An in vivo "turning model" reveals new RanBP9 interactions in lung macrophages

Abstract

The biological functions of the scaffold protein Ran Binding Protein 9 (RanBP9) remain elusive in macrophages or any other cell type where this protein is expressed together with its CTLH (C-terminal to LisH) complex partners. We have engineered a new mouse model, named RanBP9-TurnX, where RanBP9 fused to three copies of the HA tag (RanBP9-3xHA) can be turned into RanBP9-V5 tagged upon Cre-mediated recombination. We created this model to enable stringent biochemical studies at cell type specific level throughout the entire organism. Here, we have used this tool crossed with LysM-Cre transgenic mice to identify RanBP9 interactions in lung macrophages. We show that RanBP9-V5 and RanBP9-3xHA can be both co-immunoprecipitated with the known members of the CTLH complex from the same whole lung lysates. However, more than ninety percent of the proteins pulled down by RanBP9-V5 differ from those pulled-down by RanBP9-HA. The lung RanBP9-V5 associated proteome includes previously unknown interactions with macrophage-specific proteins as well as with players of the innate immune response, DNA damage response, metabolism, and mitochondrial function. This work provides the first lung specific RanBP9-associated interactome in physiological conditions and reveals that RanBP9 and the CTLH complex could be key regulators of macrophage bioenergetics and immune functions.

Fig. 1. Generation of the RanBP9-TurnX allele.

A The RanBP9-TurnX allele includes 3 copies (3x) of the HA tag and a stop codon at the C-terminus of the third HA copy flanked by loxP sites. Cre recombination eliminates the 3xHA cassette and licenses the expression of a V5 tag (followed by a stop codon). B Genotyping of RanBP9-TurnX homozygous mice shows the presence of the expected 455 bp knock-in product.

Fig. 2. Cre-lox recombination in RanBP9-TurnX mice turns RanBP9-3xHA into RanBP9-V5.

A Schematic representation of the turning from RanBP9-3xHA to RanBP9-V5 expression upon Cre-lox recombination. loxP site sequences work as linkers between RanBP9 and the two different tags. B Genotyping by PCR amplification with the indicated primers produces the expected 334 bp fragment in RanBP9-V5 mice. WB analysis of lung (C), liver (D), kidney (E), spleen (F), and thymus (G) performed with the indicated antibodies. WT, RanBP9 knock-out (KO), and RanBP9-TT Mouse Embryonic Fibroblast immortalized with the large-T antigen (T-MEFs) [21] were used as positive and negative controls for RanBP9 WT protein and RanBP9-HA or RanBP9-V5 tagged proteins. RanBP9-3xHA is present only in RanBP9-TurnX homozygous animals while RanBP9-V5 is only detected in RanBP9-V5 turned homozygous animals where the EIIa-Cre is also present. Gapdh and Vinculin are used as loading controls.

Fig. 3. RanBP9-V5 can be efficiently immunoprecipitated from RanBP9 “turned” mouse organs.

Immunoprecipitation using anti-V5 conjugated beads shows the successful pull-down of RanBP9-V5 from A lung, B liver, and C bone marrow derived macrophages (BMDM). In all three organs, RanBP9 co-immunoprecipitates with the two CTLH members and binding partners Gid8 and Maea. Vinculin is used as loading control.

Fig. 4. LysM-Cre licenses the expression of RanBP9-V5 in lung.

WB analysis shows that in LysMBP9X animals both RanBP9-3xHA and RanBP9-V5 are clearly detectable while they are absent in LysMBP9 WT controls. WT, RanBP9 KO, RanBP9-TT T-MEFs were used as positive and negative controls. Vinculin is used as loading control.

Fig. 5. Gene ontology cluster enrichment of comprehensive RanBP9 interactors in mouse lung lysates.

The lung RanBP9-immunoprecipitated interactome suggests the involvement of the CTLH complex in the regulation of fundamental biological processes in lung cells. Metascape analyses of A 446 unique entries (recognized by the algorithm of the total 450) in the complete list of RanBP9-associated lung proteome or B 275 entries (recognized by the algorithm of the total 278) of the RanBP9-3xHA interactome or C 198 entries (recognized by the algorithm of the total 202) of the RanBP9-V5 interactome. Express gene set enrichment analyses were performed at www.metascape.org [25].

Fig. 6. RanBP9 is present in mitochondrial enriched fractions and co-immunoprecipitates with HtrA2.

A WB analysis of RanBP9 and RanBP10 WT, RanBP9 KO (9KO), RanBP10 KO (10KO) and RanBP9/RanBP10 double KO (DKO) mitochondrial, cytosolic, and whole cell fractions performed with the indicated antibodies. RanBP9, together with its paralog RanBP10, is clearly present in mitochondrial fractions. Alpha-tubulin is used as positive control for all fractions. Cox4 is used as positive control for mitochondrial fractions. Experiments were repeated three times (biological replicates) with similar results. B Immunoprecipitation using anti-V5 conjugated magnetic beads shows the successful pull-down of RanBP9, RanBP10, Gid8, and HtrA2. IP using magnetic beads conjugated with IgG does not pull-down RanBP9 nor other CTLH proteins nor HtrA2. Vinculin is used as loading control. Cell lysates were obtained from RanBP9-TT MEFs grown in standard media containing 10% serum (serum +) or in media deprived of serum for 4 h before harvesting (serum −). Shown blots are representative of two independent experiments.