Defective release of α granule and lysosome contents from platelets in mouse Hermansky-Pudlak syndrome models

Abstract

Hermansky-Pudlak syndrome (HPS) is characterized by oculocutaneous albinism, bleeding diathesis, and other variable symptoms. The bleeding diathesis has been attributed to δ storage pool deficiency, reflecting the malformation of platelet dense granules. Here, we analyzed agonist-stimulated secretion from other storage granules in platelets from mouse HPS models that lack adaptor protein (AP)-3 or biogenesis of lysosome-related organelles complex (BLOC)-3 or BLOC-1. We show that α granule secretion elicited by low agonist doses is impaired in all 3 HPS models. High agonist doses or supplemental adenosine 5'-diphosphate (ADP) restored normal α granule secretion, suggesting that the impairment is secondary to absent dense granule content release. Intravital microscopy following laser-induced vascular injury showed that defective hemostatic thrombus formation in HPS mice largely reflected reduced total platelet accumulation and affirmed a reduced area of α granule secretion. Agonist-induced lysosome secretion ex vivo was also impaired in all 3 HPS models but was incompletely rescued by high agonist doses or excess ADP. Our results imply that (1) AP-3, BLOC-1, and BLOC-3 facilitate protein sorting to lysosomes to support ultimate secretion; (2) impaired secretion of α granules in HPS, and to some degree of lysosomes, is secondary to impaired dense granule secretion; and (3) diminished α granule and lysosome secretion might contribute to pathology in HPS.

Figure 1

Agonist-induced surface expression of the α granule membrane protein CD62p is impaired in HPS model platelets. Washed platelets (8 × 10⁷) isolated from WT, pallid, pearl, or light ear mice were stimulated with the indicated concentrations of thrombin (A and C) or convulxin (B and D) for 10 minutes in the absence or presence of 10 μM ADP as indicated, and then analyzed by flow cytometry for CD62p surface expression. Shown is the percentage of cells with labeling above the background level observed on unstimulated cells (see supplemental Figures 1 and 2 for gating strategies and examples of flow cytometry profiles.). Data represent mean ± standard deviation from at least 3 experiments. *P < .05; **P < .01; ***P < .005 for HPS models vs WT. ns., nonsignificant.

Figure 2

Induced secretion of α granule contents from HPS model platelets is impaired at low doses of thrombin. Washed platelets (8 × 10⁷) isolated from WT, pallid, pearl, or light ear mice were stimulated with the indicated concentrations of thrombin in 100 μL of phosphate-buffered saline/0.1% bovine serum albumin for 10 minutes in the absence or presence of 10 μM adenosine 5′-diphosphate (ADP) as indicated, and then supernatants were collected and analyzed by enzyme-linked immunosorbent assay for PF-4 (A) or PBP (C). Untreated platelets (8 × 10⁷) were lysed in 100 μL of lysis buffer, and lysates were analyzed directly for content of PF-4 (B) or PBP (D). In panels A,C, the percentage of PF-4 and PBP in releasates relative to untreated cell lysates for each sample was plotted relative to the highest percentage of release observed in a single assay for WT platelets (range: 50% to 84% for WT). Data represent mean ± standard deviation from at least 3 independent experiments. *P < .05; **P < .01; ***P < .005.

Figure 3

Impaired thrombus formation and α granule secretion after laser injury in HPS model mice. WT, pallid, pearl, or light ear mice were injected with fluorophore-conjugated antibodies to CD41 (to detect total platelets) and to P-selectin (to detect α granule secretion) and then subjected to laser-induced injury in the cremaster muscle microvasculature. Accumulation of signal for CD41 and P-selectin at injury sites was imaged by live intravital video microscopy. The total area of CD41⁺ (A-D) and P-selectin⁺ (I-L) platelet accumulation at injury sites was quantified over time. The time course in separate sets of experiments is shown for WT vs pallid and pearl mice for 4 minutes (A and I) or WT vs light ear mice for 3 minutes (C and K). Panels B,D,J,L show the peak area of accumulation at the end of the time course (mean ± standard deviation from at least 3 independent experiments). (E-H) Frames from movies at the 3-minute time point of representative thrombi labeled for P-selectin (green) and CD41 (red) in WT, pallid, pearl, and light ear mice (red and green overlay is yellow). *P < .05.

Figure 4

LAMP1 and LAMP2 do not localize predominantly to dense granules in platelets. Platelets from WT mice were fixed, permeabilized, and labeled with a rabbit antibody to the LAMP1 cytoplasmic domain together with rat monoclonal antibodies to LAMP1 (A), LAMP2 (B), or multidrug resistance protein 4 (MRP4) (C and F), or with a rabbit antibody to syntaxin 13 (STX13) and rat monoclonal antibodies to LAMP1 (D) or MRP4 (E) and fluorophore-conjugated secondary antibodies. Platelets were then analyzed by deconvolution immunofluorescence microscopy. Shown are 4 single-plane images of 1 or 2 platelets each labeled by each antibody combination. (F) Shown are 5 sequential z planes (separated by 0.2 μm) and a 3-dimensional (3D)-rendered model of a single platelet labeled for LAMP1 and MRP4. Bar represents 1 μm.

Figure 5

Agonist-induced surface expression of lysosomal membrane proteins is impaired in HPS model platelets. Washed platelets from WT, pallid, pearl, or light ear mice were stimulated as indicated with thrombin (A-D) or convulxin (E and F) for 10 minutes in the absence or presence of 10 μM ADP and then analyzed by flow cytometry for surface LAMP1 or LAMP2. Shown is the percentage of cells with labeling above the background observed on unstimulated cells (see supplemental Figure 4A,B for gating strategy and examples of flow cytometry profiles). Data represent mean ± standard deviation from at least 3 independent experiments. *P < .05; **P < .01; ***P < .005 for HPS models vs WT.

Figure 6

Impaired lysosomal enzyme release upon thrombin simulation of HPS model platelets. Washed WT, pallid, pearl, or light ear platelets (8 × 10⁷) were stimulated with varying concentrations of thrombin (A and C) for 10 minutes in the absence or presence of 10 μM ADP as indicated. Supernatants were collected and analyzed for activity of the lysosomal enzymes β-hexosaminidase (A) or β-glucuronidase (C) using colorimetric or fluorogenic substrates, respectively. Panels A′,C′ (insets) show values at 0.025 U/mL of thrombin with or without ADP on an expanded scale. Untreated platelets (8 × 10⁷) were lysed in 100 μL of lysis buffer and analyzed directly for β-hexosaminidase (B) or β-glucuronidase (D) activity. Data represent mean corrected A₄₀₅ (A, A′, and B) or fluorescence (C, C′, and D) values (mean ± standard deviation)for undiluted supernatant, normalized to the highest value in a given experiment, from at least 3 separate experiments. *P < .05; **P < .01; ***P < .005. (E) Lysates from washed WT, pallid, pearl, or light ear platelets were fractionated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and analyzed by immunoblotting for syntaxin-11, VAMP-8, or SNAP-23 (top row), or β-actin (bottom row) as a control. Bands from 3 separate experiments were quantified and plotted as the mean signal ± standard deviation for syntaxin-11, VAMP-8, or SNAP-23 relative to β-actin.