Just as turning down the gas lowers the heat in the kitchen, dialing back an inflammatory signaling pathway quenches amyloid-β in the brain, according to the November 25 Nature Medicine. Scientists led by Burkhard Becher, University of Zurich, Switzerland, and Frank Heppner, Charité–Universitätsmedizin Berlin, Germany, found that simultaneously mollifying two microglial cytokines—IL-12 and IL-23—prevents plaque buildup and improves behavior in mouse models of Alzheimer’s disease (AD). These same cytokines flare up in the cerebrospinal fluid of human AD patients, the authors report. The new study adds to mounting evidence that neuroinflammation fans AD pathology and suggests that blocking relevant cytokine pathways could slow disease development. Ustekinumab, an antibody to IL-12/23, already received FDA approval for the treatment of psoriasis, an autoimmune disease of the skin. The AD research comes hot on the heels of the news that mutations in the microglial receptor TREM2 triple AD risk (see ARF related news story).

Released by microglia, IL-12 and IL-23 stimulate maturation of naive T cells, which are part of the adaptive immune system. Each cytokine is made up of two subunits, with one, p40, being common to both. The other subunits, p35 and p19, are specific to IL-12 and IL-23, respectively. The two cytokines have been implicated in immune diseases such as psoriasis, multiple sclerosis, and Crohn’s, but scientists had not yet looked at their potential contribution to AD. A few hints have surfaced, however. A recent multi-analyte profiling study in MCI and AD patients reported that higher plasma p40 levels matched up with lower cognitive performance on the MMSE (see ARF related news story). What’s more, Aβ immunization in mice lowered T cell expression of IL-12RB1, a major component of receptors for both cytokines (see Town et al., 2002). But where is the common p40 subunit released? On which cells does it act? And does manipulating its level modulate AD pathology?

To answer those questions, joint first authors Johannes vom Berg and Stefan Prokop first compared microglia in wild-type older mice with microglia in APPPS1 transgenic mice, which develop early and aggressive cerebral amyloidosis. Not only were microglia in APPPS1 mice more activated, as others had found previously (see Frautschy et al., 1998), but they also produced more mRNA for p40, IL-12, and IL-23. When the researchers crossed APPPS1 mice with mice lacking p40, p35, or p19, cortical plaque load fell in four-month-old knockouts relative to APPPS1 controls. Mice without p40, and therefore without both IL-12 and IL-23, showed the most marked effect—a 63 percent decline. At eight months, APPPS1 p40 knockouts also had fewer microglia, their astrocytes were less aggravated, and their soluble and insoluble Aβ40 and 42 in the brain were half those in regular APPPS1 mice. Together, the results suggest that p40, and therefore IL-12 and IL-23, play a role in AD pathology, and that reining in those cytokines might suppress Aβ.

What cells make these cytokines? One major question in AD research is whether reactive glia reside in the brain or migrate in from the body’s peripheral blood circulation (see ARF related news story). To find out which pool of cells mediated the p40 effects, the researchers irradiated APPPS1 and APPPS1 p40 knockout mice to eliminate their peripheral macrophages (the skull protects brain cells from the radiation used in this experiment). They then replenished peripheral cells by injecting bone marrow from wild-type or p40 knockouts. With this approach, the scientists created mice with p40 restricted to either the central nervous system or the blood. The former accumulated plaque loads comparable to APPPS1 mice, while the latter generated far fewer plaques. “This told us that the p40 action takes place within the brain, not outside,” said Heppner.

“As far as I am aware, this is the first time that IL-12 and IL-23 have been implicated directly in an innate immune phenomenon,” said Richard Ransohoff of the Cleveland Clinic, Ohio, who was not involved with the work. He suggested it would be important next to characterize tau pathology in Heppner’s mice, since work by others suggests a tradeoff between stronger amyloid clearance and accelerated tau pathology (see Lee et al., 2010, and Bhaskar et al., 2010).

Could drugs mimic p40 knockout effects? Based on this study, it appears so. Compared to untreated mice, one-third fewer plaques developed in APPPS1 mice that got a twice-weekly injection of an anti-p40 antibody starting when they were four weeks of age, before the onset of plaques, and continuing until they were four months old. The same treatment helped older mice (six months old) with fully established Aβ pathology as well. A 60-day course of anti-p40 delivered directly to the brain rescued short-term memory deficits in the Barnes maze test and in a novel object recognition task. While plaque load did not budge in these older mice, soluble Aβ species in brain homogenates of these animals took a dive. The results hint that drugs targeting IL-12 and IL-23 could treat AD pathology and cognitive decline in humans, the scientists believe. In indirect support of this, Heppner and colleagues detected more p40 in the CSF of 39 AD patients compared to 20 controls. Furthermore, p40 levels correlated with worse cognitive performance on the MMSE.

IL-12 and IL-23 are two of many cytokines that control inflammatory and non-inflammatory signaling in glia. Researchers are only just beginning to understand how they all fit with AD pathology (see ARF related news story). While IL-12/23 may promote Aβ pathology in mice, other inflammatory cytokines, including IL-6 (see Chakrabarty et al., 2010) and interleukin 1β (Matousek et al., 2012), might help clear the peptide. On the other hand, blocking CD40 (see ARF related news story) or TGF-β (see ARF related news story) achieves a similar effect.

Vom Berg and colleagues add to mounting evidence that scientists will probably have to pick and choose which parts of the immune system to stimulate or suppress, said Terrence Town, Cedars-Sinai Medical Center, Los Angeles, California. "Pan-blocking inflammation is not going to be the way forward against AD. We have to focus on these very specific molecular cascades,” said Town. “This work demonstrates that blocking IL-12/IL-23 signaling could well be a critically important therapeutic target,” he added.

Heppner says his group plans to probe effects downstream of p40, while testing ustekinumab (see Weber and Keam, 2009), a human antibody to p40, in people at risk for AD. The antibody has been used to treat psoriasis since 2009, and “could be tested in a Phase 2 trial pretty quickly,” Heppner told Alzforum. Phase 3 trials are also underway to test its effectiveness against Crohn’s disease (see Sandborn et al., 2012), and a Phase 2 trial will examine whether ustekinumab can treat rheumatoid arthritis. The antibody did not pass primary endpoints in a Phase 2 trial for relapsing-remitting multiple sclerosis (see Segal et al., 2008).—Gwyneth Dickey Zakaib


  1. Microglial Cytokines Light the Fire—Interleukin 12-, Interleukin 23-Mediated Inflammation May Act as Driving Force Behind Disease Progression
    The paper by vom Berg and colleagues further strengthens the hypothesis that innate immune reactions may strongly contribute to the pathogenesis of Alzheimer's disease (AD). The neuroinflammatory component of the disease, which has been for long the stepchild of AD research, is recently gaining increasing interest, instigated by findings which point to an involvement of inflammatory genes as risk factors for sporadic AD (Jones et al., 2010; Guerreiro et al., 2012; Jonsson et al., 2012). Microglial cells, representing the brain's innate immune defense, mount a sterile inflammatory reaction in response to Aβ deposition and most likely other immunostimulants in the degenerating brain (Lucin and Wyss Coray, 2009). One key feature of this reaction is the release of proinflammatory cytokines, which may further drive microglial inflammation, contribute to neuronal dysfunction and death, and also stimulate and recruit astroglial cells to the site of sterile inflammation. In their paper, the authors provide compelling evidence for a disease-promoting role of the interleukin 12, -23 signaling pathway. Genetic ablation of the common molecule p40, or the p35 and p19 components, decreased the cerebral Aβ load in APP/PS1 transgenic mice and ameliorated behavioral deficits. Likewise, intracerebroventricular administration of neutralizing p40 antibodies lowered soluble Aβ peptides and improved spatial memory. Importantly, the authors were able to identify resident microglia as responsible cells for interleukin 12, interleukin 23 generation, and a contribution from peripheral cells was proven unlikely. Another important finding is that genetic ablation of the interleukin-12/23 signaling pathway did not alter APP processing. Since the p40 receptor is strongly expressed on astrocytes, the authors suggest that microglial p40 may lead to Aβ reduction through the stimulation of astroglial Aβ uptake. However, astrocytes may interact with Aβ removal at multiple levels, including the release and lipidation of ApoE, which in turn can regulate microglial Aβ phagocytosis (Terwell et al., 2011).

    Findings of increased p40 levels in human CSF from AD patients suggest that the interleukin-12/23 signaling pathway is activated in AD. However, further studies will have to identify the precise time course of this particular immunological pathway in AD before it may be exploited as a therapeutic target. Cerebral deposition of immunostimulating Aβ most likely occurs years, if not decades, before the presentation of clinical symptoms. While inflammation normally is of a self-limiting nature, in AD the persistent Aβ deposition acts as a constant stimulus which helps to establish a chronic inflammatory state. Given the detrimental effects of sustained neuroinflammation, a potential therapy will have to interfere with inflammatory mechanisms at the earliest time point possible, and therefore has to be of preventive rather than acute nature. While increasing evidence supports the hypothesis that innate immunity is intimately involved in AD progression, targets will have to be chosen wisely, and only those that can be blocked at pre- or very early disease states may ultimately provide therapeutic benefit.


    . TREM2 variants in Alzheimer's disease. N Engl J Med. 2013 Jan 10;368(2):117-27. Epub 2012 Nov 14 PubMed.

    . Genetic evidence implicates the immune system and cholesterol metabolism in the aetiology of Alzheimer's disease. PLoS One. 2010;5(11):e13950. PubMed.

    . Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med. 2013 Jan 10;368(2):107-16. Epub 2012 Nov 14 PubMed.

    . Immune activation in brain aging and neurodegeneration: too much or too little?. Neuron. 2009 Oct 15;64(1):110-22. PubMed.

    . Critical role of astroglial apolipoprotein E and liver X receptor-α expression for microglial Aβ phagocytosis. J Neurosci. 2011 May 11;31(19):7049-59. PubMed.

    View all comments by Michael Heneka

Make a Comment

To make a comment you must login or register.


News Citations

  1. Enter the New Alzheimer’s Gene: TREM2 Variant Triples Risk
  2. Plasma Markers for Alzheimer’s—Slowly But Surely?
  3. Copper Mountain: Origins and Actions of Glial Cells in AD
  4. SfN: Glial-Neuronal Signaling and AD Pathology
  5. Orlando Conference: The Chicken and the Egg
  6. Macrophages Storm Blood-brain Barrier, Clear Plaques—or Do They?

Paper Citations

  1. . Reduced Th1 and enhanced Th2 immunity after immunization with Alzheimer's beta-amyloid(1-42). J Neuroimmunol. 2002 Nov;132(1-2):49-59. PubMed.
  2. . Microglial response to amyloid plaques in APPsw transgenic mice. Am J Pathol. 1998 Jan;152(1):307-17. PubMed.
  3. . CX3CR1 deficiency alters microglial activation and reduces beta-amyloid deposition in two Alzheimer's disease mouse models. Am J Pathol. 2010 Nov;177(5):2549-62. PubMed.
  4. . Regulation of tau pathology by the microglial fractalkine receptor. Neuron. 2010 Oct 6;68(1):19-31. PubMed.
  5. . Massive gliosis induced by interleukin-6 suppresses Abeta deposition in vivo: evidence against inflammation as a driving force for amyloid deposition. FASEB J. 2010 Feb;24(2):548-59. PubMed.
  6. . Chronic IL-1β-Mediated Neuroinflammation Mitigates Amyloid Pathology in a Mouse Model of Alzheimer's Disease without Inducing Overt Neurodegeneration. J Neuroimmune Pharmacol. 2012 Mar;7(1):156-64. PubMed.
  7. . Ustekinumab induction and maintenance therapy in refractory Crohn's disease. N Engl J Med. 2012 Oct 18;367(16):1519-28. PubMed.
  8. . Repeated subcutaneous injections of IL12/23 p40 neutralising antibody, ustekinumab, in patients with relapsing-remitting multiple sclerosis: a phase II, double-blind, placebo-controlled, randomised, dose-ranging study. Lancet Neurol. 2008 Sep;7(9):796-804. PubMed.

Other Citations

  1. APPPS1

External Citations

  1. Phase 3 trials
  2. Phase 2 trial

Further Reading


  1. . Microglia and neuroinflammation: a pathological perspective. J Neuroinflammation. 2004 Jul 30;1(1):14. PubMed.
  2. . Repeated subcutaneous injections of IL12/23 p40 neutralising antibody, ustekinumab, in patients with relapsing-remitting multiple sclerosis: a phase II, double-blind, placebo-controlled, randomised, dose-ranging study. Lancet Neurol. 2008 Sep;7(9):796-804. PubMed.
  3. . Cytokine polymorphisms and Alzheimer disease: possible associations. Neurol Sci. 2010 Jun;31(3):321-5. PubMed.
  4. . Non-steroidal anti-inflammatory drugs and Alzheimer's disease: the epidemiological evidence. CNS Neurol Disord Drug Targets. 2010 Apr;9(2):132-9. PubMed.

Primary Papers

  1. . Inhibition of IL-12/IL-23 signaling reduces Alzheimer's disease-like pathology and cognitive decline. Nat Med. 2012 Dec;18(12):1812-9. PubMed.