that converted neuronal progenitors to immature neurons, Iba-1 Ab, showing that .98% were Iba-1+, whereas astrocyte cultures and this effect was not inhibited by immunosuppressive drug. This were labeled with anti-GFAP Ab and 85% were found to be GFAP+. is consistent with observation that knockout of the TAM receptor ligand Gas6 decreases BrdU incorporation into the proliferating Preparation of microglia- or astrocyte-conditioned medium neuronal progenitors in the brain subventricular zone (SVZ) (39). and treatment of NSCs We conclude that TAM receptors play a dual role in either modu- Microglia or astrocytes at 80–90% confluence were activated by addition of lating microglia innate immune responses by damping down pro- LPS (0.1 mg/ml, serotype 055:B5; Sigma-Aldrich) for 6 h and washed three duction of proinflammatory factors detrimental to neuronal function times with PBS to remove all traces of LPS, and then fresh medium was added and culture continued for another 24 h before collection of conditioned and NSC renewal or by providing trophic support for neuronal medium. Microglia-conditioned medium (25%) was added to NSC cultures. differentiation. Unbiased stereological estimation and statistical analysis of Materials and Methods granular cells in the granular layer of the hippocampal dentate Animals gyrus The Tyro32/2Axl2/2Mertk2/2 TKO and Axl2/2Mertk2/2 double-knockout To estimate the volume and cell number of the granular cell layer in (DKO) mice have been described previously (40, 41). C57BL/6J, Il62/2, hippocampus, the number of granule cells was determined in every sixth and Tnfr12/2/Tnfr22/2 (Tnfr1/22/2) DKO mice were purchased from The section in a series of 40-mm coronal sections using the unbiased stereology Jackson Laboratory. Il62/2 mice or Tnfr1/22/2 DKO mice were bred with method described above. Granule cells were identified by blue nuclear TKO mice to generate Il62/2Tyro32/2Axl2/2Mertk2/2 and Tnfr1/22/2 staining with Hoechst 33342 (Molecular Probes, Eugene, OR; 0.5 mg/ml Tyro32/2Axl2/2Mertk2/2 compound mice. All animals were housed in in TBS for 15 min) and, in most cases, immunostained with anti-NeuN Ab a pathogen-free facility and were handled according to the regulations of (1:100; EMD Millipore), and any that lay across the top focal plane or the the Institutional Animal Care and Use Committee. exclusion boundaries of the unbiased sampling frame were excluded from The Journal of Immunology 6167 counting. The granule cell layer reference volume was determined by the a significant difference. All experiments were performed at least three traced granule cell areas for each section multiplied by the distance be- times in triplicate. Data are expressed as the means 6 SD. tween the sections sampled, and then the mean granule cell number per disector volume was multiplied by the reference volume to estimate the total granule cell number. Results Adult brain neurogenesis is reduced in the hippocampus of the Injection of LPS, indomethacin, and BrdU Tyro32/2Axl2/2Mertk2/2 TKO brain To induce acute inflammation, a single i.p. injection of LPS (1 mg/kg body weight) was given and BrdU (100 mg/kg body weight) or PBS as a control Because all three TAM family receptors are expressed in brain, was injected i.p. 30 h later and daily for the following 6 d, and then the mice especially in the hippocampus (42, 43), this prompted us to in- were sacrificed 24 h after the last injection. In the groups treated with the anti- vestigate whether neurogenesis in the mutant adult brain was inflammatory drug indomethacin, the first injection of indomethacin (2.5 mg/ defective. To assess NSC proliferation, we first labeled prolifer- kg body weight) was given 18 h before the LPS injection, and then the same dose was injected twice daily for 7 consequent days. Twenty-four hours after ating NSCs with BrdU, a thymidine analog that can be inserted the last injection of BrdU or PBS, the mice were processed for BrdU im- into newly synthetic genomic DNA, on days 1–7, and then anti- munohistochemistry. For measurement of LPS-induced proinflammatory BrdU immunohistochemistry was performed on brain sections cytokines in hippocampus, a single i.v. injection of LPS (1 mg/kg body prepared on day 8. BrdU reactivity in the subgranular layer of the weight) was given at 8 h prior to RNA isolation from the hippocampal tissues. dentate gyrus (DG) represents proliferating NSCs, and the TKO Immunocytochemistry brain section clearly showed reduced BrdU labeling (Fig. 1B) ∼ Free-floating brain sections or differentiated cells on chamber slides were compared with the WT control (Fig. 1A), with an 50% reduction in the TKO brains (Fig. 1C, **p , 0.05). To assess neuronal incubated overnight at 4˚C with Abs against BrdU, b-tubulin III (anti-TUJ-1, Downloaded from 1:200; StemCell Technologies), NeuN or GFAP (1:200; EMD Millipore), differentiation of the surviving TKO NSCs, we next studied im- DCX (1:100; Santa Cruz Biotechnology), or Iba-1 (1:200; Wako Chemicals) mature neurons in the subgranular region by immunostaining and analyzed on a Carl Zeiss Axio Imager M2 cell imaging microscope hippocampal sections for the immature neuron marker DCX, equipped with Apotome and AxioCam systems (Carl Zeiss, Jena, Germany). a cytoskeleton-associated protein expressed transiently during Statistical analysis adult neurogenesis, and we found that the number of DCX+/BrdU+ Data were analyzed using ANOVA, followed by a Student t test for in- double-positive immature neurons in TKO brains showed an even dependent means where appropriate to determine which results showed more dramatic decrease than did the BrdU+ cells when compared http://www.jimmunol.org/

FIGURE 1. Decreased neurogenesis in the SGZ of the TKO mouse hippocampus. (A and B) WT and TKO mice at 2 mo old were injected i.p. with 100 mg/kg (body weight) of BrdU for 7 consecutive days. On day 8, the mice were deeply anesthetized with an overdose of 2.5% avertin prior to intracardial perfusion-fixation with 100 ml 4% paraformaldehyde and immersive postfixation by guest on September 24, 2021 at 4˚C for overnight after the brain was removed, then 40-mm serial coronal brain sections were prepared and immunostained with Abs against BrdU (green) or Dcx (red), and the nuclei were stained with Hoechst 33342 dye (blue). Prolifer- ating NSCs are labeled with BrdU, whereas immature neurons are BrdU+/Dcx+. The images show representative sections of the SGZ of the WT (A)or TKO (B)DGatbregma21.70 mm and the entire region from bregma 21.34 mm to bregma 23.08 mm was analyzed. Five mice were used for each group, and 45 sections were analyzed per mouse. Scale bar, 100 mm. (C) Quantification of proliferating cells (BrdU+) and immature neurons (BrdU+/ Dcx+) per coronal section at bregma 21.70 mm; n =5.(D and E) Stereological analysis of the total volume (D) and cell number (E) of the granule cell layer. The region analyzed was bregma 21.34 mm to bregma 23.08 mm; n =5.(F) Quantification of NSC transcriptional factor mRNA levels in hippocampus by real-time qPCR. Total RNA (1 mg) from pooled hippocampi (n = 3) of 4-mo-old WT or TKO mice was reverse transcribed using qScript cDNA SuperMix kits, and real-time qPCR was performed to quantify the relative expression of Sox 2, c-Myc, and KLF4 in the WT and TKO hippocampus. Data are shown as means 6SD of the mRNA expression level relative to b-actin mRNA level and include one representative of three experiments. The open and filled bars represent data for WT and TKO mice, respectively. *p , 0.05, **p , 0.01 in one-way ANOVA test using ProStat version 5.5. GrDG, Granular DG; hif, hippocampal fissure. 6168 TAM RECEPTORS AND ADULT NEUROGENESIS with the WT controls (Fig. 1A–C, Supplemental Fig. 1), with an and astrocytes that had been cultured for .3wkrepresentsthe average 60% reduction in the TKO brains (Fig. 1C, **p , 0.01). in vivo expression profile is currently not clear. Decreased NSC proliferation and immature neuron generation in the TKO hippocampal DG led to a slightly reduced total volume Primary microglia from TKO brains are hyperreactive to (Fig. 1D, p = 0.203, n = 5) and cell population (Fig. 1E, p = 0.024, activation of several different TLRs n = 5) in the DG granular layer as studied by unbiased stereo- Systemic or local chronic inflammation inhibits adult hippocampal logical quantification. Reduced hippocampal neurogenesis was neurogenesis (21, 24, 47). Microglia are brain-resident immune- also noticed in the Axl2/2Mertk2/2 DKO mice, but not in all three responsive cells with the similar developmental origin as periph- single gene knockout mice (Supplemental Fig. 2). eral macrophages. Given that TAM TKO macrophages show hy- Adult NSCs express the transcriptional factors Sox2, c-myc, and perreactivity to LPS stimulation (15, 33), TAM receptor–deficient KLF4 (44, 45). Comparison of the relative expression of these three microglia in the CNS might also show an unrestricted innate transcriptional factors in the adult hippocampus by real-time qPCR immune response and produce increased amounts of proin- showed that mRNA levels of both Sox2 and c-myc were signifi- flammatory cytokines detrimental to hippocampal neurogenesis. cantly reduced in 8-wk-old TKO mice (Fig. 1F, p , 0.05), whereas To test this hypothesis, we isolated and purified primary microglia the KLF4 mRNA showed slight reduction (Fig. 1F, p = 0.073). This from both WT and TKO brains (Supplemental Fig. 3) and ex- suggests a reduction in the cell population expressing these factors. amined proinflammatory cytokine expression by primary cultured microglia stimulated with LPS, poly(I:C), or CpG and found that TAM receptors are expressed in both microglia and astrocytes these components of infectious pathogens induced a rapid immune

TAM receptors are expressed in macrophages and DCs (30, 34, response by both WT and TKO microglia, with peak expression of Downloaded from 35, 40, 46), in a few nonprofessional phagocytes including the mRNAs coding for the proinflammatory cytokines IL-1b, IL-6, Sertoli cells in testis (38, 40) and retinal pigmental epithelium and inducible NO synthase (iNOS) within 4–8 h (Fig. 3A). Re- cells in eye (32), and in the CNS (42, 43). In the absence of TAM markably, at the peak, the TKO cells produced more than twice receptors, macrophages and DCs produce increased levels of proin- the amount of these factors as did WT cells. This enhanced pro- flammatory cytokines (34). To determine whether brain glial cells duction of proinflammatory cytokines by TKO microglia was

express all three members of TAM receptors that might negatively found not only in response to LPS, which acts through TLR4, but http://www.jimmunol.org/ regulate brain glial cell cytokine production responsible for inhibition also to poly(I:C) and CpG, activating, respectively, TLR3 and of neurogenesis in TKO mice, we first examined the expression TLR9 (48). After 24 h of stimulation, levels of mRNA for these profile of TYRO3, AXL, and MERTK proteins in primary cultured cytokines were reduced to basal levels in both the WT and TKO brain microglia and astrocytes by Western blotting. Although all groups, whereas, as shown in Fig. 3B, IL-6 release into the culture three receptors were expressed in both microglia and astrocytes, medium increased continuously during the 24-h stimulation pe- the expression of each receptor varied significantly. WT micro- riod. Additionally, TKO microglia produced more TNF-a than did glia showed a dominant Mertk signal and weak expression of WT controls in response to LPS treatment (Fig. 3C). These data both Tyro3 and Axl, whereas WT astrocytes showed equally high demonstrate that TAM receptors play a negative role in the TLR- expression of Tyro 3 and Axl and weak Mertk expression (Fig. 2, initiated signal transduction pathways. by guest on September 24, 2021 left panel). Both cell types from TKO brains were negative for all three proteins. WT mouse testis, a tissue known to express all LPS stimulation leads to increased activation of the MAPK and three receptors (40), was used as the positive control, and testes IKK/NF-kB signal transduction pathways in TKO microglia from Axl2/2Mertk2/2 DKO mice and Tyro32/2Mertk2/2 DKO and macrophages mice were used as negative controls (Fig. 2, right panel). Whether Given that TKO microglia showed enhanced expression of proin- such relative expression profile of the three receptors in the microglia flammatory mediators in response to TLR activation, we asked what signal transduction pathways were responsible. It is well estab- lished that TLR activation triggers multiple downstream signaling pathways, including the Erk1/2 and p38 MAPK pathway and the IKK/NF-kB pathway (14–17). Microglia express TLRs and all components of the IKK/NF-kB signaling pathway and produce proinflammatory mediators on activation (8, 49). To examine the negative functional role of TAM receptors on microglia in response to TLR activation, we activated TLR4 on primary cultured microglia with LPS and found that LPS stimulation caused rapid activation of both ERK1/2 and p38 in TKO cells compared with WT microglia (Fig. 4A). p38 has been suggested to be a major MAPK for regulating proinflammatory cytokine production by microglia (17). Indeed, we observed early activation of p38 in the 0 time sample (Fig. 4A, p-p38 row at 0 h) and rapid induction of FIGURE 2. TAM receptors are expressed by microglia and astrocytes. phosphorylation within the first hour of LPS stimulation in the Primary microglia and astrocytes were isolated from newborn WT or TKO TKO microglia (Fig. 4A, top row). This suggests that negative mice and the same number of cultured cells were directly lysed in 13 regulation of TLR signaling by TAM receptors is conserved in all NuPAGE LDS sample buffer (Invitrogen, Carlsbad, CA) and the lysate was myeloid-derived cell lineages and that loss of TAM receptors on subjected to SDS-PAGE electrophoresis, transferred onto a nitrocellulose microglia is responsible for the overproduction of proinflammatory membrane, and immunoblotted for Tyro3, Axl, Mertk, and b-actin. The TKO cell lysate was used as the negative control and the WT testicular lysate was cytokines provoked by LPS stimulation. used as a positive control. Lysates of Axl and Mertk DKO (AM) testis were We also studied NF-kB activation by LPS in TKO macrophages. used as negative controls for Axl and Mertk expression, whereas that from Activation of the NF-kB pathway after LPS stimulation can be the Tyro3 and Mertk DKO testis was used as the negative control for Tyro3 assessed by IKK phosphorylation and IkB degradation (16, 50). and Mertk expression. b-actin was used to indicate the sample loading. LPS stimulation caused rapid phosphorylation of IKK1/2 and The Journal of Immunology 6169

FIGURE 3. Enhanced expression of IL-1b, IL-6, and iNOS by TKO microglia. (A) Real-time qPCR quantiﬁcation of cytokine production. Microglial cells were isolated from newborn mouse brains and cultured in vitro for 21 d before separation from the cocultured astrocytes by low concentration of trypsin digestion. After 24 h recovery, the microglia in culture dishes were stimulated by LPS (100 ng/ml), poly(I:C) (30 mg/ml), and CpG-oligo- deoxynucleotide 1668 (1 mM; InvivoGen, San Diego, CA), accordingly, for indicated time points. The total RNA was extracted by TRIzol and reverse transcribed using a qScript cDNA SuperMix kit. Real-time qPCR was performed to measure the relative mRNA levels of IL-1b, IL-6, and iNOS genes in the WT (s) and TKO (n) microglia. The mRNA expression levels relative to b-actin mRNA levels are shown as means 6 SD; n =3.**p , 0.05. (B) ELISA measurement of IL-6 by the LPS-, Downloaded from poly(I:C)-, and CpG-stimulated microglia or of TNF-a expression by the LPS-stimulated microglia. The microglial cells were prepared and treated as described above. At each time point, the culture medium was collected and subjected to either IL-6 or TNF-a measurement using speciﬁc ELISA kits

(eBioscience, San Diego, CA), and the assays were http://www.jimmunol.org/ performed following the manufacturer’s instruction. The concentration of IL-6 (B)orTNF-a (C)concentration in the WT (s)ortheTKO(n) microglia medium was plotted against stimulation times. Data are shown as means 6 SD for five wells per group in a single experiment and are representative of those in three experiments. *p , 0.05, n =3;**p , 0.01. by guest on September 24, 2021 degradation of IkB in both cell types, but the mutant cells dis- stimulation for 6 h, followed by extensive wash-off of residual played earlier IKK phosphorylation (15 min in TKO versus 30 LPS, and then the cells were transferred to normal medium for min in WT) and enhanced, prolonged IkB degradation (Fig. 4B), 24 h, after which the conditioned medium was collected and pu- suggesting that macrophages lacking TAM receptors respond to rified through a 0.2-mm filter and added to NSC cultures for 48 h, TLR activation with increased NF-kB activation and thus down- and then proliferating NSCs were labeled with BrdU (shown in stream target gene expression. Fig. 5A). Although conditioned medium from LPS-stimulated The results shown above indicate that activation of TLRs in TKO WT cells had an inhibitory effect on NSC proliferation (Fig. 5B, microglia or macrophages resulted in high and prolonged activation WT-by-LPS), the conditioned medium from LPS-treated TKO of MAPK and NF-kB signaling, which leads to elevated proin- microglia led to a more dramatic reduction in BrdU incorporation flammatory cytokine production, suggesting that TAM signaling into either WT or TKO proliferating NSCs (Fig. 5B, TKO-by- may inhibit TLR-induced proinflammatory cytokine production. LPS), indicating that TKO microglial cells released higher lev- Alternatively, we also examined whether activation of TAM els of neurotoxic mediators after LPS stimulation. Notably, TKO receptors on WT microglia by their ligand Gas6 would inhibit NSCs were more vulnerable to the neurotoxicity of LPS-treated innate immune responses. We pretreated cultured WT microglia conditioned medium than were WT NSCs (Fig. 5B). Furthermore, with Gas6 for 2 h, then with LPS and Gas6 for 6 h, and quantified LPS-stimulated TKO cell–conditioned medium caused more ex- levels of mRNAs for inflammatory factors by real-time qPCR. As tensive NSC death than did LPS-stimulated WT cell–conditioned shown in Fig. 4C, the LPS-induced expression of IL-1b, IL-6, and medium (Fig. 5C). Interestingly, close examination of both NSC iNOS mRNAs was inhibited by Gas6 pretreatment, further con- genotypes clearly showed that TKO NSCs were more committed firming that TAM receptors play a negative regulatory role on to programmed cell death than were WT cells. those cells. We next studied the effect of TKO-conditioned medium on NSC differentiation into b-tubulin III+ neurons. TKO microglia–con- Conditioned medium from LPS-treated TKO microglia inhibits ditioned medium in the absence of LPS preactivation had a sig- NSC proliferation and neuronal differentiation in vitro and this nificant inhibitory effect on WT NSC differentiation into b-tubulin effect is inhibited by anti–IL-6 Ab III+ neurons (Fig. 5D, open bars, TKO-by-PBS versus WT-by- Because mounting evidence showed that TKO microglia produced PBS), and this effect was as high as that using conditioned me- increased levels of proinflammatory cytokines, especially after dium from LPS-pretreated WT microglia (Fig. 5D, open bars, exposure to different TLR cognate ligands, we examined whether TKO-by-PBS versus WT-by-LPS). Interestingly, differentiation of conditioned medium from LPS-stimulated TKO microglia cul- TKO NSCs into b-tubulin III+ neurons was significantly reduced tures was neurotoxic for cultured NSCs. To generate conditioned compared to the WT cells, and LPS pretreatment had a negligible medium, primary cultures of microglia were subjected to LPS inhibitory effect (Fig. 5D, open bars), suggesting that TAM 6170 TAM RECEPTORS AND ADULT NEUROGENESIS

deed alleviated the neurotoxicity of the conditioned medium from TKO microglia (Fig. 5F). This result shows that the enhanced release of IL-6 from LPS-activated TKO microglia was responsible for the enhanced neuronal toxicity and suggests that this factor is a major candidate for the downregulation of hippocampal neurogenesis in TKO mice. Astrocytes, similar to microglia, produce higher levels of proinflammatory cytokines deleterious to adult neurogenesis Astrocytes have been postulated to exaggerate the neurotoxic effect of microglial inflammation by amplifying the release of inflammatory neurotoxic mediators (53). Given that astrocytes express TLRs and IL-1b receptors mediating the innate immune response (9, 49, 53–57) and express all three TAM receptor members (shown in Fig. 2), we postulated that the paracrine action of proinflammatory cytokines on TKO astrocytes might also be un- checked in the absence of negative regulation by TAMs. We therefore investigated inflammatory mediator production by cul-

tured astrocytes. Consistent with the effect seen on microglia (Fig. Downloaded from 3B), LPS-elicited IL-6 production by TKO astrocytes was sig- niﬁcantly higher than that seen with WT cells (Fig. 6A, 6C). Because astrocytes show high expression of IL-1b receptor (53), it was possible that the enhanced production of IL-1b by TKO microglia might stimulate astrocytes to generate more inﬂamma-

tory cytokines. This was indeed found to be the case, as, similar to http://www.jimmunol.org/ LPS, IL-1b stimulation triggered increased expression of mRNAs for IL-6 and IL-1b itself (Fig. 6B) and increased release of IL-6 into the culture medium (Fig. 6D). These data suggest that mutant astrocytes and microglia work together to impair hippocampal neurogenesis. LPS, mimicking pathogen infection, inhibits NSC proliferation FIGURE 4. Activation of TLR induces enhanced activation of MAPK in vivo A

and NF-kB signal transduction pathways. ( ) Western blotting analysis of by guest on September 24, 2021 Neuroinflammation inhibits adult NSC replication and differenti- p38 and Erk activation in the LPS-stimulated microglia. Microglia isolation and purification follow the procedure described in Materials and ation (24, 58). Because we previously showed that TAM receptors Methods. After 48 h recovery from mild trypsin digestion, both WT and played a negative role on microglia in the innate immune response TKO microglia were stimulated with 100 ng/ml LPS for the indicated time and that naive TKO mice exhibited reduced NSC proliferation points. By the end of activation, the equal numbers of cells were directly and differentiation into immature neurons, we then examined lysed into cold 13 NuPAGE LDS sample buffer (Invitrogen) and subjected whether systemic administration of LPS induced increased proin- to SDS-PAGE electrophoresis, transferred onto nitrocellular membrane, flammatory cytokine production, and, if so, whether this contrib- and subsequently immunoblotted for phospho-p38, total p38, phospho- uted to inhibition of neurogenesis. LPS is a potent inducer of B ERK, total ERK, and b-actin. ( ) Western blot analysis of IKK phos- microglia activation and proinflammatory cytokine release in the k phorylation and I B degradation in the purified peritoneal macrophage CNS (8, 59, 60). A single i.v. injection of LPS (1 mg/kg) resulted protein lysate. Western blots were probed with anti-IKK1/2 pSer, anti- in an increase in mRNA levels for the proinflammatory cytokines IKK2, and anti-IkB Abs. (C) TAM receptor ligand Gas6 inhibits proinflammatory cytokine production by microglia. The WT microglia were IL-1b, IL-6, and TNF-a in 8 h, and this effect was greater in the pretreated with 50 nM GAS6 for 2 h and then stimulated with 100 ng/ml TKO hippocampus (Fig. 7A). LPS pretreatment for 4 h prior to the LPS for 6 h. The RNA isolation, cDNA synthesis, and real-time qPCR initial BrdU injection significantly inhibited BrdU incorporation measurement of Il-1b, IL-6, and iNOS gene expression are described in into proliferating NSCs in WT brains, in which the percentage of Materials and Methods and in the legend to Fig. 3. the BrdU+ NSCs was reduced to a level similar to that seen in the untreated TKO brains (Fig. 7B, compare PBS in the TKO group receptors play an intrinsic functional role in supporting neuro- and LPS in the WT group). Notably, LPS treatment further re- genesis and that TAM receptor knockout renders NSCs more duced BrdU incorporation into TKO brains (Fig. 7B, PBS versus susceptible to toxic mediators in conditioned medium. LPS groups). LPS treatment also inhibited differentiation into In an effort to search for neurotoxic mediators, we focused on BrdU+/DCX+ immature neuronal progenies in the WT brains (Fig. one well-characterized proinflammatory cytokine, IL-6, which 7C, PBS versus LPS in the WT groups); however, this differen- could be detrimental to NSC renewal and survival (51, 52). ELISA tiation was still higher than in the naive TKO brain (Fig. 7C, TKO quantification clearly showed increased release of IL-6 by LPS- in PBS versus WT in LPS groups). Interestingly, examination of activated TKO microglia compared with LPS-activated WT cells BrdU+/DCX+ double-positive immature neurons in the TKO brain (Fig. 5E). To examine whether the toxicity of TKO microglia– revealed that LPS treatment did not cause a further decrease in the conditioned medium could be attributed to the enhanced produc- number of immature neurons (Fig. 7C, solid bars in the PBS and tion of IL-6, we prepared conditioned media from LPS-treated LPS groups). The percentage of immature neurons in the BrdU+ Il62/2 TKO and Il62/2 microglia culture and compared their subgranular zone (SGZ) cell population was significantly lower effects on WT NSC proliferation with those of the WT- and TKO- than that seen in WT brains (Fig. 7C), even when compared with conditioned media. Knockout of IL-6 in the TKO microglia in- LPS-stimulated WT brains, suggesting that TAM receptors play an The Journal of Immunology 6171 Downloaded from http://www.jimmunol.org/

FIGURE 5. The TKO microglia–conditioned medium inhibits NSC proliferation, neuronal differentiation, and promotes NSC death. (A) Schematic by guest on September 24, 2021 diagram shows how the conditioned medium is prepared. After mild trypsinization, both WT and TKO microglia were cultured for 48 h prior to 6 h of LPS (100 ng/ml) stimulation, after which the residual LPS was intensively washed off with sterile 13 PBS and replaced with fresh medium (DMEM/F12 containing 10% FBS and antibiotics) for preparation of microglia-conditioned medium (MG-cMD). The conditioned media prepared from the LPS- stimulated TKO or WT microglial cells are labeled as TKO-by-LPS or WT-by-LPS, respectively. The control conditioned medium was prepared from the cells treated with same volume of 13 PBS without LPS, and those MG-cMD are named either TKO-by-PBS or WT-by-PBS. The conditioned medium was collected 24 h later and passed through a 0.2-mm filter, and the resulting medium, served as MG-cMD, was subsequently mixed with 10% differentiation supplements (Invitrogen) and used in the following NSC culture for another 48 h. (B) Proliferating NSCs were labeled by addition of 10 mM BrdU into the culture medium in the last 12 h. The incorporated BrdU was revealed by anti-BrdU immunostaining. The percentage of BrdU+ cells in the total DAPI- stained cells were counted and calculated. (C) Apoptotic cells after 48 h of conditioned medium culture were identified by TUNEL assay (Roche) following the manufacturer’s instructions. The TUNEL+ cells in the DAPI-stained cells were calculated and plotted. (D) After the cells were cultured in the conditioned medium for 48 h, the medium was removed and the cells on the chamber slides were washed and fixed with 4% paraformaldehyde for 20 min at room temperature before immunostaining with anti–b-tubulin Ab. The b-tubulin+ cells in total DAPI-stained cells were counted and the percentage was calculated. (E) The IL-6 level in the conditioned medium was measured by ELISA. The cells were stimulated with LPS for 6 h and replaced with fresh medium without LPS, and the cell culture media used for ELISA assays were collected from the media either by the end of 6 h of LPS stimulation (LPS-6 h, chase 0 h) or by 24 h after chase with fresh medium (chase for 24 h after 6 h of LPS treatment) fed with fresh medium for 24 h. The IL-6 was measured by an ELISA kit (eBioscience) following the manufacturer’s instructions. (F) The MG-cMD was collected from primary cultured microglial cells prepared from the genotypes of WT, TKO, Il62/2Tyro32/2Axl2/2Mertk2/2 (Il62/2 TKO), and Il62/2 mice and was used to treat the WT NSCs for 48 h. The percentages of b-tubulin+ and BrdU+ cells were calculated and the fold changes over the WT-conditioned medium was plotted. The open and filled bars in (B)–(D) represent data obtained from the WT or TKO NSC culture, respectively. Means 6 SD are shown. *p , 0.05, n $ 3; **p , 0.01. intrinsic role in NSC differentiation and that loss of these recep- tion affects neurogenesis in adult brains, but TAM knockout cre- tors dramatically hampers differentiation into immature neurons in ates a harsher environment antagonizing the immunosuppressive the hippocampus. effect of indomethacin. Interestingly, when comparing the number To confirm that inflammation elicited by LPS stimulation was of BrdU+/DCX+ immature neurons in the PBS-treated TKO brains responsible for reduced NSC proliferation, we treated the mice with with those in the LPS-, LPS/indomethacin-, or indomethacin- the anti-inflammatory drug indomethacin prior to, and during, LPS treated TKO brains, LPS, LPS/indomethacin, or indomethacin treatment and BrdU labeling and found that administration of alone did not change the number of the BrdU+/DCX+ immature indomethacin to LPS-stimulated mice was able to antagonize the neurons, suggesting that LPS-triggered inflammation or an anti- inhibitory effects of LPS on hippocampal neurogenesis in WT inflammatory drug does not contribute much to the decreased mice, but had no effect in TKO mice (Fig. 7B, LPS versus LPS/ generation of BrdU+/DCX+ immature neurons seen in the PBS- indomethacin groups). This observation suggests that inflamma- treated TKO mice (Fig. 7C). However, whether those observations 6172 TAM RECEPTORS AND ADULT NEUROGENESIS

FIGURE 6. Enhanced innate immune response by TKO astrocytes. (A) Astrocytes were isolated and culture as described in Materials and Methods. After mild trypsin digestion and separation from microglia, astrocytes were recovered for 48 h before stimulation by LPS (100 ng/ml) for indicated time points. The IL-6 in the culture medium was quantiﬁed by an ELISA kit (eBioscience). (B) Real-time qPCR quantiﬁcation of IL-1b and IL-6 by WT and TKO astrocytes that had been stimulated with 10 ng/ml IL-1b for 4 h. By the end of IL-1b activation, the cells were directly lysed

into TRIzol reagents for total RNA isolation. cDNA Downloaded from synthesis and qPCR quantiﬁcation are described in Materials and Methods.(C and D) ELISA measurement of IL-6 in the astrocyte-conditioned medium after activation with LPS (C) or IL-1b (D). All data are shown as means 6 SD and experiments were repeated three times. **p , 0.05. http://www.jimmunol.org/

suggest an alternative Cox-independent mechanism accounting for that is, Tnfr1 (p55) and Tnfr2 (p75). To block the TNF-a action on by guest on September 24, 2021 the increased vulnerability of TAM TKO mice or an endogenous the target cells, we crossed the Tnfr12/2Tnfr22/2 into Axl2/2 regulatory role that the TAM receptors play in the hippocampal Mertk2/2 knockout mice to assess the hippocampal neurogenesis neuronal stem cells remains to be further elucidated. in the resulting compound knockout mice. Block of the TNF-a action by knocking out its receptor did not restore the impaired a 2 2 2 2 Knockout of IL-6 but not the TNF- receptor in the TAM TKO neurogenesis in the Axl / Mertk / DKO mice. These data or AM DKO mice restores neurogenesis suggest that overproduction of IL-6 by chronically activated TKO Direct evidence for proinflammatory cytokines being detrimental or DKO microglia is responsible for impaired hippocampal neu- to adult neurogenesis came from a study showing that transgenic rogenesis. mice overexpressing IL-6 uniquely in astrocytes showed impaired hippocampal neurogenesis (61). To further test the hypothesis that Discussion the increase in microglial cell–produced proinflammatory cyto- Mounting evidence shows that activated microglia produce mul- kines in TKO mice, especially that of IL-6, was deleterious to tiple proinflammatory cytokines that are essential for normal tissue hippocampal NSCs, we crossed Il62/2 mice with Axl2/2Mertk2/2 protection from pathogen invasion and tissue damage insults (62); DKO mice to generate Il62/2Axl2/2Mertk2/2 (Il62/2 DKO) however, uncontrolled activation of the innate immune system and compound gene knockout mice and assessed adult hippocampal an excessive inflammatory response in the CNS may contribute neurogenesis in these compound mutant mice at 2–3 mo old. The significantly to the pathogenesis and progression of many neuro- Il62/2 DKO mice displayed no additional overall defects com- degenerative diseases (9, 63, 64) and to impairment of adult pared to Axl2/2Mertk2/2 DKO parental strains. As assessed using neurogenesis (21, 24, 47, 65). The inhibitory function of TAM a 7-d BrdU labeling procedure (24), neurogenesis in the sub- receptors in innate immunity has received much attention recently granular DG of Il62/2 DKO mice exhibited a substantial recovery (15, 29, 31, 34, 46, 66). This family of receptors plays a regulatory compared with the WT control (Fig. 8A–E, n = 5). Additionally, role in DCs and monocyte-derived cells and causes late-stage differentiation of neuronal progenitor cells into the DCX/BrdU inhibition of cytokine receptor signaling either by upregulation double-positive immature neurons was also significantly im- of SOCS, which are inducible negative regulators of cytokine provedintheIl62/2 DKO mice versus the Axl2/2Mertk2/2 signaling (15, 34, 67), or by induction of expression of an NF-kB knockout mice (Fig. 8F). Consistent with the IL-6 effects on repressor, Twist, which binds to E-box elements in the TNF pro- neurogenesis, its receptor and co-receptor, gp130, were constantly moter and inhibits NF-kB–dependent transcription (29). Our detected in the RNA from hippocampal and NSCs by real-time present data demonstrated that TAM receptors on microglia also qPCR (Fig. 8G) and NSC protein lysate by Western blotting (Fig. have a suppressive effect on cytokine signaling. Microglia, a 8H). Furthermore, we also investigated the role of TNF-a on the brain-resident innate cell type with most of the major character- impaired neurogenesis. There are two subunits of TNF receptors, istics of peripheral macrophage, express all major TLRs (11–13) The Journal of Immunology 6173

FIGURE 7. Systemic inflammation inhibits neurogenesis and neuronal differentiation. Injection of LPS i.v. or i.p. caused increased expression of proinflammatory cytokines (A) and decreased the proliferating BrdU+ (B) and BrdU+/Dcx+ (C) cells numbers in the subgranular cell layer. (A) Eight hours after i.v. injection of LPS (1 mg/kg body weight), the total RNA was isolated from hippocampus of either WT or TKO mice and reverse transcribed into cDNA. The relative mRNA of IL-1b, IL-6, and TNF-a based on the b-actin mRNA level was measured by real-time qPCR. (B and C) The mice at 2 mo of age from each genotype were divided into four groups: PBS, LPS, LPS/Indocin (indomethacin), and Indocin. To induce acute inflammation, a single i.p. injection of LPS at dose of 1 mg/kg body weight was applied, 30 h prior to the first i.p. injection of BrdU (100 mg/kg body weight). The same dose of BrdU was given daily for the following 6 d. Downloaded from The mice were sacrificed 24 h after the last BrdU injection and processed as described in Fig. 1. In the groups treated with the anti-inflammatory drug indomethacin, the first injection of indomethacin at dose of 2.5 mg/kg body weight was given at 18 h prior to the first LPS injection and administrated twice daily at the same dosage. Data were collected from the WT (open http://www.jimmunol.org/ bars) and TKO (solid bars) brains and are expressed as average numbers of BrdU+ (B)andBrdU+/Dcx+ (C)cells per section at SGZ of bregma 21.70 mm. Results are shown as means 6 SEM. *p , 0.05; **p , 0.01, n =5.

and, in our study, activation with LPS, poly(I:C), or CpG triggered controls, TKO brains showed significantly lower BrdU incorpo- proinflammatory cytokine production. Notably, our data showed ration in dividing NSCs in the hippocampal subgranular layer, and by guest on September 24, 2021 that pathogen-induced proinflammatory cytokine expression was systemic administration of LPS, a potent inducer of proinflammatory much higher in TKO microglia than in WT controls, suggesting cytokine production as a result of its activating TLR4-initiated sig- that TAM receptors indeed function as negative regulators to keep naling in microglia and astrocytes, resulted in even more dramatic microglia cytokine signaling in check and maintain a homeostatic inhibition of neurogenesis. balance for microglia innate immune responses. Consistent with To further assess the neurotoxic role of TKO microglia and to test this observation, TKO microglia exhibited activated MAPK and whether the enhanced production of proinflammatory mediators by IKK/NF-kB signaling before treatment and responded to TLR TKO microglia inhibited NSC proliferation and differentiation, we activation with prolonged phosphorylation of MAPKs, especially used an in vitro culture system and exposed WT NSC cultures to p38, and increased degradation of IkB, which normally inhibits conditioned medium collected from TKO microglia cultures and NF-kB transcription activity. Interestingly, in microglia, p38 has found that TKO microglia released harmful factors that signifi- been postulated to be the major MAPK responsible for the pro- cantly inhibited WT NSC proliferation and neuronal differentiation duction of proinflammatory cytokines upon activation (17, 68, 69). and caused apoptosis, and that these effects were increased by LPS Inhibition of p38 by specific inhibitors or gene deletion impairs pretreatment of the cultured microglia, suggesting that unrestricted proinflammatory cytokine production by microglia in the inflam- innate immune responses by the TKO microglia released high level matory or aging brain (17, 68, 69). The unrestricted activation of of proinflammatory mediators that were detrimental to the cultured p38 and NF-kB signaling in TKO microglia is therefore considered NSCs. Recent studies of the effect of proinflammatory cytokines on a major signaling pathway for increased proinflammatory cytokine neurogenesis have shown that the detrimental or supportive effects production in the TKO brains. of microglia on neurogenesis are determined by their activation The unrestrained cytokine signaling and enhanced production status. In response to infection, most commonly mimicked by LPS of proinflammatory cytokines by TKO microglia inspired us to stimulation in animal models, activated microglia release proin- evaluate the effect of loss-of-function mutation on adult hippo- flammatory cytokines, such as IL-1b, IL-6, and TNF-a, which campal neurogenesis. Adult NSCs are restricted to a few defined have a negative effect on hippocampal neurogenesis and neuronal regions in the brain, including the subventricular zone and hip- differentiation (21–24, 27, 28, 86). However, microglia activated pocampal SGZ (70–74), and they are characterized by a self- with anti-inflammatory cytokines, such as IL-4, IL-15, or low- renewal capacity and the potential to differentiate into neurons dose IFN-g, support neurogenesis and glial differentiation (87– or glia (75–78). Adult NSCs with differentiation potential provide 90). Of the proinflammatory factors that suppress neurogenesis, a rich source for the replacement of cells lost during normal cell IL-6, TNF-a, and IL-1b have been intensely studied, and IL-6 has turnover and after brain injury (79–82). Interruption of adult been shown to play the major role based on the fact that IL-6– neurogenesis leads to impairment of hippocampus-dependent neutralizing Ab inhibits the negative effect of LPS-activated learning and behavior (24, 25, 79, 81, 83–85). Compared to WT microglia on neurogenesis (24). This inhibitory effect of IL-6 6174 TAM RECEPTORS AND ADULT NEUROGENESIS Downloaded from http://www.jimmunol.org/

FIGURE 8. Knockout of IL-6 in Axl2/2Mertk2/2 DKO mice restored neurogenesis. (A–D) Mice with four different genotypes of WT (A), Axl2/2Mertk2/2 DKO (B), Il62/2Axl2/2Mertk2/2 (Il62/2 DKO) (C), and IL62/2 (D), at age of 2 mo, were i.p. injected with 100 mg/kg weight of BrdU for 7 d. Forty- micrometer brain sections were cut and then stained by BrdU (green), Dcx (red), and Hoechst (blue). The immunostained sections at bregma at 1.94 mm are shown as representatives of the entire extent of DG analyzed between bregma 21.34 mm to bregma 23.08 mm. Scale bar, 100 mm. (E) Summary of the number of BrdU+ cells per 500 mm in length along the SVZ of DG of WT, Axl2/2Mertk2/2 (A2/2M2/2), Il62/2Axl2/2Mertk2/2 (Il62/2A2/2M2/2), Il62/2, Tnfr1/22/2,andTnfr1/22/2Axl2/2Mertk2/2 mice. Tnfr1/22/2 represents the DKO of the TNF receptor superfamilies 1 (p55) and 2 (p75). Shown are means 6 SD. (F) Percentage of the DCX/BrdU double-positive (d.p.) cells over the total BrdU+ cells in the SGZ of DG to indicated genotypes. Shown are means 6 SD. (G and H) Expression of IL-6 receptor and IL-6 coreceptor gp130 in hippocampal and primary cultured NSCs. Total RNA was isolated from ex vivo hippocampal tissue and primary NSC culture, and the expression of both IL-6 receptor and coreceptor gp130 was shown by RT- by guest on September 24, 2021 PCR (G). NSC lysate was subjected to Western blotting and detected with anti–IL-6 receptor Ab (H). The b-actin band shows equal loading of the protein samples. **p , 0.01, n =5. was also demonstrated in IL-6 transgenic mice, in which over- tokine production by activated microglia (91). Indomethacin, a non- expression of IL-6 uniquely in astrocytes dramatically inhibited steroidal anti-inflammatory drug, inhibits cyclooxygenase (COX)- neurogenesis (61). We indeed detected a higher level of IL-6 by mediated PG synthesis and thus inhibits LPS-induced systemic ELISA in the conditioned medium or by real-time qPCR from TKO inflammation (24). Notably, indomethacin did not alleviate the microglia with or without LPS activation. To further test whether impaired neurogenesis seen in untreated or LPS-treated TKO increased IL-6 produced by the mutant glial cells was responsible mice, suggesting that the brain local inflammation may play for the neurotoxic effect, we crossed Il62/2 mice with Axl2/2 a major role in the impaired neurogenesis seen in the adult TKO Mertk2/2 DKO mice to eliminate IL-6 expression in DKO mice and brain. There is no strong evidence that a single injection of LPS tested the effect on hippocampal neurogenesis and found that, in vivo can cause permanent neuronal damage and neurodegeneration compared with Axl2/2Mertk2/2 DKO or Tyro32/2Axl2/2Mertk2/2 (92), except in the case of dopaminergic neurons in the substantia TKO mice, the Il62/2Axl2/2Mertk2/2 knockout mice exhibited nigra (53, 93). However, long-term exposure to infectious pathogens significant improvement in adult hippocampal neurogenesis to has been shown to have detrimental effects on the nervous system levels approaching those in WT and Il62/2 mice. Meanwhile, by (19, 94–96). This again suggests that TKO brains undergo chronic using the similar approach, we blocked the TNF-a signaling by inflammation that is harmful for adult neurogenesis. knocking out the a and b subunits of TNF receptor 1 in the Axl2/2 Microglial cells actively patrol the CNS and sense any danger Mertk2/2 DKO mice and found no significant improvement for the signals. The next question was why TKO microglia were constantly impaired neurogenesis. This indicates that it is IL-6 but not TNF-a activated. On the one hand, TAM receptors play a trophic sup- responsible for the impaired neurogenesis in the mutants. porting role in a variety of cell types, which, in the absence of TAM LPS-induced inflammation has been shown to inhibit hippocampal receptors, undergo apoptosis (30, 40, 46). Additionally, all three neurogenesis (21, 24). Our results are in agreement with this finding, TAM receptors have been shown to be expressed in brain blood and furthermore, systemic administration of the anti-inflammatory vessel endothelial cells, and loss of function mutation destroys drug indomethacin to a great extent overcame the LPS-induced blood vessel integrity (97, 98), rendering the CNS highly vul- inhibition of hippocampal neurogenesis in the WT brain. LPS- nerable to local or peripheral insults (99). Microglia can efficiently induced inhibition of neurogenesis is thought to be caused by sense such risk factors and are constantly activated and produce primarily peripheral inflammatory mediators that penetrate the brain- neurotoxic levels of proinflammatory factors. On the other hand, blood barrier and cause strong upregulation of proinflammatory cy- TAM receptors themselves regulate inhibitory pathways that The Journal of Immunology 6175 prevent prolonged activation of microglia in a similar way to that References demonstrated for DCs and macrophages (15, 29, 31). We observed 1. Wake, H., A. J. Moorhouse, S. Jinno, S. Kohsaka, and J. Nabekura. 2009. constitutive activation of MAPKs, especially p38 MAPK, and Resting microglia directly monitor the functional state of synapses in vivo and determine the fate of ischemic terminals. J. Neurosci. 29: 3974–3980. increased IKK/NF-kB signaling in response to LPS stimulation. 2. Tremblay, M. E., R. L. Lowery, and A. K. Majewska. 2010. Microglial inter- Unrestricted activation of IKK and downstream NF-kB and sus- actions with synapses are modulated by visual experience. PLoS Biol. 8: tained secretion of proinflammatory cytokines have also been seen e1000527. 3. Ousman, S. S., and P. Kubes. 2012. Immune surveillance in the central nervous in apoptotic cell–induced inhibition of the inflammatory response system. Nat. Neurosci. 15: 1096–1101. by monocyte-derived DCs from Mertk knockout mice (66). In our 4. Prinz, M., J. Priller, S. S. Sisodia, and R. M. Ransohoff. 2011. Heterogeneity of CNS myeloid cells and their roles in neurodegeneration. Nat. Neurosci. 14: study, when stimulated with LPS in vitro, TKO microglia pro- 1227–1235. duced increased levels of proinflammatory cytokines, as shown at 5. Hanisch, U. K., and H. Kettenmann. 2007. Microglia: active sensor and ver- either the mRNA or protein level, creating a chronic inflammatory satile effector cells in the normal and pathologic brain. Nat. Neurosci. 10: 1387–1394. environment detrimental to adult brain stem cell renewal and 6. Lynch, M. A. 2010. Age-related neuroinflammatory changes negatively impact neuronal differentiation. on neuronal function. Front. Aging Neurosci 1: 6. TAM receptors may also play a role in supporting NSC dif- 7. Perry, V. H. 2004. The influence of systemic inflammation on inflammation in + the brain: implications for chronic neurodegenerative disease. Brain Behav. ferentiation into DCX immature neuroblasts or immature neu- Immun. 18: 407–413. rons. Given the NSCs expressing all three receptors (39, 100) and 8. Rivest, S. 2009. Regulation of innate immune responses in the brain. Nat. Rev. Immunol. 9: 429–439. the TAM receptor ligand, Gas6, lowering incorporation of BrdU 9. Glass, C. K., K. Saijo, B. Winner, M. C. Marchetto, and F. H. Gage. 2010. into the proliferating neuronal progenitor cells located in brain Mechanisms underlying inflammation in neurodegeneration. Cell 140: 918–934.

SVZ (39), the defective neurogenesis could also be caused by loss 10. Ekdahl, C. T., Z. Kokaia, and O. Lindvall. 2009. Brain inflammation and adult Downloaded from neurogenesis: the dual role of microglia. Neuroscience 158: 1021–1029. of direct support from TAM signaling on NSC proliferation and 11. Rivest, S. 2003. Molecular insights on the cerebral innate immune system. differentiation, in addition to the overreactive innate immune re- Brain Behav. Immun. 17: 13–19. sponse by TKO glial cells discussed above. Detailed comparison 12. Chakravarty, S., and M. Herkenham. 2005. Toll-like receptor 4 on non- + hematopoietic cells sustains CNS inflammation during endotoxemia, indepen- of differentiation of WT and TKO NSCs into b-tubulin III neu- dent of systemic cytokines. J. Neurosci. 25: 1788–1796. rons, presented in Fig. 5D, showed that a significantly lower 13. Reed-Geaghan, E. G., J. C. Savage, A. G. Hise, and G. E. Landreth. 2009. CD14 and Toll-like receptors 2 and 4 are required for fibrillar Ab-stimulated percentage of immature neurons was produced by TKO NSCs and microglial activation. J. Neurosci. 29: 11982–11992. http://www.jimmunol.org/ that LPS-activated microglia-conditioned medium did not result in 14. Dong, C., R. J. Davis, and R. A. Flavell. 2002. MAP kinases in the immune a further decrease, whereas WT NSCs showed an equal decrease response. Annu. Rev. Immunol. 20: 55–72. + 15. Rothlin, C. V., S. Ghosh, E. I. Zuniga, M. B. Oldstone, and G. Lemke. 2007. in differentiation into b-tubulin III neurons after exposure to the TAM receptors are pleiotropic inhibitors of the innate immune response. Cell microglia-conditioned medium from either PBS-treated TKO or 131: 1124–1136. LPS-treated TKO and WT cells, implying that TAM receptors 16. Li, Q., Q. Lu, V. Bottero, G. Estepa, L. Morrison, F. Mercurio, and I. M. Verma. 2005. Enhanced NF-kB activation and cellular function in macrophages lack- might play a functional role in NSC proliferation and neuronal ing IkB kinase 1 (IKK1). Proc. Natl. Acad. Sci. USA 102: 12425–12430. differentiation. In vivo LPS treatment not only inhibited NSC 17. Bachstetter, A. D., B. Xing, L. de Almeida, E. R. Dimayuga, D. M. Watterson, + + and L. J. Van Eldik. 2011. Microglial p38a MAPK is a key regulator of proliferation, but also hindered differentiation into BrdU /DCX proinflammatory cytokine up-regulation induced by toll-like receptor (TLR) neuronal progenies in WT brains, but this LPS inhibition of im- ligands or b-amyloid (Ab). J. Neuroinflammation 8: 79. by guest on September 24, 2021 mature neuron differentiation in the WT brain never reached the 18. Quan, N., M. Whiteside, L. Kim, and M. Herkenham. 1997. Induction of inhibitory factor kBa mRNA in the central nervous system after peripheral li- level seen in naive TKO brain, as shown in Fig. 7C. In contrast, popolysaccharide administration: an in situ hybridization histochemistry study LPS treatment did not cause a further decrease in the number of in the rat. Proc. Natl. Acad. Sci. USA 94: 10985–10990. immature neurons (BrdU+/DCX+) in the TKO brain, indicating 19. Lehnardt, S., L. Massillon, P. Follett, F. E. Jensen, R. Ratan, P. A. Rosenberg, J. J. Volpe, and T. Vartanian. 2003. Activation of innate immunity in the CNS that differentiation into immature neurons may be affected by triggers neurodegeneration through a Toll-like receptor 4-dependent pathway. other endogenous factors normally under positive regulation by Proc. Natl. Acad. Sci. USA 100: 8514–8519. 20. Rolls, A., R. Shechter, A. London, Y. Ziv, A. Ronen, R. Levy, and M. Schwartz. TAM receptors and/or that TAM knockout had already exerted 2007. Toll-like receptors modulate adult hippocampal neurogenesis. Nat. Cell a maximal deleterious effect on immature neuron production. Biol. 9: 1081–1088. In summary, TAM receptors are expressed on microglia. Loss of 21. Ekdahl, C. T., J. H. Claasen, S. Bonde, Z. Kokaia, and O. Lindvall. 2003. In- flammation is detrimental for neurogenesis in adult brain. Proc. Natl. Acad. Sci. these receptors results in adult hippocampal neurogenesis being USA 100: 13632–13637. significantly comprised. TAM receptors have a negative regulatory 22. Kuzumaki, N., D. Ikegami, S. Imai, M. Narita, R. Tamura, M. Yajima, effect on microglia to prevent overproduction of proinflammatory A. Suzuki, K. Miyashita, K. Niikura, H. Takeshima, et al. 2010. Enhanced IL- 1b production in response to the activation of hippocampal glial cells impairs mediators detrimental to NSC proliferation and differentiation. Loss neurogenesis in aged mice. Synapse 64: 721–728. of TAM receptors that the microglia produces enhanced IL-6 and 23. Iosif, R. E., C. T. Ekdahl, H. Ahlenius, C. J. Pronk, S. Bonde, Z. Kokaia, S. E. Jacobsen, and O. Lindvall. 2006. Tumor necrosis factor receptor 1 is IL-1b, and IL-1b acting as a paracrine factor, in turn, stimulates a negative regulator of progenitor proliferation in adult hippocampal neuro- astrocytes to produce more IL-6. Overproduction of IL-6 by both genesis. J. Neurosci. 26: 9703–9712. microglia and astrocytes inhibits NSC proliferation and differen- 24. Monje, M. L., H. Toda, and T. D. Palmer. 2003. Inflammatory blockade restores adult hippocampal neurogenesis. Science 302: 1760–1765. tiation, leading to an impaired neurogenesis in adult hippocampus. 25. Monje, M. L., S. Mizumatsu, J. R. Fike, and T. D. Palmer. 2002. Irradiation Additionally, autonomous trophic function of the TAM receptors induces neural precursor-cell dysfunction. Nat. Med. 8: 955–962. on the proliferating neuronal progenitors may also promote neu- 26. Picard-Riera, N., L. Decker, C. Delarasse, K. Goude, B. Nait-Oumesmar, R. Liblau, D. Pham-Dinh, and A. Baron-Van Evercooren. 2002. Experimental ronal differentiation of progenitor into immature neurons. autoimmune encephalomyelitis mobilizes neural progenitors from the subventricular zone to undergo oligodendrogenesis in adult mice. Proc. Natl. Acad. Sci. USA 99: 13211–13216. Acknowledgments 27. Cacci, E., M. A. Ajmone-Cat, T. Anelli, S. Biagioni, and L. Minghetti. 2008. In vitro neuronal and glial differentiation from embryonic or adult neural We thank the Research to Prevent Blindness Foundation and the core fa- precursor cells are differently affected by chronic or acute activation of cility at the Department of Ophthalmology, University of Louisville, for microglia. Glia 56: 412–425. general support. 28. Liu, Y. P., H. I. Lin, and S. F. Tzeng. 2005. Tumor necrosis factor-a and interleukin-18 modulate neuronal cell fate in embryonic neural progenitor culture. Brain Res. 1054: 152–158. 29. Sharif, M. N., D. Sosic, C. V. Rothlin, E. Kelly, G. Lemke, E. N. Olson, and Disclosures L. B. Ivashkiv. 2006. Twist mediates suppression of inflammation by type I The authors have no financial conflicts of interest. IFNs and Axl. J. Exp. Med. 203: 1891–1901. 6176 TAM RECEPTORS AND ADULT NEUROGENESIS

30. Lu, Q., and G. Lemke. 2001. Homeostatic regulation of the immune system by 59. Frank, M. G., M. V. Baratta, D. B. Sprunger, L. R. Watkins, and S. F. Maier. 2007. receptor tyrosine kinases of the Tyro 3 family. Science 293: 306–311. Microglia serve as a neuroimmune substrate for stress-induced potentiation of 31. Ye, F., L. Han, Q. Lu, W. Dong, Z. Chen, H. Shao, H. J. Kaplan, Q. Li, and CNS pro-inflammatory cytokine responses. Brain Behav. Immun. 21: 47–59. Q. Lu. 2011. Retinal self-antigen induces a predominantly Th1 effector re- 60. Cunningham, C., D. C. Wilcockson, S. Campion, K. Lunnon, and V. H. Perry. sponse in Axl and Mertk double-knockout mice. J. Immunol. 187: 4178–4186. 2005. Central and systemic endotoxin challenges exacerbate the local inflam- 32. Prasad, D., C. V. Rothlin, P. Burrola, T. Burstyn-Cohen, Q. Lu, P. Garcia de matory response and increase neuronal death during chronic neuro- Frutos, and G. Lemke. 2006. TAM receptor function in the retinal pigment degeneration. J. Neurosci. 25: 9275–9284. epithelium. Mol. Cell. Neurosci. 33: 96–108. 61. Vallie`res, L., I. L. Campbell, F. H. Gage, and P. E. Sawchenko. 2002. Reduced 33. Camenisch, T. D., B. H. Koller, H. S. Earp, and G. K. Matsushima. 1999. A hippocampal neurogenesis in adult transgenic mice with chronic astrocytic novel receptor tyrosine kinase, Mer, inhibits TNF-a production and production of interleukin-6. J. Neurosci. 22: 486–492. lipopolysaccharide-induced endotoxic shock. J. Immunol. 162: 3498–3503. 62. Fetler, L., and S. Amigorena. 2005. Neuroscience. Brain under surveillance: the 34. Lemke, G., and C. V. Rothlin. 2008. Immunobiology of the TAM receptors. microglia patrol. Science 309: 392–393. Nat. Rev. Immunol. 8: 327–336. 63. Wyss-Coray, T., and L. Mucke. 2002. Inflammation in neurodegenerative dis- 35. Seitz, H. M., T. D. Camenisch, G. Lemke, H. S. Earp, and G. K. Matsushima. ease: a double-edged sword. Neuron 35: 419–432. 2007. Macrophages and dendritic cells use different Axl/Mertk/Tyro3 receptors 64. Minghetti, L. 2005. Role of inflammation in neurodegenerative diseases. Curr. in clearance of apoptotic cells. J. Immunol. 178: 5635–5642. Opin. Neurol. 18: 315–321. 36. Lemke, G., and Q. Lu. 2003. Macrophage regulation by Tyro 3 family recep- 65. Voloboueva, L. A., and R. G. Giffard. 2011. Inflammation, mitochondria, and tors. Curr. Opin. Immunol. 15: 31–36. the inhibition of adult neurogenesis. J. Neurosci. Res. 89: 1989–1996. 37. Cohen, P. L., R. Caricchio, V. Abraham, T. D. Camenisch, J. C. Jennette, 66. Sen, P., M. A. Wallet, Z. Yi, Y. Huang, M. Henderson, C. E. Mathews, R. A. Roubey, H. S. Earp, G. Matsushima, and E. A. Reap. 2002. Delayed H. S. Earp, G. Matsushima, A. S. Baldwin, Jr., and R. M. Tisch. 2007. Apo- apoptotic cell clearance and lupus-like autoimmunity in mice lacking the c-mer ptotic cells induce Mer tyrosine kinase-dependent blockade of NF-kB activa- membrane tyrosine kinase. J. Exp. Med. 196: 135–140. tion in dendritic cells. Blood 109: 653–660. 38. Xiong, W., Y. Chen, H. Wang, H. Wang, H. Wu, Q. Lu, and D. Han. 2008. Gas6 67. Hanada, T., H. Yoshida, S. Kato, K. Tanaka, K. Masutani, J. Tsukada, and the Tyro 3 receptor tyrosine kinase subfamily regulate the phagocytic Y. Nomura, H. Mimata, M. Kubo, and A. Yoshimura. 2003. Suppressor of function of Sertoli cells. Reproduction 135: 77–87. cytokine signaling-1 is essential for suppressing dendritic cell activation and 39. Gely-Pernot, A., V. Coronas, T. Harnois, L. Prestoz, N. Mandairon, A. Didier, systemic autoimmunity. Immunity 19: 437–450. Downloaded from J. M. Berjeaud, A. Monvoisin, N. Bourmeyster, P. G. De Frutos, et al. 2012. An 68. Lloyd, E., K. Somera-Molina, L. J. Van Eldik, D. M. Watterson, and endogenous vitamin K-dependent mechanism regulates cell proliferation in the M. S. Wainwright. 2008. Suppression of acute proinflammatory cytokine and brain subventricular stem cell niche. Stem Cells 30: 719–731. chemokine upregulation by post-injury administration of a novel small mole- 40. Lu, Q., M. Gore, Q. Zhang, T. Camenisch, S. Boast, F. Casagranda, C. Lai, cule improves long-term neurologic outcome in a mouse model of traumatic M. K. Skinner, R. Klein, G. K. Matsushima, et al. 1999. Tyro-3 family brain injury. J. Neuroinflammation 5: 28. receptors are essential regulators of mammalian spermatogenesis. Nature 398: 69. Munoz, L., H. Ralay Ranaivo, S. M. Roy, W. Hu, J. M. Craft, L. K. McNamara, 723–728. L. W. Chico, L. J. Van Eldik, and D. M. Watterson. 2007. A novel p38 a MAPK

41. Ye, F., Q. Li, Y. Ke, Q. Lu, L. Han, H. J. Kaplan, H. Shao, and Q. Lu. 2011. inhibitor suppresses brain proinflammatory cytokine up-regulation and http://www.jimmunol.org/ TAM receptor knockout mice are susceptible to retinal autoimmune induction. attenuates synaptic dysfunction and behavioral deficits in an Alzheimer’s disease Invest. Ophthalmol. Vis. Sci. 52: 4239–4246. mouse model. J. Neuroinflammation 4: 21. 42. Prieto, A. L., S. O’Dell, B. Varnum, and C. Lai. 2007. Localization and sig- 70. Gould, E., A. J. Reeves, M. Fallah, P. Tanapat, C. G. Gross, and E. Fuchs. 1999. naling of the receptor protein tyrosine kinase Tyro3 in cortical and hippocampal Hippocampal neurogenesis in adult Old World primates. Proc. Natl. Acad. Sci. neurons. Neuroscience 150: 319–334. USA 96: 5263–5267. 43. Prieto, A. L., J. L. Weber, and C. Lai. 2000. Expression of the receptor protein- 71. Eriksson, P. S., E. Perfilieva, T. Bjo¨rk-Eriksson, A. M. Alborn, C. Nordborg, tyrosine kinases Tyro-3, Axl, and mer in the developing rat central nervous D. A. Peterson, and F. H. Gage. 1998. Neurogenesis in the adult human hip- system. J. Comp. Neurol. 425: 295–314. pocampus. Nat. Med. 4: 1313–1317. 44. Ellis, P., B. M. Fagan, S. T. Magness, S. Hutton, O. Taranova, S. Hayashi, 72. Kempermann, G., and F. H. Gage. 1999. New nerve cells for the adult brain. A. McMahon, M. Rao, and L. Pevny. 2004. SOX2, a persistent marker for Sci. Am. 280: 48–53. multipotential neural stem cells derived from embryonic stem cells, the embryo 73. Gage, F. H. 2000. Mammalian neural stem cells. Science 287: 1433–1438.

or the adult. Dev. Neurosci. 26: 148–165. 74. Okano, H. 2002. Stem cell biology of the central nervous system. J. Neurosci. by guest on September 24, 2021 45. Kim, J. B., H. Zaehres, G. Wu, L. Gentile, K. Ko, V. Sebastiano, M. J. Arau´zo- Res. 69: 698–707. Bravo, D. Ruau, D. W. Han, M. Zenke, and H. R. Scho¨ler. 2008. Pluripotent 75. Toni, N., E. M. Teng, E. A. Bushong, J. B. Aimone, C. Zhao, A. Consiglio, stem cells induced from adult neural stem cells by reprogramming with two H. van Praag, M. E. Martone, M. H. Ellisman, and F. H. Gage. 2007. Synapse factors. Nature 454: 646–650. formation on neurons born in the adult hippocampus. Nat. Neurosci. 10: 727– 46. Scott, R. S., E. J. McMahon, S. M. Pop, E. A. Reap, R. Caricchio, P. L. Cohen, 734. H. S. Earp, and G. K. Matsushima. 2001. Phagocytosis and clearance of apo- 76. Toni, N., D. A. Laplagne, C. Zhao, G. Lombardi, C. E. Ribak, F. H. Gage, and ptotic cells is mediated by MER. Nature 411: 207–211. A. F. Schinder. 2008. Neurons born in the adult dentate gyrus form functional 47. Stanojcic, M., T. Burstyn-Cohen, N. Nashi, G. Lemke, and B. Sakic. 2009. synapses with target cells. Nat. Neurosci. 11: 901–907. Disturbed distribution of proliferative brain cells during lupus-like disease. 77. Marıń-Burgin, A., L. A. Mongiat, M. B. Pardi, and A. F. Schinder. 2012. Brain Behav. Immun. 23: 1003–1013. Unique processing during a period of high excitation/inhibition balance in 48. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. adult-born neurons. Science 335: 1238–1242. Immunol. 21: 335–376. 78. Ming, G. L., and H. Song. 2011. Adult neurogenesis in the mammalian brain: 49. Olson, J. K., and S. D. Miller. 2004. Microglia initiate central nervous system significant answers and significant questions. Neuron 70: 687–702. innate and adaptive immune responses through multiple TLRs. J. Immunol. 79. Zhao, C., W. Deng, and F. H. Gage. 2008. Mechanisms and functional impli- 173: 3916–3924. cations of adult neurogenesis. Cell 132: 645–660. 50. Li, Q., and I. M. Verma. 2002. NF-kB regulation in the immune system. Nat. 80. van Praag, H., A. F. Schinder, B. R. Christie, N. Toni, T. D. Palmer, and Rev. Immunol. 2: 725–734. F. H. Gage. 2002. Functional neurogenesis in the adult hippocampus. Nature 51. Bauer, S. 2009. Cytokine control of adult neural stem cells. Ann. N. Y. Acad. 415: 1030–1034. Sci. 1153: 48–56. 81. Zhang, C. L., Y. Zou, W. He, F. H. Gage, and R. M. Evans. 2008. A role for 52. Carpentier, P. A., and T. D. Palmer. 2009. Immune influence on adult neural adult TLX-positive neural stem cells in learning and behaviour. Nature 451: stem cell regulation and function. Neuron 64: 79–92. 1004–1007. 53. Saijo, K., B. Winner, C. T. Carson, J. G. Collier, L. Boyer, M. G. Rosenfeld, 82. Deng, W., J. B. Aimone, and F. H. Gage. 2010. New neurons and new mem- F. H. Gage, and C. K. Glass. 2009. A Nurr1/CoREST pathway in microglia and ories: how does adult hippocampal neurogenesis affect learning and memory? astrocytes protects dopaminergic neurons from inflammation-induced death. Nat. Rev. Neurosci. 11: 339–350. Cell 137: 47–59. 83. Madsen, T. M., P. E. Kristjansen, T. G. Bolwig, and G. Wo¨rtwein. 2003. 54. Carpentier, P. A., W. S. Begolka, J. K. Olson, A. Elhofy, W. J. Karpus, and Arrested neuronal proliferation and impaired hippocampal function following S. D. Miller. 2005. Differential activation of astrocytes by innate and adaptive fractionated brain irradiation in the adult rat. Neuroscience 119: 635–642. immune stimuli. Glia 49: 360–374. 84. Shors, T. J., G. Miesegaes, A. Beylin, M. Zhao, T. Rydel, and E. Gould. 2001. 55. Gorina, R., M. Font-Nieves, L. Ma´rquez-Kisinousky, T. Santalucia, and Neurogenesis in the adult is involved in the formation of trace memories. A. M. Planas. 2011. Astrocyte TLR4 activation induces a proinflammatory Nature 410: 372–376. environment through the interplay between MyD88-dependent NFkB signaling, 85. Dupret, D., J. M. Revest, M. Koehl, F. Ichas, F. De Giorgi, P. Costet, MAPK, and Jak1/Stat1 pathways. Glia 59: 242–255. D. N. Abrous, and P. V. Piazza. 2008. Spatial relational memory requires 56. Welser-Alves, J. V., S. J. Crocker, and R. Milner. 2011. A dual role for microglia hippocampal adult neurogenesis. PLoS One 3: e1959. in promoting tissue inhibitor of metalloproteinase (TIMP) expression in glial 86. Wu, M. D., A. M. Hein, M. J. Moravan, S. S. Shaftel, J. A. Olschowka, and cells in response to neuroinflammatory stimuli. J. Neuroinflammation 8: 61. M. K. O’Banion. 2012. Adult murine hippocampal neurogenesis is inhibited by 57. Farina, C., F. Aloisi, and E. Meinl. 2007. Astrocytes are active players in ce- sustained IL-1b and not rescued by voluntary running. Brain Behav. Immun. 26: rebral innate immunity. Trends Immunol. 28: 138–145. 292–300. 58. Huehnchen, P., T. Prozorovski, P. Klaissle, A. Lesemann, J. Ingwersen, 87. Butovsky, O., Y. Ziv, A. Schwartz, G. Landa, A. E. Talpalar, S. Pluchino, S. A. Wolf, A. Kupsch, O. Aktas, and B. Steiner. 2011. Modulation of adult G. Martino, and M. Schwartz. 2006. Microglia activated by IL-4 or IFN-g hippocampal neurogenesis during myelin-directed autoimmune neuro- differentially induce neurogenesis and oligodendrogenesis from adult stem/ inflammation. Glia 59: 132–142. progenitor cells. Mol. Cell. Neurosci. 31: 149–160. The Journal of Immunology 6177

88. Butovsky, O., A. E. Talpalar, K. Ben-Yaakov, and M. Schwartz. 2005. Acti- TLR4 is necessary for lipopolysaccharide-induced oligodendrocyte injury in vation of microglia by aggregated b-amyloid or lipopolysaccharide impairs the CNS. J. Neurosci. 22: 2478–2486. MHC-II expression and renders them cytotoxic whereas IFN-g and IL-4 render 95. Lehnardt, S., P. Henneke, E. Lien, D. L. Kasper, J. J. Volpe, I. Bechmann, them protective. Mol. Cell. Neurosci. 29: 381–393. R. Nitsch, J. R. Weber, D. T. Golenbock, and T. Vartanian. 2006. A 89. Go´mez-Nicola, D., B. Valle-Argos, N. Pallas-Bazarra, and M. Nieto-Sampedro. mechanism for neurodegeneration induced by group B streptococci 2011. Interleukin-15 regulates proliferation and self-renewal of adult neural through activation of the TLR2/MyD88 pathway in microglia. J. Immunol. stem cells. Mol. Biol. Cell 22: 1960–1970. 177: 583–592. 90. Walton, N. M., B. M. Sutter, E. D. Laywell, L. H. Levkoff, S. M. Kearns, 96. Nau, R., and W. Bru¨ck. 2002. Neuronal injury in bacterial meningitis: mech- G. P. Marshall, II, B. Scheffler, and D. A. Steindler. 2006. Microglia instruct anisms and implications for therapy. Trends Neurosci. 25: 38–45. subventricular zone neurogenesis. Glia 54: 815–825. 97. Korshunov, V. A., A. M. Mohan, M. A. Georger, and B. C. Berk. 2006. Axl, 91. Turrin, N. P., D. Gayle, S. E. Ilyin, M. C. Flynn, W. Langhans, G. J. Schwartz, and a receptor tyrosine kinase, mediates flow-induced vascular remodeling. Circ. C. R. Plata-Salamań. 2001. Pro-inflammatory and anti-inflammatory cytokine Res. 98: 1446–1452. mRNA induction in the periphery and brain following intraperitoneal adminis- 98. Zhu, D., Y. Wang, I. Singh, R. D. Bell, R. Deane, Z. Zhong, A. Sagare, tration of bacterial lipopolysaccharide. Brain Res. Bull. 54: 443–453. E. A. Winkler, and B. V. Zlokovic. 2010. Protein S controls hypoxic/ischemic 92. Kim, W. G., R. P. Mohney, B. Wilson, G. H. Jeohn, B. Liu, and J. S. Hong. blood-brain barrier disruption through the TAM receptor Tyro3 and sphingosine 2000. Regional difference in susceptibility to lipopolysaccharide-induced 1-phosphate receptor. Blood 115: 4963–4972. neurotoxicity in the rat brain: role of microglia. J. Neurosci. 20: 6309–6316. 99. Li, Q., Q. Lu, H. Lu, S. Tian, and Q. Lu. 2013. Systemic autoimmunity in TAM 93. Castanõ, A., A. J. Herrera, J. Cano, and A. Machado. 1998. Lipopolysaccharide triple knockout mice causes inflammatory brain damage and cell death. PLoS intranigral injection induces inflammatory reaction and damage in nigrostriatal ONE 8: e64812. dopaminergic system. J. Neurochem. 70: 1584–1592. 100. Wang, J., H. Zhang, A. G. Young, R. Qiu, S. Argalian, X. Li, X. Wu, G. Lemke, 94. Lehnardt, S., C. Lachance, S. Patrizi, S. Lefebvre, P. L. Follett, F. E. Jensen, and Q. Lu. 2011. Transcriptome analysis of neural progenitor cells by a genetic P. A. Rosenberg, J. J. Volpe, and T. Vartanian. 2002. The Toll-like receptor dual reporter strategy. Stem Cells 29: 1589–1600. Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021