Behavioral deficits and progressive neuropathology in progranulin-deficient mice: a mouse model of frontotemporal dementia

Abstract

Progranulin haploinsufficiency causes frontotemporal dementia with tau-negative, ubiquitin-positive neuronal inclusion pathology. In this study, we showed that progranulin-deficient mice displayed increased depression- and disinhibition-like behavior, as well as deficits in social recognition from a relatively young age. These mice did not have any deficit in locomotion or exploration. Eighteen-month-old progranulin-deficient mice demonstrated impaired spatial learning and memory in the Morris water maze. In addition to behavioral deficits, progranulin-deficient mice showed a progressive development of neuropathology from 12 mo of age, including enhanced activation of microglia and astrocytes and ubiquitination and cytoplasmic accumulation of phosphorylated TDP-43. Thus, progranulin deficiency induced FTD-like behavioral and neuropathological deficits. These mice may serve as an important tool for deciphering underlying mechanisms in frontotemporal dementia.

Figure 1.

Normal motor behavior in PGRN-deficient mice. A) Open field: WT and PGRN-deficient (KO) mice were tested at 5 (n=15/group), 12 (n=15/ group), and 18 (n=10/group) mo of age in the open-field test. Distances moved during 3 consecutive days are shown. B) Accelerated rotarod: WT and KO mice were tested at 5 (n=15/group), 12 (n=15/group), and 18 (n=10/group) mo of age in the accelerated rotarod. Results show latencies before falling. Results are expressed as means ± se.

Figure 2.

Emotional behavior deficit in PGRN-deficient mice. A) Depression test: WT and PGRN-deficient (KO) mice at 4 (n=15/group), 12 (n=15/group), and 18 (n=10/group) mo of age were suspended by the tail (left panel) or placed in a forced swimming tank (right panel). Time stayed immobile was recorded, and results were expressed as means ± se. P value, Student's t test. B) Anxiety test: Wild-type (WT) and PGRN-deficient (KO) mice at 3 (n=15/group), 12 (n=15/group), and 18 (n=10/group) mo of age were subjected to the elevated plus maze test. Time spent in the open arms or the closed arms were recorded; results are expressed as percentage of time spent in the open arms. Results are expressed as means ± se. Values of P calculated by Student's t test.

Figure 3.

Social recognition deficit in PGRN-deficient mice. A) Social recognition: WT and PGRN-deficient (KO) mice were tested at 1 (n=15/group), 7 (n=15/group), and 18 (n=10/group) mo of age. In the first 10 min (top upper panel), time spent sniffing an empty wire container (white) or an identical container holding an unfamiliar mouse (gray, stranger 1) was recorded. In the next 10 min (bottom panel), time spent sniffing the container with the now familiar stranger 1 (gray) or the container with a new mouse (black, stranger 2) was recorded. B) Olfactory discrimination: WT and PGRN-deficient (KO) mice were subjected to olfactory discrimination test at 2 (n=15/group), 6 (n=14/group), and 18 (n=10/group) mo of age. Time spent sniffing a filter paper spotted either with water or sesame oil was recorded for 3 min. Results are expressed as means ± se. Values of P calculated by Student's t test.

Figure 4.

Cognitive deficit in old PGRN-deficient mice. A) Morris water maze hidden platform: WT and PGRN-deficient (KO) mice at 5 (n=15/group), 12 (n=15/group), and 18 (n=10/group) mo of age were trained to locate the hidden platform for 5 consecutive days. Time spent finding the platform was recorded each day. B) Morris water maze probe trial: On d 5, mice were subjected to the probe trial without platform, and percentage of time spent in the target quadrant (left panel) and number of target quadrant crosses (right panel) were measured. Results are expressed as means ± se. Values of P calculated by Student's t test.

Figure 5.

Activation of microglial in PGRN-deficient mice. A–C) Coronal sections of hippocampus (A), cortex (B), and thalamus (C) from age-matched wild-type (WT) and PGRN-deficient (KO) mice (n=5) were immunostained for CD68 expression. D–F) Quantitative analysis of CD68⁺, expressed as cells per square millimeter. Results are expressed as means ± se. *P < 0.0001 vs. WT; Student's t test. Scale bars = 100 μm (A), 200 μm (B, C).

Figure 6.

Activation of astrocytes in PGRN-deficient mice. A–C) Coronal sections of hippocampus (A), cortex (B), and thalamus (C) from age-matched WT and PGRN-deficient (KO) mice (n=5) were immunostained for GFAP expression. D–F) Quantitative analysis of GFAP⁺, expressed as cells per square millimeter. Results are means ± se. *P < 0.0001 vs. WT; Student's test. Scale bars = 100 μm (A), 200 μm (B, C).

Figure 7.

Cytosolic phosphorylated TDP-43 accumulation in PGRN-deficient mice. Coronal sections of hippocampus (A) and thalamus (B, C) from age-matched WT and PGRN-deficient (KO) mice (n=5) were immunostained for phosphorylated TDP-43. Solid arrowheads indicate cytosolic accumulation of phosphorylated TDP-43. Scale bars = 100 μm (A), 200 μm (B), 10 μm (C).

Figure 8.

Hippocampal ubiquitination in PGRN-deficient mice. Coronal hippocampal sections from 3, 12, and 18 mo-old WT and PGRN-deficient (KO) mice (n=5) were stained for antiubiquitin reactivity. Representative images demonstrate increased ubiquitination in 12- and 18-mo-old KO mice. Scale bar = 100 μm.