Membrane Localization of HspA1A, a Stress Inducible 70-kDa Heat-Shock Protein, Depends on Its Interaction with Intracellular Phosphatidylserine

Abstract

HspA1A is a cytosolic molecular chaperone essential for cellular homeostasis. HspA1A also localizes at the plasma membrane (PM) of tumor and stressed cells. However, it is currently unknown how this cytosolic protein translocates to the PM. Taking into account that HspA1A interacts with lipids, including phosphatidylserine (PS), and that lipids recruit proteins to the PM, we hypothesized that the interaction of HspA1A with PS allows the chaperone to localize at the PM. To test this hypothesis, we subjected cells to mild heat-shock and the PM-localized HspA1A was quantified using confocal microscopy and cell surface biotinylation. These experiments revealed that HspA1A's membrane localization increased during recovery from non-apoptotic heat-shock. Next, we selectively reduced PS targets by overexpressing the C2 domain of lactadherin (Lact-C2), a known PS-biosensor, and determined that HspA1A's membrane localization was greatly reduced. In contrast, the reduction of PI(4,5)P₂ availability by overexpression of the PLCδ-PH biosensor had minimal effects on HspA1A's PM-localization. Implementation of a fluorescent PS analog, TopFluor-PS, established that PS co-localizes with HspA1A. Collectively, these results reveal that HspA1A's PM localization and anchorage depend on its selective interaction with intracellular PS. This discovery institutes PS as a new and dynamic partner in the cellular stress response.

Keywords: heat-shock proteins; lipid–protein interactions; phosphatidylserine; plasma membrane.

Figure 1

Plasma membrane localization of HspA1A increases during recovery from mild heat-shock at 42 °C. (A) Representative images of Henrietta Lacks’ ‘Immortal’ (HeLa) cells expressing empty EGFP or GFP-HspA1A. The cells were grown under normal conditions (37 °C) or heat-shocked for 1 h at 42 °C and allowed to recover for 8 h. The plasma membrane was stained using the wheat germ agglutinin lectin Alexa Fluor^® 555 conjugate (WGA-AF555). The images showing the +ve Product of the Differences from the Mean (PDM) (see materials and methods) values were included to show co-localization. Scale bar = 10 μm. (B) Quantification of the corrected total cell fluorescence (CTCF) as a ratio between the total GFP-HspA1A fluorescence of the plasma membrane and the rest of the cell. Center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by R software; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; crosses represent sample means. The experiment was repeated three times and the number of cells used to generate each boxplot was (from left to right) n = 16, 34, 48, 17, 38, and 41. The P values of the student t-test were: WGA-AF555-42/HspA1A-GFP-42 < 0.0001; GFP-HspA1A-37/GFP-HspA1A-42 < 0.0001; GFP-37/GFP-42 = 0.2019; GFP-37/GFP-HspA1A-37 < 0.0001; GFP-42/GFP-HspA1A-42 < 0.0001.

Figure 2

The plasma membrane (PM)-localization of HspA1A is minimally affected by the total membrane charge. Quantification of the corrected total cell fluorescence (CTCF) as a ratio between the total GFP-HspA1A fluorescence of the plasma membrane and the rest of the cell in the presence or absence of sphingosine. The polybasic charge sensor R-Pre-GFP was used as a control. Center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by R software; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; crosses represent sample means. The experiment was repeated three times and the number of cells used to generate each boxplot was (from left to right) n = 48, 32, 28, 26, 41, 39.

Figure 3

The PM-localization of HspA1A is inhibited by the removal of phosphatidylserine (PS) (presence of Lact-C2), but it is minimally affected by the removal of PI(4,5)P₂ (presence of PLC-δ-PH). The quantification of the corrected total cell fluorescence (CTCF) as a ratio between the total GFP-HspA1A fluorescence of the plasma membrane and the rest of the cell is presented for the different conditions used. Center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by R software; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; crosses represent sample means. The experiment was repeated three times and the number of cells used to generate each boxplot was (from left to right) n = 48, 44, 26, 41, 52, 34.

Figure 4

HspA1A co-localizes with TopFluor-PS in human cells. (A) Representative images of HeLa cells showing the co-localization of HspA1A (top panel) and Lact-C2 (lower panel) with PS at 37 °C. (B) Quantification of the co-localization using the Manders’ co-localization coefficient analyses using 20, 26, and 25 cells from three independent experiments. Center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by R software; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; crosses represent sample means. The images showing the +ve PDM (see materials and methods) values were included to show co-localization. Scale bar = 10 μm. The P values of the student t-test were: RFP/HspA1A < 0.0001; RFP/ LactC2 < 0.0001; HspA1A/LactC2 < 0.0001.

Figure 5

Cell surface biotinylation reveals that HspA1A’s PM-localization depends on its binding to PS. (A) Representative Western blots showing the total and biotinylated fractions of HEK293 cell lysates transfected with HspA1A-GFP and RFP-C1 vector, and HspA1A-GFP and Lact-C2-mCherry. The antibody used was the OmicsLink™ Anti-GFP Tag Antibody Mouse Monoclonal IgG1 ((CGAB-GFP-0050); 1:1000); (B) The nitrocellulose membranes used in (A) were stained with the Reversible Protein Stain Kit (Pierce) and were also blotted with the control antibodies listed below. The antibodies used were the Beta Actin, Clone: 13E5, Cell Signaling ((Rabbit mAb #4970) 1:1000); Na+/K+ ATPase α (ATP1A1) antibody RabMAb^® ((EP1845Y); (2047-1); 1:1000); and RFP Tag Monoclonal Antibody (RF5R) ((MA5-15257); 1:1000)). M: molecular size marker (BioRad Dual Color Protein Ladder; approximate sizes shown on the left side of the blots). (C) Quantification of the antibody detected signals of the HspA1A (GFP tagged) in the presence or absence of Lact-C2, as well as the Lact-C2 (mCherry tagged) presented as a ratio between the biotinylated (PM) fraction and the total cell lysate. Densitometry values are averages of three independent experiments (n = 3). Center lines show the medians; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; crosses represent sample means. The P values of the student t-test were: HspA1A-37/HspA1A-42 = 0.003; HspA1A-42/HspA1A-42-LactC2 < 0.0001; HspA1A-37-LactC2/HspA1A-42-LactC2 = 0.0413; HspA1A-37/HspA1A-37-LactC2 = 0.0212.