C-terminus of progranulin interacts with the beta-propeller region of sortilin to regulate progranulin trafficking

Abstract

Progranulin haplo-insufficiency is a main cause of frontotemporal lobar degeneration (FTLD) with TDP-43 aggregates. Previous studies have shown that sortilin regulates progranulin trafficking and is a main determinant of progranulin level in the brain. In this study, we mapped the binding site between progranulin and sortilin. Progranulin binds to the beta-propeller region of sortilin through its C-terminal tail. The C-terminal progranulin fragment is fully sufficient for sortilin binding and progranulin C-terminal peptide displaces progranulin binding to sortilin. Deletion of the last 3 residues of progranulin (QLL) abolishes its binding to sortilin and also sortilin dependent regulation of progranulin trafficking. Since progranulin haplo-insufficiency results in FTLD, these results may provide important insights into future studies of progranulin trafficking and signaling and progranulin based therapy for FTLD.

Figure 1. Progranulin interacts with the beta-propeller region of sortilin.

(A) A schematic drawing of sortilin fragments cloned in the pDisplay vector (full length sortilin ecto domain (residues 81–757), sortilin 10CC region (residues 612–757) or sortilin beta propeller domain (residues 81–611)) and western blot to confirm their expression levels in COS-7 cells using anti-myc antibodies. (B) Binding of 100 nM alkaline phosphatase tagged progranulin (AP-PGRN) to COS-7 cells expressing various sortilin construct as indicated. Expression of these constructs was verified by anti-myc staining. (C) S283E mutation of sortilin abolished progranulin binding. Cell lysates from HEK293T cells transfected with vector control, wild type sortilin or the S283E mutant of sortilin were incubated with anti-Flag beads bound to Flag-PGRN. The presence of sortilin in the 3% input and anti-Flag IP was detected using anti-sortilin antidodies. (D) COS-7 cells expressing wild type sortilin or sortilin S283E mutant were allowed to bind to 100 nM AP-PGRN. The expression of S283E mutant was confirmed with western blot using anti-sortilin antibodies. Scale bar = 100 µm.

Figure 2. Progranulin C-terminus mediates its binding to sortilin.

(A) COS-7 cells expressing wild type sortilin were allowed to bind AP tagged progranulin fragments as indicated. (B) Quantification of the binding assay as described in (A). The intensity of AP staining was measured using the Metaexpress software (Y axis: average integrated optical density per cell (×100)). (C) C24 fragment of progranulin binds to sortilin with a similar affinity as the C100 fragment and full length progrnaulin. Sortilin expressing COS-7 cells were allowed to bind to AP tagged C100 and C24 at increasing concentrations. Scatchard plot was used to determine the Kd of the binding. (D) Flag tagged PGRN and C100 but not PGRNΔ3aa or C97 pull down sortilin in the cell lysates of NSC-34, a motor neuron cell line. (E) Alkaline phosphatase fused full length PGRN, C-terminal C100, C9 or C6 proteins binds to sortilin expressed in COS-7 cells at 100 nM, but not full length PGRN protein with additional 7 residues (GPMHETR) at its C-terminus. (F) Quantification of binding intensity for the experiment in (E) using the MetaXpress software. Scale bar = 100 µm.

Figure 3. Progranulin C-terminal peptide displaces progranulin binding to sortilin.

(A) Alkaline phosphatase fused full length PGRN and C24 binding to sortilin can be displaced by purified recombinant his-PGRN, C18 peptide (EAPRWDAPLRDPALRQLL) but not by his-PGRNΔ3aa. Sortilin expressing COS-7 cells were incubated with 150 nM purified his-PGRN, his-PGRNΔ3aa or C18 peptide of indicated concentrations for 1 hour before adding 10 nM of AP-PGRN or AP-C24. (B) Quantification of binding intensity for experiments in (A) using the ImageXpress system. Scale bar = 100 µm. *, p<0.05;**, p<0.01, n = 3–6, paired Student's T-test.

Figure 4. The C-terminal carboxylate of progranulin mediates its binding to sortilin.

(A) Illustration of binding of progranulin C-terminal tail to sortilin. Drawing was made with Pymol software. Residue backbones of sortilin are labeled in gray and progranulin labeled in green. (B) Alignment of the C-terminal residues of progranulin from different species of mammals (from top to bottom: gi|4504151|[Homo sapiens], gi|114666849|[Pan troglodytes], gi|296476263|[Bos taurus], gi|224967126|[Mus musculus], gi|204224|[Rattus norvegicus] ) and neurotensin. (C) Mouse progranulin interacts with mouse sortilin through its C-terminal tail. COS-7 cells transfected with mouse sortilin were incubated with 10 or 50 nM of AP tagged C24 and C21 of mouse progranulin (mC24 and mC21) or C24 of human progranulin (hC24). Scale bar = 100 µm.

Figure 5. Deletion of progranulin C-terminal tail abolishes sortilin dependent control of progranulin trafficking but does not affect progranulin secretion.

(A) Deletion of the PGRN C-terminal 3 residues (PGRNΔ3aa) or S283E mutation in sortilin abolishes sortilin dependent control of PGRN trafficking. AP tagged full length progranulin (FL), PGRNΔ3aa (-3aa) or AP alone were co-transfected with pcDNA vector control, wild type or S283E mutant of sortilin into HEK 293T cells or N2A cells. The AP activities in the medium were measured 4 days after transfection. The AP activities in sortilin expressing cells were expressed as percentage of vector control. *, p<0.05;**, p<0.01, n = 3–5, paired Student's T-test. (B) Deletion of progranulin C-terminal residues does not affect progranulin secretion. Progranulin expression constructs as AP fusion proteins were transfected in HEK293T and N2Acells. Progranulin levels in the cell lysates and media were determined by western blots using anti-AP antibodies.