Tamibarotene: a Candidate Retinoid Drug for Alzheimer's Disease

Home , Bexarotene, Tamibarotene

1206 Biol. Pharm. Bull. 35(8) 1206–1212 (2012) Vol. 35, No. 8 Review Tamibarotene: A Candidate Retinoid Drug for Alzheimer’s Disease Hiroshi Fukasawa,*,a Madoka Nakagomi,a Naoko Yamagata,a Hiroshi Katsuki,b Kohichi Kawahara,c Kazuyoshi Kitaoka,d Takami Miki,e and Koichi Shudoa a Research Foundation ITSUU Laboratory; 2–28–10 Tamagawa, Setagaya-ku, Tokyo 158–0094, Japan: b Department of Chemico-Pharmacological Sciences, Kumamoto University; c Department of Molecular Cell Function, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto University; 5–1 Ohe-honmachi, Kumamoto 862–0973, Japan: d Department of Physiology, Institute of Health Biosciences, The University of Tokushima Graduate School; 3–18–15 Kuramoto-cho, Tokushima 770–8503, Japan: and e Department of Geriatrics and Neurology, Graduate School of Medicine, Osaka City University; 1–4–3 Asahimachi, Abeno-ku, Osaka 545–8585, Japan. Received April 6, 2012; accepted May 22, 2012

Tamibarotene (Am80), a synthetic retinoid approved in Japan for treatment of acute promyelocytic leukemia (APL), is a retinoic acid receptor (RAR) agonist with high specificity for RARα and RARβ over RARγ. Temporarily and spatially specific expression of RARs suggests their pivotal roles in the adult brain. Am80 is considered to be a promising candidate drug for treatment of Alzheimer’s disease (AD) because of its tran- scriptional controls of multiple target genes involved in etiology and pathology of AD. In APP23 AD model mice, administration of Am80 decreased the deposition of insoluble amyloid-β(42). In senescence-accelerated mice (SAMP8), Am80 ameliorated the decrease of cortical acetylcholine, as well as reducing anxiety in behavioral tests and improving the sleep deficit. Am80 also effected a significant improvement of memory in the rat scopolamine-induced memory deficit model. Like other retinoids, Am80 also has an immunomodula- tory effect and reduces secretion of proinflammatory cytokines and chemokines by astrocytes and microglia surrounding amyloid-β plaques. In a rat experimental autoimmune encephalomyelitis model, Am80 reduced inflammatory cytokines and showed significant efficacy. Retinoids also promote differentiation of neural stem cells, and Am80 improved the recovery of spinal cord-injured rats. Am80 may also improve vascular factors involved in onset and/or progression of AD. Am80 has been in clinical use for treatment of APL in Japan since 2005, and has been reported to have fewer side effects than other retinoids. We have recently started a clinical study to evaluate the efficacy and safety of Am80 for the treatment of Alzheimer’s disease. Key words retinoid; tamibarotene; Alzheimer’s disease; amyloid beta; acetylcholine; memory

INTRODUCTION approach. Another possibility would be to use a single drug that acts on plural targets related to Aβ, inflammation, neuro- Alzheimer’s disease (AD) mainly affects elderly individu- transmission, etc., involving different pathways or physiologi- als, and is becoming an increasing socioeconomic problem as cal categories.3) Retinoids may be good candidates for the lat- average life expectancy increases, particularly in developed ter approach, because they directly regulate a number of key countries. Drugs currently used to treat AD are neurotrans- genes and proteins in diverse pathways, and their principal mission modifiers, which are expected only to ameliorate biological targets include neurotransmission, Aβ, inflamma- the symptoms. Anti-amyloid therapies designed to reduce tion, neurogenesis, the blood–brain barrier, and others.4,5) amyloid-β (Aβ) production or to promote Aβ clearance, including active or passive immunotherapies, have been pro- 1. RETINOIDS AND THEIR BIOLOGICAL EFFECTS posed, and tau protein and mitochondrial dysfunction are also potential targets for therapy,1,2) though the latter approaches Retinoids are analogues of all-trans-retinoic acid (ATRA), are at a very early stage. To assist the clinical evaluation of an active metabolite of vitamin A (retinol), and are specific candidate drugs, it remains important to identify and validate modulators of cell proliferation, differentiation, morphogen- appropriate biomarkers. esis and immunology in vertebrates, as has been extensively The etiology of AD is thought to be complex and the cause reviewed.6–10) The term retinoid is often used loosely, e.g., to of early onset also remains unclear, though genetic factors def- include (1) structurally vitamin A-related compounds, includ- initely participate in the case of early-onset familial AD. Envi- ing vitamin A and its biological precursor carotenoids,11) (2) ronmental factors and infection may also play a role. A variety all-trans-retinoic acid (a metabolite of vitamin A that activates of factors may also contribute to progression of the disease. retinoic acid receptor (RAR)α, β, and γ) and its synthetic Therefore, multi-target therapies may prove to be more ef- analogs that bind to RARs with high affinity in an agonistic fective than single-target therapies. Multi-drug therapy, with (similar biological activities to ATRA) or antagonistic manner; each drug having a different target, would be one possible this definition is close to the original definition of Sporn,10) (3) compounds that activate retinoid X receptors (RXRα, β, and The authors declare no conflict of interest. γ), which are nuclear receptors structurally and functionally

Fig. 1. Chemical Structures of ATRA, Tamibarotene/Am80 and Am580 related to RARs and activated by 9-cis-RA (9cRA), but dis- viewpoints of low adverse effects and high target selectiv- tinct from them,12,13) (4) compounds that modify the activities ity. For example, ATRA is highly irritant to skin, in which it of ATRA by influencing metabolism, biosynthesis or other activates RARγ,27,28) and it also activates RXRs, which in turn pathways acting on co-factors or transporters, without bind- activate various other nuclear receptors. ing to RARs or RXRs. We prefer to use the term retinoid in a strict sense, that is, confined to compounds in category (2). 3. PREREQUISITES OF DRUGS FOR TREATMENT It should be noted that the activity of vitamin A is essentially OF AD due to RA generated by metabolism in vivo, except for the participation in vision via rhodopsin. Normally, endogenous Candidate drugs for use in the treatment of AD should in ATRA is maintained at low nanomolar levels in serum and principle have the following effects: (1) suppression of sub- tissues under strict metabolic control, although much larger acute accumulation of Aβ, a component of senile plaque of (low micromolar) amounts of retinol are found in serum and Alzheimer’s patients, which is produced from amyloid pre- tissues, while retinyl esters are stored in liver at high mi- cursor protein (APP) by β- and γ-secretases, in the brain of cromolar levels.14) 9cRA, which activates RARs as ATRA, gene-modified animal models, (2) suppression of brain inflam- can be hardly detected in serum and tissues.15) Therefore, its mation in animal models generated by intracerebral injection physiological role in vivo is still a subject of controversy. It is of lipopolysaccharide (LPS) or other agents, (3) stimulation of important to recognize that vitamin A itself is not active at all. functional, emotional and histological regeneration in animal The endogenous active retinoid, ATRA, equally binds to models of cholinergic or glutaminergic nerve destruction, etc., and activates RARα, β, and γ. It is important to note that as reflected by elevation of acetylcholine (Ach) level, choline ATRA also could bind to and activate RXRα, β, and γ, as does acetyltransferase (ChAT) activity, etc., (4) improvement of 9cRA16) (Shudo, unpublished). Many synthetic analogs with a cognition and learning in animal models of impairment, (5) variety of structures have been prepared, and often show dif- stimulation of nerve regeneration and neurite growth, and (6) ferent receptor selectivity and pharmacological effects from tolerability during prolonged use. ATRA, including different adverse effects. Some act as an- During screening of drugs for the treatment of AD, require- tagonists which inhibit the activity of ATRA.17,18) ments (1), (2) and (3) are evaluated first. If pharmacologically It is now clear that retinoids play a key role in the differ- acceptable compounds that satisfy these requirements are entiation and proliferation of various stem cells and progeni- discovered, the next requirement (4) is a behavioral pharma- tor cells, including nerve cells.19) Retinoids initially attracted cological study. However, the results of such a study depend attention in the field of skin science, because they markedly upon the experimental conditions, choice of animals, method suppress skin keratinization (an effect that stimulates skin of evaluation and so on, and are not always reliable or repro- rejuvenation). In 1978, Strickland and Madhvi discovered that ducible. More importantly, effects observed in experimental retinoic acid stimulates differentiation of embryonic stem cells animals cannot necessarily be extrapolated to humans. Never- into various types of cells.20) At approximately the same time, theless such studies have an important role. it was reported that retinoids stimulate final differentiation of Neuro-regeneration with small molecules is not yet widely the acute promyelocytic leukemia (APL) cell line HL60, in accepted as a potential clinical strategy, and alternatives such which differentiation is blocked midway.21) as cell transplantation and gene therapy are often considered more promising. However, neuro-regeneration with retinoids 2. ATRA AND TAMIBAROTENE FOR APL THERA- appears to be feasible.19) The central nervous system con- PY tains stem cells whose differentiation undoubtedly involves retinoids. In addition, safety is an absolute requirement, and in Nearly 50000 papers on retinoids have been indexed in this respect, the use of small-molecular drugs may have some PubMed, most of them concerning ATRA. Clinically, ATRA advantages. is extremely effective in the treatment of APL; most patients achieve complete remission.22) Several clinically used synthet- 4. THERAPEUTIC POTENTIAL OF TAMIBAROTENE ic retinoids have advantages over ATRA in terms of receptor FOR AD selectivity, potency, stability or bioavailability.17,18) Among them, tamibarotene (Am80) is effective in APL patients who So far, no compound satisfying all of the aforementioned experience relapse after ATRA treatment and acquire RA requirements has been found. Indeed, most drug discovery resistance.23–25) Am80 is a receptor subtype-selective retinoid efforts have been focused on a single target. However, Am80 (RARα= : RARβ≫RARγ) and is inactive towards RXRα, β, is intrinsically a multi-target drug and it meets at least some and γ26); this selectivity is practically advantageous from the of the above criteria. As it is already in clinical use, there is 1208 Vol. 35, No. 8 a considerable body of safety data, though for the new indica- are decreased significantly in hippocampus of these mice.40) tion of AD it will be necessary to carry out additional animal Recently, Cramer et al. reported that an RXR-selective studies. First, however, we will review the pharmacological agonist (so-called “rexinoid”) bexarotene (LG1069, Targre- data obtained during the last decade, in order to illustrate the tin), which is approved for treatement of cutaneous T cell rationale for clinical treatment of AD with Am80. lymphoma, enhance apolipoprotein E (ApoE)-dependent Aβ 4.1. Amyloid β Etchamendy et al. reported alleviation clearance from the brain and improve behavioral defects in of age-related deficit in CA1 long-term potentiation efficacy, APP/PS1 mice.41) Probably, bexarotene also binds to RXR and improvement of age-related relational memory deficit by in RAR-RXR heterodimer and enhances RAR signaling as ATRA in vivo.29) On the other hand, disruption of the retinoid HX630 does.18,42,43) signaling pathway caused deposition of Aβ in the adult rat 4.2. Suppression of Inflammation Neuroinflammation brain.30) Accumulation of Aβ in the brain is considered a hall- is a prominent feature in Alzheimer’s pathology and a po- mark of AD, though it may not be a gold standard marker.31) tential target for therapy. Anti-inflammatory therapies, par- Even though, reducing Aβ production or aggregation is gener- ticularly nonsteroidal anti-inflammatory drugs, have received ally considered to be an important requirement for therapy. considerable experimental and clinical attention. In APP23 transgenic mice with Swedish mutation, pro- LPS-interferon (IFN)γ-induced inflammation of central ner- longed administration of Am80 (0.5 mg/kg/d from 20 to 34 vous system cells was powerfully suppressed by Am80. The weeks of age) tended to reduce the insoluble Aβ(40) level in loss of neuronal viability in organotypic midbrain slice cul- the brain and significantly decreased the level of insoluble tures was rescued by submicromolar concentrations of Am80. Aβ(42), though the soluble Aβ level did not differ from that in The neuroprotective effect of Am80 on midbrain neurons control mice.32) This decrease of insoluble Aβ(42) may be an was also observed in an LPS-induced inflammation model in outcome of retinoid-induced potentiation of either α-secretase vivo.44) In that study, gene expression and protein synthesis of transcription or phagocytosis by alternatively activated (= brain-derived neurotrophic factor (BDNF) were markedly en- phagocytic) microglia (vide infra).33,34) Although Am80 had hanced, suggesting that the neuroprotective activity of Am80 no significant effect on the learning deficit of APP23 mice was mediated at least in part by BDNF.44,45) In an intracerebral in the Morris water maze (MWM) test, coadministration of hemorrhage model in mice, Am80 significantly attenuated Am80 with a specific RXR agonist HX630, which strongly neuronal damage in the striatum, and this effect was associ- enhances the biological effects of RARs in the presence of ated with suppression of activation of microglia/macrophages RAR agonist, resulted in a highly significant improvement in in the perihematoma region.46) A similar suppression of in- the MWM test concomitantly with a decrease of Aβ produc- flammatory cell death caused by oxidative stress, nitroxide tion (Kawahara, unpublished). or other stresses seems likely to occur in other brain areas. Decreased intracellular and extracellular Aβ production Jarvis et al.35) reported that Am580 significantly suppressed 35) was reported by Jarvis et al. in Tg2576 mice given synthetic cell death of cultured cortical neurons exposed to 10 µM Aβ. retinoid Am580 (1 mg/kg intraperitoneally (i.p.) every 3 d, or RA also showed evident suppression of inflammation around 3.6 mg/kg fed daily from 3 to 7 months of age), an isomer of sites of aggregation.36) Am80, as a selective RARα agonist (this compound also acti- Experimental autoimmune encephalomyelitis (EAE) is a vates RARβ, but not RARγ or RXRs, like Am80). Am580 had model that develops inflammation and demyelination in the similar effects to Am80 on Aβ(40) and Aβ(42). Further, i.p. central nervous system, resembling multiple sclerosis in hu- administration of Am580 to Tg2576 mice for 3 to 7 months mans. DA rats immunized with complete Freund's adjuvant dramatically increased sAPPα, a product of α-secretase pro- (CFA) supplemented with myelin basic protein developed cessing, in cortices. Am580 did not influence the production severe EAE, which reached its peak 12 to 14 d after immu- of β- and γ-secretases in cell cultures. Similarly, treatment nization. Am80 (1 and 3 mg/kg/d) administered orally for with ATRA (20 mg/kg, 3 times a week, starting at 5 months 12 d after immunization significantly diminished the clinical age, for 8 weeks) in APP-PS1 double-transgenic mice, which symptoms and the infiltration of inflammatory cells in a dose- exhibit faster Aβ accumulation and disease progression than dependent manner. The mRNA levels of interleukin-6 (IL-6), Tg2576 mice, caused a marked decrease of Aβ aggregation and IFN-γ and tumor necrosis factor-α (TNF-α) in spinal cord improvement of three-dimensional learning and memorization were found to parallel the clinical symptoms of the disease in the MWM test.36) Decreased activation of non-phagocytic in the Am80-treated rats. The expression of IL-6 mRNA was inflammatory microglia and reactive astrocytes was also ob- more rapidly and substantially reduced than that of the other served, together with a reduction of tau phosphorylation. two cytokine mRNAs. These findings suggest that the inhi- Retinoids, including Am80, increase gene transcription of bition of EAE is, at least in part, related to the inhibition of ADAM10 (a member of the disintegrin and metalloprotease, IL-6 production.47) ADAM, family) and increase the activity of α-secretase, A similar experiment in B6 mice examined in more detail which cleaves a different site of APP protein, precluding Aβ the immunological effects of Am80 treatment in the EAE generation and leading to release of the soluble neuroprotec- model: Am80 (3 mg/kg/d per os (p.o.)) significantly improved tive protein sAPPα.37–39) Increase or activation of ADAM10 the clinical score of the EAE mice. CD4+ effector T cells may be effective for reducing toxic Aβ without affecting were classically categorized into two subsets: T helper type γ-secretase activity; this is important because γ-secretase par- 1 (Th1) and type 2 (Th2) cells, and more recent research has ticipates in Notch signaling, which is essential for the main- delineated a further subset, Th17 cells, which are activated tenance of mammalian neural stem cells. Administration of CD4+ T cells characterized by the production of the cytokine Am80 to 13-month-old SAMP8 mice (vide infra) improved the IL-17 and involved in chronic inflammation. Am80 inhibited mRNA expression and protein production of ADAM10, which the in vitro Th17 differentiation of splenocytes, and instead August 2012 1209 up-regulated Foxp3 expression, a key transcription factor in vesicular acetylcholine transporter (VAChT) in the brainstem regulatory T cells. Am80 administration suppressed both the and TTR in the hippocampus. Furthermore, the decreased differentiation and the inflammatory function of Th17 cells in cortical acetylcholine (ACh) content in SAMP8 was improved vivo.48) These mice have been viewed as a model of multiple by Am80 administration. Decreased non-REM sleep and delta sclerosis, but the inflammatory lesions seen in this model (e.g., oscillation were also observed in SAMP8 mice.53) expression of IL-6, IFNγ and MCP-1) resemble those observed It was reported that expression of the RARβ gene seems in neurodegenerative diseases, including AD. to determine the contribution of delta oscillation to the sleep 4.3. Neural Growth In addition to the prevention of EEG patterns in mice.54) Four-week LE540 (RAR antagonist) neural cell death and amelioration of inflammation, retinoids administration induced a significant attenuation of wakeful- have a regenerative effect on neural cells.19) Goncalves et al.49) ness and delta oscillation magnitude in non-rapid eye move- reported that differentiation of adult forebrain neural pro- ment sleep.55) There was also a significant decrease in the genitor cells (NPC) into neurons was induced by RA through expression of dopamine D1 receptor (D1DR) in the striatum sequential RARβ and RARα signaling in vivo; RARγ seemed and tyrosine hydroxylase in the midbrain of the treated mice. not to be involved. The proliferated embryonic forebrain NPC Therefore, the improvement of sleep disorder by Am80 may expressed glial fibrillary acidic protein and were predomi- be induced by activation of the dopaminergic pathway as well nantly uni/bipolar, two characteristics of neuronal progenitor as the cholinergic pathway. cells. RARα agonist stimulated the differentiation of NPC into Prolonged administration of Am80 to SAMP8 mice sig- cholinergic neurons. Neuroblastoma cells NH-12 were differ- nificantly suppressed the symptoms of low anxiety, resulting entiated to neuronal cells by Am80 more efficiently than by in improvement of the moving latency of the animals in a RA.50) Am80 combined with BDNF promoted extensive neu- light-dark box and in open field behavior experiments to the rite outgrowth and increased expression of neurotrophic tyro- level seen in the control mice (SAMR1). The distance moved, sine kinase receptor B (TrkB) gene, which encodes the cog- the number of movements from one compartment to another nate receptor for BDNF, in cultured SH-SY5Y neuroblastoma and the number of standing-up motions were all improved.56) cells, as did RA. Thymidine incorporation was dramatically Am80 may have a beneficial effect on ambulatory tendency, a suppressed, but there was little effect on cell viability. Thus, symptom seen in patients with AD. the effect of Am80 on spinal cord injury (SCI) in rats as a Scopolamine is used in cognitive drug research testing, neural differentiation model was investigated. Treatment with because it impairs power of attention, continuity of attention, Am80 (orally or subcutaneously) significantly promoted re- quality of working memory, quality of episodic secondary covery from SCI-induced motor dysfunction. On day 28 after memory, and speed of memory in a dose- and time-dependent injury, the lesion cavity size was markedly reduced. Interest- manner, and reversal of scopolamine-induced cognitive im- ingly, expression of TrkB was over 3-fold higher after Am80 pairment is a viable model for identifying procognitive com- treatment than in SCI controls. Many TrkB-positive cells, as pounds. Donepezil as well as non-cholinergic compounds are well as BDNF-positive ones, were observed around the injured shown to be effective in this test.57) The learning disorder site. These findings suggest that Am80 has potential for treat- caused by scopolamine was improved by concomitant treat- ing neurodegenerative disorders.51) ment with Am80 in a one-trial, step-through, light-dark pas- 4.4. Improvement of Behaviors and Neurotransmission sive avoidance paradigm.58) Spontaneous models may have advantages over gene-modified 4.5. Vascular Events Vascular dysfunction is also in- models to explore the etiopathogenesis of aging-related disor- volved in AD pathology. Cerebral hypoperfusion and impaired ders. Senescence-accelerated mouse strains (SAM), a group Aβ clearance across the blood–brain barrier (BBB) may con- of related inbred strains consisting of senescence-prone in- tribute to the onset and progression of AD. The BBB is com- bred strains (SAMP) and senescence-resistant inbred strains prised of specialized capillary endothelial cells firmly linked (SAMR), have been developed by selective inbreeding of via intercellular tight junctions and surrounding pericytes and AKR/J strain mice donated by the Jackson Laboratory in glial cells. The BBB plays a critical role in Aβ clearance from 1968. The characteristic feature of SAMP is accelerated se- the brain. The accumulation of Aβ on cerebral blood vessels, nescence, whereas SAMR shows normal aging and can be known as cerebral amyloid angiopathy (CAA), is a feature of used as a reference. The prone 8 (SAMP8) mice show age- aging and AD.59) related behavioral deterioration, including deficits in learning Am58060) and Am80 (Nishikiori and Osanai, unpublished) and memory and emotional disorders such as reduced anxiety- enhanced expression of glial cell line-derived neurotrophic like behavior.52) factor (GDNF), but decreased expression of vascular endothe- SAMP8 mice exhibit age-related deterioration in sleep- lial growth factor (VEGF)/vascular permeability factor in glial wake architecture compared with SAMR1 mice. Treatment cells of the blood–retinal barrier (BRB) in diabetic patients. of 9- to 10-month-old SAMP8 mice with Am80 (2 mg/kg fed These actions normalize the excessive vascular permeability daily) improved their sleep defect as evaluated by electro- of BRB by modulating expression of tight junction proteins in encephalogram (EEG) and electromyogram analyses. One or capillary endothelium. Suppressive action of Am80 on Th17 four weeks of Am80 administration improved the decrease in lymphocytes can be also beneficial for maintenance of BBB rapid eye movement (REM) sleep characteristic of SAMP8 function. BBB endothelial cells express IL-17 receptor on their mice. Real-time reverse transcription-polymerase chain reac- surface and IL-17 disrupts BBB tight junctions.61) tion (RT-PCR) analysis demonstrated an impairment in the RA and Am80 exert potent antiangiogenic activities in the hippocampal retinoid cascade genes RARα and transthyretin chicken chorioallantoic membrane (CAM) assay and modulate (TTR) in SAMP8 in comparison to SAMR1 mice. Am80 vascular remodeling.62) Am80 has potent effects on vascular treatment induced an increase in mRNA expression levels of reconstruction after physical and drug injury via inhibition 1210 Vol. 35, No. 8

Fig. 2. Retinoid Dramatis Personae (Characters Involved in Retinoid Therapy for AD) Neuron, damaged neuron, astrocyte, microglia (MG), blood brain barrier (BBB), and other cellular members participate in the pathology of Alzheimer’s disease. Reti- nol bound to retinol-binding protein (RBP) and TTR is transported from blood into brain. Cellular ATRA is synthesized from retinol by enzymes, such as retinaldehyde dehydrogenase (RALDH). Intracellular levels of active RA is regulated by cellular RA binding protein (CRABP) and metabolizing enzymes, such as CYP26.7) Retinoids decrease amyloid-β (Aβ) deposition from APP by increase of α-secretase activity through ADAM10 induction and by induction of phogocytic MG. Inflammation caused by amyloid-β and other causes such as xenobiotics or chemicals, inflammatory or proinflammatory cytokines and chemokines (IL-1β, IL-6, IL-17, TNFα, MCP-1, VEGF). Their production is suppressed by retinoids through RARs expressed in astrocyte and inflammatory microglia. In addition, differentiation and/or function of Th17 cells, which are involved in autoimmune pathways and BBB disruption, are suppressed by retinoid. Cell death may be inhibited by the decrease of amyloid-β and suppression of inflammation. From the viewpoint of neuroregeneration, retinoids have critical roles in maintenance and differentiation of neural stem cells. Neurotransmission of ACh, DA and other molecules are effected by ChAT, dopamine receptor (D2R), BDNF, TrkB and t-PA which are transcriptionally regulated directly or indirectly by retinoid, influencing memory and learning. In vascular system, retinoids keep BBB integrity and suppress angiogenesis of cerebral microvessels. Suppressive effect of retinoid on KLF5, which is involved in vascular remodeling, in vascular smooth muscle cells may be beneficial. In the scheme, + and − mean up-regulation and down-regulation of the events or activities, by retinoids, respectively. This figure is a revised version of Fig. 2 from our previous review article.5) of transcription factor Kruppel-like factor 5 (KLF5).63,64) of APL,69) or in clinical studies of patients with Crohn’s dis- Microvessels derived from AD patients express or release a ease and multiple myeloma (unpublished results). In the post- myriad of factors implicated in vascular activation and angio- marketing surveillance of adverse reactions, common adverse genesis. Extensive cerebral angiogenesis in Tg2576 mouse and reactions were found to be reversible TG elevation, bone pain, AD patients is reported.65) Sunitinib, a multi-target tyrosine rash, etc. Overall, there were fewer adverse reactions to Am80 kinase inhibitor (TKI) that predominantly targets VEGF, was than to RA, which frequently induces skin inflammation. reported to improve behaviors and to reduce production of Am80, like other retinoids, is absolutely contraindicated in various inflammatory cytokines in Tg2576 mice,66) and Am80 pregnant women. The clearance time of Am80 from the body may act through similar mechanisms. is short and Am80 is not accumulated in tissues.

5. SAFETY ISSUES 6. CLINICAL STUDY

Am80 has already been approved for use in the treatment of On the basis of these pharmacological and safety data, a acute promyelocytic leukemia (APL) in Japan since 2005.67) A clinical trial (ClinicalTrials.gov. NCT 01120002; JAPICCTI- serious adverse reaction in the treatment of APL is differen- 101115(ja)) of Am80 for Alzheimer’s disease is in progress. tiation syndrome, formerly known as retinoic acid syndrome, The trial is an interventional, randomized, placebo-controlled but this adverse reaction to retinoids has been deﬁnitively study, in which patients receive two tamibarotene 2 mg tablets concluded to be associated with the underlying condition of or two placebo tablets once daily. Primary outcome measure APL.68) This reaction was not seen among 400 patients dur- is the change in Alzheimer’s disease assessment scale (ADAS- ing prolonged (over two years) use of Am80 after remission JCog). The current trial location is Osaka City University. August 2012 1211

CONCLUSION 20) Strickland S, Mahdavi V. The induction of differentiation in tera- tocarcinoma stem cells by retinoic acid. Cell, 15, 393–403 (1978). Am80, tamibarotene, is a potent multi-target drug that acts 21) Breitman TR, Selonick SE, Collins SJ. Induction of differentiation on multiple pathways considered to be involved in the etiology of the human promyelocytic leukemia cell line (HL-60) by retinoic acid. Proc. Natl. Acad. Sci. U.S.A., 77, 2936–2940 (1980). and pathophysiology of Alzheimer’s disease, as summarized 22) Huang ME, Ye YC, Chen SR, Chai JR, Lu JX, Zhoa L, Gu LJ, in Fig. 2. It is therefore considered to be a promising agent for Wang ZY. Use of all-trans retinoic acid in the treatment of acute treatment of Alzheimer’s disease, and a clinical trial is under promyelocytic leukemia. Blood, 72, 567–572 (1988). way. 23) Kagechika H, Kawachi E, Hashimoto Y, Shudo K. New type in- ducers of differentiation of human HL-60 promyelocytic leukemia REFERENCES cells. Terephthalic anilides. Chem. Pharm. Bull., 32, 4209–4212 (1984). 1) Citron M. Alzheimer’s disease: strategies for disease modification. 24) Kagechika H, Kawachi E, Hashimoto Y, Himi T, Shudo K. Retino- Nat. Rev. Drug Discov., 9, 387–398 (2010). benzoic acids. 1. Structure–activity relationships of aromatic am- 2) Selkoe DJ. Resolving controversies on the path to Alzheimer’s ides with retinoidal activity. J. Med. Chem., 31, 2182–2192 (1988). therapeutics. Nat. Med., 17, 1060–1065 (2011). 25) Miwako I, Kagechika H. Tamibarotene. Drugs Today, 43, 563–568 3) Mangialasche F, Solomon A, Winblad B, Mecocci P, Kivipelto M. (2007). Alzheimer’s disease: clinical trials and drug development. Lancet 26) Umemiya H, Fukasawa H, Ebisawa M, Eyrolles L, Kawachi E, Neurol., 9, 702–716 (2010). Eisenmann G, Gronemeyer H, Hashimoto Y, Shudo K, Kagechika 4) Lee HP, Casadesus G, Zhu X, Lee HG, Perry G, Smith MA, H. Regulation of retinoidal actions by diazepinylbenzoic acids. Gustaw-Rothenberg K, Lerner A. All-trans retinoic acid as a novel Retinoid synergists which activate the RXR-RAR heterodimers. J. therapeutic strategy for Alzheimer’s disease. Expert Rev. Neuro- Med. Chem., 40, 4222–4234 (1997). ther., 9, 1615–1621 (2009). 27) Standeven AM, Teng M, Chandraratna RA. Lack of involvement 5) Shudo K, Fukasawa H, Nakagomi M, Yamagata N. Towards of retinoic acid receptor alpha in retinoid-induced skin irritation in retinoid therapy for Alzheimer’s disease. Curr. Alzheimer Res., 6, hairless mice. Toxicol. Lett., 92, 231–240 (1997). 302–311 (2009). 28) Look J, Landwehr J, Bauer F, Hoffmann AS, Bluethmann H, 6) Balmer JE, Blomhoff R. Gene expression regulation by retinoic LeMotte P. Marked resistance of RAR gamma-deficient mice to the acid. J. Lipid Res., 43, 1773–1808 (2002). toxic effects of retinoic acid. Am. J. Physiol., 269, E91–E98 (1995). 7) Blomhoff R, Blomhoff HK. Overview of retinoid metabolism and 29) Etchamendy N, Enderlin V, Marighetto A, Vouimba RM, Pallet V, function. J. Neurobiol., 66, 606–630 (2006). Jaffard R, Higueret P. Alleviation of a selective age-related relation- 8) Germain P, Chambon P, Eichele G, Evans RM, Lazar MA, Leid M, al memory deficit in mice by pharmacologically induced normaliza- De Lera AR, Lotan R, Mangelsdorf DJ, Gronemeyer H. Internation- tion of brain retinoid signaling. J. Neurosci., 21, 6423–6429 (2001). al Union of Pharmacology. LX. Retinoic acid receptors. Pharmacol. 30) Corcoran JP, So PL, Maden M. Disruption of the retinoid signalling Rev., 58, 712–725 (2006). pathway causes a deposition of amyloid beta in the adult rat brain. 9) Sporn MB, Roberts AB. Goodman DS, The Retinoids. Academic Eur. J. Neurosci., 20, 896–902 (2004). Press, Orlando (1984). 31) Zhang YW, Thompson R, Zhang H, Xu H. APP processing in 10) Sporn MB, Roberts AB, Goodman DS. Gudas LJ. The Retinoids: Alzheimer’s disease. Mol. Brain, 4, 3 (2011). biology, chemistry, and medicine (2nd ed.). Raven Press, New York, 32) Kawahara K, Nishi K, Suenobu M, Ohtsuka H, Maeda A, Na- (1994). gatomo K, Kuniyasu A, Staufenbiel M, Nakagomi M, Shudo K, 11) IUPAC-ICB Joint Commission on Biochemical Nomenclature Nakayama H. Oral administration of synthetic retinoid Am80 (Ta- (JCBN). Nomenclature of retinoids. Recommendations 1981. Eur. J. mibarotene) decreases brain beta-amyloid peptides in APP23 mice. Biochem., 129, 1–5 (1982). Biol. Pharm. Bull., 32, 1307–1309 (2009). 12) Germain P, Chambon P, Eichele G, Evans RM, Lazar MA, Leid M, 33) Shimizu E, Kawahara K, Kajizono M, Sawada M, Nakayama De Lera AR, Lotan R, Mangelsdorf DJ, Gronemeyer H. Internation- H. IL-4-induced selective clearance of oligomeric beta-amyloid al Union of Pharmacology. LXIII. Retinoid X receptors. Pharmacol. peptide (1–42) by rat primary type 2 microglia. J. Immunol., 181, Rev., 58, 760–772 (2006). 6503–6513 (2008). 13) Mangelsdorf DJ, Evans RM. The RXR heterodimers and orphan 34) Cameron B, Landreth GE. Inflammation, microglia, and receptors. Cell, 83, 841–850 (1995). Alzheimer’s disease. Neurobiol. Dis., 37, 503–509 (2010). 14) Kane MA, Napoli JL. Quantification of endogenous retinoids. Meth- 35) Jarvis CI, Goncalves MB, Clarke E, Dogruel M, Kalindjian SB, ods Mol. Biol., 652, 1–54 (2010). Thomas SA, Maden M, Corcoran JP. Retinoic acid receptor-α sig- 15) Kane MA, Folias AE, Pingitore A, Perri M, Obrochta KM, Krois nalling antagonizes both intracellular and extracellular amyloid-β CR, Cione E, Ryu JY, Napoli JL. Identification of 9-cis-retinoic production and prevents neuronal cell death caused by amyloid-β. acid as a pancreas-specific autacoid that attenuates glucose-stimu- Eur. J. Neurosci., 32, 1246–1255 (2010). lated insulin secretion. Proc. Natl. Acad. Sci. U.S.A., 107, 21884– 36) Ding Y, Qiao A, Wang Z, Goodwin JS, Lee ES, Block ML, 21889 (2010). Allsbrook M, McDonald MP, Fan GH. Retinoic acid attenu- 16) Mangelsdorf DJ, Ong ES, Dyck JA, Evans RM. Nuclear receptor ates β-amyloid deposition and rescues memory deficits in an that identifies a novel retinoic acid response pathway. Nature, 345, Alzheimer’s disease transgenic mouse model. J. Neurosci., 28, 224–229 (1990). 11622–11634 (2008). 17) de Lera AR, Bourguet W, Altucci L, Gronemeyer H. Design of se- 37) Endres K, Fahrenholz F. Regulation of alpha-secretase ADAM10 lective nuclear receptor modulators: RAR and RXR as a case study. expression and activity. Exp. Brain Res., 217, 343–352 (2012). Nat. Rev. Drug Discov., 6, 811–820 (2007). 38) Koryakina A, Aeberhard J, Kiefer S, Hamburger M, Küenzi P. 18) Kagechika H, Shudo K. Synthetic retinoids: recent developments Regulation of secretases by all-trans-retinoic acid. FEBS J., 276, concerning structure and clinical utility. J. Med. Chem., 48, 5875– 2645–2655 (2009). 5883 (2005). 39) Lichtenthaler SF. Alpha-secretase in Alzheimer’s disease: molecular 19) Mey J. New therapeutic target for CNS injury? The role of retinoic identity, regulation and therapeutic potential. J. Neurochem., 116, acid signaling after nerve lesions. J. Neurobiol., 66, 757–779 (2006). 10–21 (2011). 1212 Vol. 35, No. 8

40) Kitaoka K, Shimizu N, Chikahisa S, Nakagomi M, Shudo K, Sei sleep. Science, 310, 111–113 (2005). H. “Chronic administration of retinoic acid receptor agonist Am80 55) Kitaoka K, Shimizu M, Shimizu N, Chikahisa S, Nakagomi M, induces the ADAM10 expression in the hippocampus of SAMP8 Shudo K, Yoshizaki K, Séi H. Retinoic acid receptor antagonist mice, The 89th Annual Meeting of the Physiological Society of Ja- LE540 attenuates wakefulness via the dopamine D1 receptor in pan, March 29–31, 2012, Matsumoto, Japan.”: ‹http://seiri89.umin.jp/ mice. Brain Res., 1423, 10–16 (2011). images/program/abs_p3.pdf›, cited 15 Feburary, 2012. 56) Nakagomi M, Shudo K. Presentation at 10th International Confer- 41) Cramer PE, Cirrito JR, Wesson DW, Lee CY, Karlo JC, Zinn AE, ence AD/PD 2011, Barcelona). Neurodegener. Dis., 8, S1 (2011). Casali BT, Restivo JL, Goebel WD, James MJ, Brunden KR, Wil- 57) Lenz RA, Baker JD, Locke C, Rueter LE, Mohler EG, Wesnes K, son DA, Landreth GE. ApoE-directed therapeutics rapidly clear Abi-Saab W, Saltarelli MD. The scopolamine model as a pharmaco- β-amyloid and reverse deficits in AD mouse models. Science, 335, dynamic marker in early drug development. Psychopharmacology 1503–1506 (2012). (Berl.), 220, 97–107 (2012). 42) Nishimaki-Mogami T, Tamehiro N, Sato Y, Okuhira K, Sai K, 58) Shudo K, Kagechika H, Yamazaki N, Igarashi M, Tateda C. A Kagechika H, Shudo K, Abe-Dohmae S, Yokoyama S, Ohno Y, synthetic retinoid Am80 (tamibarotene) rescues the memory defi- Inoue K, Sawada J. The RXR agonists PA024 and HX630 have dif- cit caused by scopolamine in a passive avoidance paradigm. Biol. ferent abilities to activate LXR/RXR and to induce ABCA1 expres- Pharm. Bull., 27, 1887–1889 (2004). sion in macrophage cell lines. Biochem. Pharmacol., 76, 1006–1013 59) Bell RD, Zlokovic BV. Neurovascular mechanisms and blood–brain (2008). barrier disorder in Alzheimer’s disease. Acta Neuropathol., 118, 43) Pérez E, Bourguet W, Gronemeyer H, de Lera AR. Modulation of 103–113 (2009). RXR function through ligand design. Biochim. Biophys. Acta, 1821, 60) Nishikiori N, Osanai M, Chiba H, Kojima T, Mitamura Y, Ohguro 57–69 (2012). H, Sawada N. Glial cell-derived cytokines attenuate the breakdown 44) Katsuki H, Kurimoto E, Takemori S, Kurauchi Y, Hisatsune A, of vascular integrity in diabetic retinopathy. Diabetes, 56, 1333– Isohama Y, Izumi Y, Kume T, Shudo K, Akaike A. Retinoic acid 1340 (2007). receptor stimulation protects midbrain dopaminergic neurons from 61) Kebir H, Kreymborg K, Ifergan I, Dodelet-Devillers A, Cayrol R, inflammatory degeneration via BDNF-mediated signaling. J. Neuro- Bernard M, Giuliani F, Arbour N, Becher B, Prat A. Human TH17 chem., 110, 707–718 (2009). lymphocytes promote blood–brain barrier disruption and central 45) Kurauchi Y, Hisatsune A, Isohama Y, Sawa T, Akaike T, Shudo K, nervous system inflammation. Nat. Med., 13, 1173–1175 (2007). Katsuki H. Midbrain dopaminergic neurons utilize nitric oxide/cy- 62) Suzuki Y, Komi Y, Ashino H, Yamashita J, Inoue J, Yoshiki A, clic GMP signaling to recruit ERK that links retinoic acid receptor Eichmann A, Amanuma H, Kojima S. Retinoic acid controls blood stimulation to up-regulation of BDNF. J. Neurochem., 116, 323–333 vessel formation by modulating endothelial and mural cell interac- (2011). tion via suppression of Tie2 signaling in vascular progenitor cells. 46) Matsushita H, Hijioka M, Hisatsune A, Isohama Y, Shudo K, Blood, 104, 166–169 (2004). Katsuki H. A retinoic acid receptor agonist Am80 rescues neurons, 63) Nagai R, Suzuki T, Aizawa K, Shindo T, Manabe I. Significance attenuates inflammatory reactions, and improves behavioral recov- of the transcription factor KLF5 in cardiovascular remodeling. J. ery after intracerebral hemorrhage in mice. J. Cereb. Blood Flow Thromb. Haemost., 3, 1569–1576 (2005). Metab., 31, 222–234 (2011). 64) Fujiu K, Manabe I, Ishihara A, Oishi Y, Iwata H, Nishimura G, 47) Wang T, Niwa S, Bouda K, Matsuura S, Homma T, Shudo K, Nagai Shindo T, Maemura K, Kagechika H, Shudo K, Nagai R. Synthetic H. The effect of Am-80, one of retinoids derivatives on experi- retinoid Am80 suppresses smooth muscle phenotypic modulation mental allergic encephalomyelitis in rats. Life Sci., 67, 1869–1879 and in-stent neointima formation by inhibiting KLF5. Circ. Res., 97, (2000). 1132–1141 (2005). 48) Klemann C, Raveney BJ, Klemann AK, Ozawa T, von Hörsten S, 65) Biron KE, Dickstein DL, Gopaul R, Jefferies WA. Amyloid triggers Shudo K, Oki S, Yamamura T. Synthetic retinoid AM80 inhibits extensive cerebral angiogenesis causing blood brain barrier perme- Th17 cells and ameliorates experimental autoimmune encephalo- ability and hypervascularity in Alzheimer’s disease. PLoS ONE, 6, myelitis. Am. J. Pathol., 174, 2234–2245 (2009). e23789 (2011). 49) Goncalves MB, Agudo M, Connor S, McMahon S, Minger SL, 66) Grammas P, Yin L, Sanchez A, Evola M, Young A. Targeting Maden M, Corcoran JP. Sequential RARbeta and alpha signalling in vascular activation: A novel therapeutic strategy for Alzheimer’s vivo can induce adult forebrain neural progenitor cells to differenti- disease. J. Nutr. Health Aging, 15, S11–S12 (2011). ate into neurons through Shh and FGF signalling pathways. Dev. 67) Tobita T, Takeshita A, Kitamura K, Ohnishi K, Yanagi M, Hiraoka Biol., 326, 305–313 (2009). A, Karasuno T, Takeuchi M, Miyawaki S, Ueda R, Naoe T, Ohno 50) Shiohira H, Kitaoka A, Shirasawa H, Enjoji M, Nakashima M. R. Treatment with a new synthetic retinoid, Am80, of acute pro- Am80 induces neuronal differentiation in a human neuroblastoma myelocytic leukemia relapsed from complete remission induced by NH-12 cell line. Int. J. Mol. Med., 26, 393–399 (2010). all-trans retinoic acid. Blood, 90, 967–973 (1997). 51) Takenaga M, Ohta Y, Tokura Y, Hamaguchi A, Shudo K, Okano H, 68) Montesinos P, Sanz MA. The differentiation syndrome in patients Igarashi R. The effect of Am-80, a synthetic retinoid, on spinal cord with acute promyelocytic leukemia: experience of the pethema injury-induced motor dysfunction in rats. Biol. Pharm. Bull., 32, group and review of the literature. Mediterr. J. Hematol. Infect. 225–231 (2009). Dis., 3, e2011059 (2011). 52) Takeda T. Senescence-accelerated mouse (SAM) with special refer- 69) Asou N, Kishimoto Y, Kiyoi H, Okada M, Kawai Y, Tsuzuki M, ences to neurodegeneration models, SAMP8 and SAMP10 mice. Horikawa K, Matsuda M, Shinagawa K, Kobayashi T, Ohtake S, Neurochem. Res., 34, 639–659 (2009). Nishimura M, Takahashi M, Yagasaki F, Takeshita A, Kimura Y, 53) Kitaoka K, Sano A, Chikahisa S, Yoshizaki K, Séi H. Disturbance Iwanaga M, Naoe T, Ohno R, Japan Adult Leukemia Study Group. of rapid eye movement sleep in senescence-accelerated mouse A randomized study with or without intensified maintenance che- prone/8 mice is improved by retinoic acid receptor agonist Am80 motherapy in patients with acute promyelocytic leukemia who have (Tamibarotene). Neuroscience, 167, 573–582 (2010). become negative for PML-RARalpha transcript after consolidation 54) Maret S, Franken P, Dauvilliers Y, Ghyselinck NB, Chambon P, therapy: the Japan Adult Leukemia Study Group (JALSG) APL97 Tafti M. Retinoic acid signaling affects cortical synchrony during study. Blood, 110, 59–66 (2007).