Tumor suppressor Hippo/MST1 kinase mediates chemotaxis by regulating spreading and adhesion

Abstract

Chemotaxis depends on a network of parallel pathways that coordinate cytoskeletal events to bias cell movement along a chemoattractant gradient. Using a forward genetic screen in Dictyostelium discoideum, we identified the Ste20 kinase KrsB, a homolog of tumor suppressors Hippo and MST1/2, as a negative regulator of cell spreading and substrate attachment. The excessive adhesion of krsB(-) cells reduced directional movement and prolonged the streaming phase of multicellular aggregation. These phenotypes depended on an intact kinase domain and phosphorylation of a conserved threonine (T176) within the activation loop. Chemoattractants triggered a rapid, transient autophosphorylation of T176 in a heterotrimeric G protein-dependent and PI3K- and TorC2-independent manner. The active phosphorylated form of KrsB acts to decrease adhesion to the substrate. Taken together these studies suggest that cycling between active and inactive forms of KrsB may provide the dynamic regulation of cell adhesion needed for proper cell migration and chemotaxis. KrsB interacts genetically with another D. discoideum Hippo/MST homolog, KrsA, but the two genes are not functionally redundant. These studies show that Hippo/MST proteins, like the tumor suppressor PTEN and oncogenes Ras and PI3K, play a key role in cell morphological events in addition to their role in regulating cell growth.

Fig. 1.

KrsB is required for proper morphological development and multicellular behavior. (A) (Upper) WT and two independent krsB⁻ cell lines were plated on a Klebsiella aerogenes lawn, and plaques were imaged after 4 d. (Scale bar: 5 mm.) (Lower) High-magnification views of plaque edges. (Scale bar: 1 mm.) (B) Schematic of the expression constructs used in this study. (C) WT (Upper) and krsB⁻ (Lower) cells expressing C-terminally GFP-tagged KrsB constructs or empty vector were treated as in A. (Scale bar: 1 mm.) (D) Cells were plated on nonnutrient agar and imaged at the indicated time points. (Scale bar: 0.5 mm.) (E) Aggregation-competent cells were plated under buffer and imaged at the indicated time points. (Scale bar: 100 μm.)

Fig. 2.

KrsB is required for properly directed migration and random motility. (A and B) Aggregation-competent WT and krsB⁻ cells plated on a chambered coverglass were exposed to a micropipette filled with 10 μM cAMP and were imaged every 30 s for 30 min. (A) Tracks of individual cells over the 30-min period, as well as snapshots (of the top left quadrant only) taken at the indicated time points are shown. (Scale bar: 50 μm.) (B) Behavior of 98 WT and 96 krsB⁻ cells obtained from three independent experiments was analyzed for various chemotaxis parameters. (C) Aggregation-competent krsB⁻ cells expressing empty vector or C-terminally GFP-tagged KrsB or KrsB_ΔC were treated as in A. (Scale bar: 50 μm.) (D and E) Vegetative WT and krsB⁻ cells were plated on a chambered coverglass in growth medium. (D) Cells were imaged every 30 s for 30 min. Tracks of individual cells over the 30-min period are shown. (Scale bar: 100 μm.) (E) Behavior of 452 WT and 560 krsB⁻ cells obtained from three independent experiments was analyzed for motility, speed, and persistence.

Fig. 3.

Increased spreading and adhesion of krsB⁻ cells is responsible, in part, for their defective chemotaxis. Analysis was performed on WT and krsB⁻ cells. (A) Phase-contrast images of vegetative cells in glass-bottomed chambers in growth medium. (Scale bar: 50 μm.) (B) Vegetative cells expressing cAR1-mCherry were plated in glass-bottomed chambers in phosphate buffer, and serial sections were imaged with a spinning disk confocal microscope. The area of each slice from 12 WT and 20 krsB⁻ cells was measured to derive the surface area-to-volume ratio. Data are shown as mean ± SD; ***P < 0.001. (C) Vegetative cells were loaded into a microfluidic device, exposed to constant flow with pressure increasing by 0.1 psi every 2 min, and imaged every 30 s. The percentage of cells remaining attached to the surface was calculated before each pressure increase. (D and E) Vegetative cells expressing either empty vector or the indicated C-terminally GFP-tagged KrsB constructs were subjected to a rotational adhesion assay at 150 rpm (D) or 100 rpm (E). Data derived from four (D) or three (E) separate experiments, each performed in duplicate, are shown as mean ± SE; **P < 0.01; ***P < 0.001. (F) Aggregation-competent cells plated in glass-bottomed chambers with or without 0.2% BSA coating were exposed to a micropipette filled with 10 μM cAMP and were imaged every 15 s for 20 min. Tracks of individual cells over the 20-min period are shown. (Scale bar: 100 μm.)

Fig. 4.

Activation loop phosphorylation of KrsB is regulated by cAMP. (A) Alignment of human MST1 and D. discoideum KrsB protein sequence surrounding Thr183 in hMST1. (B) Vegetative WT and krsB⁻ cells expressing C-terminally GFP-tagged KrsB constructs or empty vector were lysed and immunoblotted (IB) with an antibody against MST1 phosphorylated on Thr183 (pMST1) or KrsB. The black circle indicates the position of the endogenous KrsB band. (C and D) C-terminally GFP-tagged KrsB constructs expressed in krsB⁻ cells were immunoprecipitated (IP) with antibodies against GFP and subjected to an in vitro kinase assay (KA) in the presence of [γ-³²P]ATP either alone (C) or in various combinations (D). Immunoprecipitates also were immunoblotted with an antibody against GFP. (E) Aggregation-competent WT cells were treated with 1 μM cAMP or vehicle, lysed at the indicated times, and immunoblotted as in B. Densitometric data for phospho-MST1 normalized for the KrsB signal were obtained from four separate experiments and are expressed as mean ± SD. ***P < 0.001.

Fig. 5.

Activation loop phosphorylation is required for KrsB function. Analysis was performed on WT and krsB⁻ cells expressing empty vector or C-terminally GFP-tagged KrsB constructs under the control of a doxycycline-inducible promoter. (A) Vegetative cells were lysed and immunoblotted with an antibody against MST1 phosphorylated on Thr183 (pMST1) or KrsB. (B) Cells were plated on nonnutrient agar and were imaged at the indicated time points. (Scale bar: 0.5 mm.) (C) Vegetative cells were subjected to a rotational adhesion assay at 150 rpm. Data derived from three separate experiments, each performed in duplicate, are shown as mean ± SE. *P < 0.05; ***P < 0.001.

Fig. 6.

Proposed model for KrsB function. Chemoattractants trigger transient activation of KrsB by intermolecular autophosphorylation in a heterotrimeric G protein-dependent and PI3K- and TorC2-independent manner. Expression of the phospho-mimetic form (T176E) or WT KrsB, which is constitutively phosphorylated, leads to reduced adhesion. On the other hand, krsB⁻ cells and cells in which KrsB is inactive (KrsB_T176A or KrsB_K52R) have increased adhesion that causes reduced motility and chemotaxis, resulting in a streamer phenotype.