Turku Centre for Biotechnology

University of Turku


Cell Adhesion and Cancer

SHARPIN deficiency causes aberrant collagen fibre organization in the mouse mammary gland stroma. Second harmonic generation (SHG) and unfiltered multiphoton (MP) imaging of collagen fibres surrounding terminal end buds in 6-7 weeks old wild-type (wt) and Sharpin-/- mouse mammary glands are shown. Adapted from Peuhu et al. 2017)


SHANKs (1-3) are a family of scaffold proteins largely known for their role in maintaining normal brain function. Genetic alterations in SHANK3 are responsible for a spectrum of neuropsychiatric disorders, including autism spectrum disorders, schizophrenia, intellectual disability and manic-like behaviour. In addition to expression in the post synaptic density of excitatory neurons, SHANKs are also expressed in peripheral organs with largely unknown functions. In an international collaborative study, we uncovered SHANK1 and SHANK3 as novel integrin inactivators. We resolved the  structure of the SHANK N-terminus and found that it contained an unexpected Ras-association (RA) motif with high affinity for active Ras and Rap G-proteins. SHANK interaction with these specific G proteins prevented integrin activation mediated by the talin-RIAM1-Rap1 complex. Loss of SHANKs, leading to enhanced integrin activity, promoted cell spreading, migration and invasion. Interestingly, we also found that autism-related mutations within the SHANK N-terminus disrupt G protein binding and abrogate SHANK-dependent downregulation of integrin function in cancer cells and in neurons (see Publications, Lilja et al., 2017 and an NCB News and Views by Atherton and Ballestrem 2017).


AMPK (5' AMP-activated protein kinase) is an important energy sensor and has been extensively studied as a central regulator of cellular metabolism; however, in recent years it has become apparent that AMPK directly phosphorylates non-metabolic proteins and regulates other cellular functions, such as transcription, cell polarity, mitosis, migration and adhesion. We found that inhibition of AMPK triggers a significant increase in integrin activity in fibroblasts, accompanied by enhanced formation of fibrillar adhesions (centrally located α5β1 integrin-ECM contacts rich in tensin) and fibronectin fibrillogenesis. In line with these data, we discovered an upregulation in tensin expression, the major component of fibrillar adhesions, following AMPK silencing or knockdown. As tensins bind to overlapping talin-binding sites within the β-integrin tail, we propose that tensins maintain integrins in an active state following talin-dependent activation. Importantly the role of AMPK in this process, as a regulator of tensin expression, provides a link between energy sensing and integrin signalling that may have wide implications in fibrosis and cancer (see Publications; Georgiadou et al., 2017).

Transition from inactive to active integrin requires separation of the integrin legs. Talin binding to the β-integrin subunit promotes an open/active integrin conformation with high affinity for extracellular ligands, whereas SHARPIN binding to the α-integrin subunit prevents association of integrin activators with the β tail, thereby maintaining integrins in an inactive state.


SHARPIN is a widely expressed multifunctional protein implicated in cancer, inflammation and linear ubiquitination. We discovered that siRNA-mediated silencing of SHARPIN expression results in increased integrin activity, implicating SHARPIN as a negative regulator of integrin activation and signalling. We identified a conserved SHARPIN binding site within the α-integrin cytoplasmic domain and found that SHARPIN-integrin interaction inhibits the recruitment of integrin activators to the β-tail (see Publications; Rantala et al., 2011).

More recently, we demonstrated a role for SHARPIN in epithelial homeostasis in the developing mouse mammary gland. We observed increased intergin activity and altered stromal matrix deposition in vivo in the absence of SHARPIN, leading to diminished mammary ductal invasion and growth (see Publications; Peuhu et al., 2017).

Left panel: The overall layout of the cell spot microarray (CSMA) used for the siRNA screen for endogenous integrin inhibitors in prostate cancer cells (PC3). The inset shows representative images of array spots stained for active integrin. Right Panel: The top 44 siRNAs with the largest effect on integrin activation represented by Z-scores. Solid coloured bars indicate siRNAs for those genes in which both individual siRNAs significantly increased integrin activity. These include SHARPIN and PRKAA2 (also known as AMPK), PRKCH (also known as nPKC-eta). For full details of the genes implicated in integrin activation see Rantala et al,. 2011 (full reference found in Publications).

Regulation of integrin activity

Keywords: AMPK, integrin activity, matrix remodelling, Ras-association domain, RNAi screens, SHANK, SHARPIN, talin, tensin

Introduction: Integrin heterodimers can exist in a bent/closed conformation with low affinity for extracellular ligands (‘inactive’) or in an extended/open conformation with a high affinity for ligands (‘active’). As a consequence of this conformational switch, integrins are able to transmit signals bidirectionally across the cell membrane. Engagement of extracellular ligand by integrins elicits signalling responses within the cell (‘outside-in’ signalling), whereas binding of intracellular proteins such as talin and kindlins to the β-integrin tail (NPXY motifs) promote the ligand-binding receptor conformation (‘inside-out’ signalling).

Our research:Our RNAi screens have thus far revealed several candidate proteins implicated in the regulation of integrin activity and importantly have led to the identification of SHARPIN, SHANKs and AMPK as negative regulators of integrin function (see Publications: Rantala et al., 2011; Lilja et al., 2017; Georgiadou et al., 2017 and below for a summary of these works).