Magnetic Cellular Switches

Abstract

This paper focuses on the development of magnetic cellular switches to enable magnetic control of intracellular functions in living mammalian cells, including receptor signal transduction and gene transcription. Our approach takes advantage of the mechanosensitivity of adenosine 3',5'-monophosphate (cAMP) induction and downstream transcription controlled by the cAMP regulatory element (CRE) to engineer gene constructs that optically report gene expression in living cells. We activate transcription of these gene reporters by applying magnetic (mechanical) stress to magnetic microbeads bound to cell surface integrin receptors. In these gene reporter constructs, CRE motifs drive the expression of fluorescent proteins or enzymes that produce fluorescent products, such as DsRed and β-lactamase (BLA), respectively. We demonstrate that a chemical inducer of cAMP (forskolin) increases expression of CRE-DsRed in living cells. More importantly, a threefold increase in CRE-BLA expression is induced by application of mechanical stress to magnetic microbeads (4.5 µm) bound to cell surface integrin receptors. Induction of cAMP could be detected within 5 min using a protein fragment complementation assay involving interactions between the KID and KIX domains of the CRE binding protein linked to complementary halves of the BLA enzyme. These studies confirm that application of magnetic stress to integrins induces gene transcription by activating the cAMP-dependent transcription factor CREB. Ongoing studies focus on optimizing sensitivity and reducing signal-to-noise by establishing stable cell lines that express these gene reporters. These studies collectively demonstrate the feasibility of using magnetic technologies to control function in living mammalian cells and, hence, support the possibility of developing magnetically-actuated cellular components for use in future micro- and nanotechnologies.

Fig. 1

Chemical activation of CRE-DsRed transcription. Phase contrast and fluorescence micrographs of BCE cells expressing CRE-DsRed treated in the presence or absence of forskolin (100 µM) for 24 h. Note the large increase in the number of DsRed-positive cells after forskolin stimulation relative to untreated controls, indicating successful induction of CRE-DsRed transcription by cAMP.

Fig. 2

Magnetic actuation of CRE-BLA transcription in living cells. Phase contrast and fluorescence micrographs of BCE cells that were cotransfected with CRE-BLA and CMV-DsRed (marker). Cells were bound to RGD-coated magnetic microbeads (4.5 µm) and magnetically twisted for 10 min. Six hours later, cells were loaded with CCF-2 and imaged using a color camera with 409-nm excitation. Expression of BLA and cleavage of CCF-2 (as indicated by a shift from green to blue fluorescence) increased threefold in plasmid-containing (DsRed-positive) cells that were magnetically twisted. These data were obtained from ten to 12 different cells for each condition; similar results were obtained in two independent experiments.

Fig. 3

Rapid readout of cAMP via induction of BLA activity using protein fragment complementation. BCE cells were transfected with CMV-DsRed (marker), KIX-BLAn, and KID-BLAc. Cells were loaded with CCF-2 and then treated with 100 µM forskolin. (a) Plasmid-containing (DsRed-positive) cells demonstrated increased CCF-2 cleavage 30 min after stimulation, seen as a change from green to blue cytoplasmic staining, indicating successful reconstitution of BLA enzyme activity via KID/KIX binding. (b) Quantitation of the relative change in blue-to-green fluorescence intensity in individual cells revealed that BLA activity can be detected within 5 min following treatment with forskolin to increase intracellular cAMP. Error bars represent SEM computed from either 5 DsRed-positive cells or 20 DsRed-negative cells from the experiment shown in (a).