TABLE 1.
Summary of PPARγ agonist effects on pre-clinical models of Alzheimer’s disease.
| Model/references | Dosing | Human Equivalent Dosea | Comments | Results |
| Treatment paradigm | ||||
| Tg2576 (Pedersen et al., 2006) | Rosiglitazone, 4 mg/kg, po (chow), vs vehicle control, 4 months. | 19.51 mg/d | Male, 9 months old at initiation of treatment; amyloid deposits, hippocampal dendrite spine loss, and defective spatial learning evident when treatment initiated. | Rosiglitazone was statistically associated with improved memory, reduced learning deficits (radial arm maze); reduced insoluble Aβ42 levels; reduced corticosterone levels. |
| Tg2576, Nenov 2014, 2015 (Nenov et al., 2014, 2015) | Rosiglitazone, 30 mg/kg, po (chow), vs vehicle control, for 30 days. | 146.35 mg/d | Equal numbers of males and females, 8 months old at initiation of treatment. | Learning, memory improvements correlated with improved spontaneous synaptic activity and short-term plasticity; engagement of the ERK pathway and expression of synaptic proteins, restoration of mature: immature DG granule cell ratio, normalized Nav-mediated currents. |
| APP (V717I), (Heneka et al., 2005) | Pioglitazone, 40 mg/kg/d; po (chow), vs vehicle control, 7 days | 195.12 mg/d | Equal numbers of males and females; 10 months old at initiation of treatment; amyloid pathology present. | Decreased BACE1, amyloid plaque deposits, soluble Aβ42 levels, reactive microglia. |
| APP (Swe/PS1)Δ9, (Mandrekar-Colucci et al., 2012; Skerrett et al., 2015) | Pioglitazone, 80 mg/kd/d, po (gavage) vs vehicle control, 9 days | 390.24 mg/d | Equal numbers of males and females; similar results for mice that were 6 or 12 months-old at initiation of treatment and that exhibited differing pathological loads. | Improved memory retention. Decreased amyloid plaque deposits, reduced soluble and insoluble Aβ40 and Aβ42 in 6-month-old mice; reduced insoluble Aβ42 and Aβ40 and soluble Aβ40 in 12-month-old mice; reduced reactive microglia and astrocytes and enhanced microglial phagocytosis; increased IL-1β, TNFα, Tm1, Fizz1, Arg1 expression; |
| APP/PSI (Chen et al., 2015) | Pioglitazone, 10 mg/kg/d, ip, vs. vehicle control, 7 and 10 days. | 48.78 mg/d | Equal numbers of males and females; 12 months old at initiation of treatment; pathology present. | Improved LTP after 7 days treatment & water maze performance after 10 days. Reduced CDK5 expression and activity as tau phosphorylation surrogate. |
| APP (Swe/PS1)Δ9 (Toba et al., 2016) | Pioglitazone, 80 mg/kg/d, po (chow), vs vehicle control, 9 days. | 390.25 mg/d | Equal numbers of male and female mice, 5 – 6 months old at initiation of treatment, emergent stages of pathology. | Increased motor coordination, LTP; decreased CDK5 regulatory protein (p25 & p35) expression. |
| J20 (V717F under PDGF promoter) (Escribano et al., 2010) | Rosiglitazone, 5 mg/kg/d, po (gavage) vs vehicle control, treated for 1 month and 4 months. | 24.39 mg/d | Equal numbers of males and females; 10 months old at initiation of treatment. | Improved object recognition after one month and progressively improved spatial memory (Morris Water Maze); reduced amyloid plaque and insoluble Aβ42 and Aβ40 levels, and phosphorylated tau, and promoted anti-inflammatory, pro-phagocytic microglial phenotype. |
| J20 (Nicolakakis et al., 2008) | Pioglitazone, 20 mg/kg/d, po (chow), vs vehicle control, for 1.5 – 2 months. | 97.56 mg/d | Equal numbers of males and females; 14 months old at initiation of treatment; amyloidosis, neuronal loss well established. | No effect on water maze performance, amyloid deposits or levels of soluble or insoluble Aβ42 or Aβ40. Improved cerebral blood flow and glucose uptake; restored cerebrovascular function; trend toward improved cortical cholinergic stimulation; reduced astrogliosis; reversed cerebral oxidative stress. |
| 3xTg, Search (Searcy et al., 2012) | Pioglitazone, 18 mg/kg, po (chow), vs vehicle control, for 3.5 months | 87.81 mg/d | Female mice, 11 – 12 months of age when treatment initiated; amyloid deposits well-established, tau aggregates present. | Improved learning on the active avoidance task; enhanced LTP; reduced amyloid deposits and hyperphosphorylated tau in CA1. |
| Prevention paradigm | ||||
| J20 (Escribano et al., 2009) | Rosiglitazone, 5 mg/kg/d, po (gavage) vs vehicle control. | 24.39 mg/d | Prevention vs. rescue study. Equal numbers of 1.5-month-old males and females were treated for 2.5 months (prevention), and equal numbers of 9-month-old mice were treated for 1 month (rescue). | Improved object recognition in both cohorts. In older mice, reduced corticosterone levels and blocked glucocorticoid receptor down-regulation. |
| J20 (Badhwar et al., 2013) | Pioglitazone, 20 mg/kg/day, po (chow) vs. control | 97.56 mg/d | Treatment initiated with 3-month-old mice for 14 weeks; small cohort for 3 days. | 14-week treatment that was initiated in young mice was statistically associated with improved spatial learning, with trend toward improved memory. Three-day treatment rescued cerebral blood flow; effect persisted in the longer-term treated mice. |
| Tg2576 (Rodriguez-Rivera et al., 2011) | Rosiglitazone, 30 mg/kg, po(chow), vs vehicle control, for 4, 8 or 12 months | 146.16 mg/d | Equal numbers of males and females, 1 month old at initiation of treatment. | Reversed associative learning and memory deficits in 9-month old animals (fed for 8 months), but not 5 (fed for 4 months) or 13 (fed for 12 months) month-old mice. |
| SCAMP8 (Seok et al., 2019) SCAMP8 is a spontaneous ‘Alzheimer’s-like’ mouse model that exhibits amyloid and tau pathology, neuron and dendrite spine loss, and CNS oxidative stress (Armbrecht et al., 2014; Cheng et al., 2014) | Pioglitazone, 2 or 5 mg/kg/d, po (gavage) versus vehicle control, for 7 weeks. | 9.76 or 24.39 mg/d | Equal numbers of male and female mice, 9 months old at initiation of treatment. | Improved water maze performance, reduced amyloid deposits and soluble Aβ40; increased LRP1 expression. All responses were attenuated at 5 mg/kg/d vs 2 mg/kg/d. |
| Cerebrovascular model | ||||
| J20/TGFβ1 (Papadopoulos et al., 2013) | Pioglitazone, 20 mg/kg/d, po (chow), vs vehicle control, 6 months | 97.56 mg/d | Also a ‘treatment vs prevention study.’ Equal numbers of males and females. Two cohorts, 6 and 12 months of age at beginning of treatment. Adult mice treated for 6 months; aged mice treated for 3 months. | No effect on spatial learning or memory; improved reversal learning in adult but not aged mice. In both adult and aged cohorts, improved cerebral blood flow, cerebral glucose uptake; suppressed astrogliosis in cortex but not in hippocampus; suppressed microglial activation in hippocampus. No effect in either cohort on amyloid pathology, or on cerebrovascular reactivity. |
| TGFβ1 (Lacombe et al., 2004) | Pioglitazone, 18 mg/kg/d, po (chow), vs vehicle control, 2 months. | 87.81 mg/d | Equal numbers of male and female mice, 2 months old at initiation of treatment. | Decreased Aβ42 levels and glia activation, and increased hydrocephalus. |
| TGFβ1 (Galea et al., 2006) | Pioglitazone, 18 mg/kg/d, po (chow), vs vehicle control, 2 months. | 87.81 mg/d | Equal numbers of male and female mice, 2 months old at initiation of treatment. | Pioglitazone inhibited cerebral glucose uptake in control (non-transgenic littermate) mice and failed to reverse TGF1b-mediated inhibition in transgenic mice. |
| Diabetes models | ||||
| ICR mice (Jiang et al., 2012) The ICR strain is a general-purpose mouse line. Diabetes was induced by feeding a high fat diet (60% fat, 20% carbohydrate, 20% protein) for 1 month to 10-week-old mice to elicit peripheral insulin resistance and hyperglycemia, followed by injection with streptozotocin (100 mg/kg), to cause insulin deficiency and cerebral hypoinsulinemia. | Pioglitazone, 18 mg/kg/d, po (chow) or 9 mg/kg/d vs vehicle control, for 6 weeks. | 87.81 mg/d or 43.9 mg/d | Equal numbers of male and female mice, 16 – 18 weeks old at the initiation of treatment, with similar body weights and degree of hyperglycemia were randomly assigned to equally sized treatment groups. Non-diabetic controls were similarly divided into treatment groups. | HFD/strep-diabetes was associated with memory impairments; pioglitazone treatment improved learning and memory, and reduced soluble Aβ42 and Aβ40, BACE1, NF-κB and RAGE. |
| Sprague-Dawley rats (Gao et al., 2017) 12-week-old rats were fed 60% fat diet for 20 weeks, then injected with streptozotocin (27 mg/kg). | Pioglitazone, 10 mg/kg/d, po (chow) vs vehicle control, for 10 weeks. | 48.78 mg/d | Equal numbers of 20-week-old male and female mice were divided into treatment groups, as described for the ICR mice. | HFD/strep rats exhibited memory impairments versus the control and diabetes+pioglitazone groups, which did not differ from each other. Pioglitazone corrected impaired ERK1/2 mRNA and protein expression caused by hyperglycemia. |
| APOE Model | ||||
| APOE TR mice (To et al., 2011) 3-month-old male mice fed 60% fat diet for 32 weeks. | Pioglitazone, 20 mg/k/d, po (gavage) vs vehicle control, for 3 weeks. | 97.56 mg/d | 40-week-old male HF or LF mice were divided into control and pioglitazone treatment groups and dosed for 3 weeks. | HF diet elicited insulin resistance and impaired glucose tolerance; reduced all phospho-tau epitopes in all mice. Neither diet nor pioglitazone affected APP metabolism. In HF mice, pioglitazone was associated with reduced AT8 p-tau in APOE ε3 mice and increased AT8 p-tau in APOE ε4 mice. |
| Aging-associated neuropathology | ||||
| Reversal of aging effects in Wistar rats (Cowley et al., 2012) | Rosiglitazone, 3 mg/k/g, po (chow), vs vehicle control, 56 days. | 14.63 mg/d | Equal numbers of male and female rats, 22 months old vs 3 months old (control) at the initiation of treatment. | Rosiglitazone improved T1 relaxation times, improved post-synaptic component of LTP, decreased astrogliosis and RANTES expression, mediated by rosiglitazone-enhanced endothelial cell-astrocyte interactions. No effect on microglial activation. |
| Prevention of aging effects in Wistar rats (Wang et al., 2012) | Rosiglitazone, 6 mg/kg/d, po (chow), vs vehicle control, 40 days. | 29.27 mg/d | Equal numbers of male and female rats, 12 – 14 months old (middle aged) vs 1-month-old controls at the initiation of treatment. | Rosiglitazone improved water maze learning, enhanced synaptic plasticity, place cell activity, improved post-synaptic component of LTP, and restored hippocampal GLUT3 expression. |
| BOLD imaging Pharmacodynamics | ||||
| Young adult Wistar Rats (Crenshaw et al., 2015) | Pioglitazone, 0.04, 0.08, 0.16, 0.32 mg/kg/d, po (gavage) versus vehicle control, for 2 and 7 days. | 0.195, 0.39, 0.78 or 1.56 mg/d | Doses were chosen to bracket the dose used in NCT02284906 (TOMMORROW) (0.8 mg/day), after allometric scaling. Study underpowered for the large number of comparisons. | Resting state functional connectivity increased between two regions after two days of 0.08 mg/kg/day; after seven days 17 connections were changed vs. baseline across all 5 dose groups. On day 7, connectivity between CA1 and ventral thalamus was increased in all pioglitazone doses but was weakest at 0.32 mg/kg/day. |