NBR1-mediated selective autophagy promotes FA disassembly. (A) Representative spinning disk confocal image of a migrating cell stably expressing paxillin-mTurquoise, Venus-LC3, and mCherry-NBR1. Whole-cell merged image shown at left and enlarged boxed insets of two- and three-color merged images shown at right. Bar, 2.5 µm. Insets are magnified 1.4-fold. (B) Quantification of mCherry-NBR1 colocalization with FA-associated Venus-LC3 vesicles. FA-associated Venus-LC3 vesicles were identified as described in Fig. 4 B, and then the number of mCherry-NBR1-positive and -negative vesicles were enumerated. Scatter plot shows individual single cells (n = 16 total cells) and mean (line), representing 170 total FA-associated Venus-LC3 vesicles analyzed from two independent experiments. (C) Quantification of the percentage of dynamic leading edge FAs per cell targeted by autophagosomes in shCTRL or shNBR1 cells. FAs were randomly chosen independent of the GFP channel and then manually tracked for evidence of direct contact by GFP-LC3 vesicles as described in Fig. 4 B. Scatter plot shows individual single cells (n = 17 cells for shCTRL and n = 18 cells for shNBR1) and mean (line), representing 165 FAs for shCTRL and 159 FAs for shNBR1 analyzed from two independent experiments. P-value determined using unpaired t test. (D) Schematic of wild-type NBR1 (left) and autophagy-defective NBR1 (NBR1 ΔLIR, right) resulting from deletion of the LIR (aa 727–738, depicted as vertical line). Bottom diagram demonstrates inability of NBR1 ΔLIR to bind LC3 (right) and be recruited into autophagosomes, unlike wild-type NBR1 (left). (E) Nutrient-starved (HBSS, 4 h) HEK-293T cells ectopically expressing wild-type GFP-NBR1 or GFP-NBR1 ΔLIR. GAPDH is loading control. (F) Quantification of FA assembly rate constants (left), disassembly rate constants (middle), and lifetime (right) for FAs in cells expressing GFP control, GFP-NBR1, GFP-NBR1 ΔLIR, or GFP-NBR1 ΔUBA. Data presented as median (line), first and third quartile (box), and whiskers extend to ±1.5 times the interquartile range. Individual data points outside of this are shown. n = 99 FAs for GFP control, n = 84 FAs for GFP-NBR1, n = 62 FAs for GFP-NBR1 ΔLIR, and n = 62 FAs for GFP-NBR1 ΔUBA, pooled from four independent experiments. P-values calculated using a nonparametric Kruskall-Wallis test followed by Dunn post-hoc test. n.s., not significant.