HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 19/14/2017 05-4628fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by De, A. Articles by Gambhir, S. S. Search for Related Content PubMed PubMed Citation Articles by De, A. Articles by Gambhir, S. S.

Noninvasive imaging of protein–protein interactions from live cells and living subjects using bioluminescence resonance energy transfer

Molecular Imaging Program at Stanford (MIPS) and Bio-X Program, Departments of Radiology and Bioengineering, School of Medicine, Stanford University, Stanford, California, USA

¹Correspondence: The James H. Clark Center, 318 Campus Dr., East Wing, First Floor, E150A, Stanford, CA 94305-5427, USA. E-mail: sgambhir{at}stanford.edu

SPECIFIC AIMS

This study demonstrates a significant advancement of in vivo imaging of bioluminescent resonance energy transfer (BRET²) signal, enabling simultaneous visualization and quantitation from live cells and cells implanted in living mice. We hypothesized that using a cooled charge coupled device (CCD) imaging system, a BRET assay can be performed to study protein-protein interactions in small living subjects.

PRINCIPAL FINDINGS

1. BRET signal shows linearity as a function of protein content and time and can be spectrally resolved in as few as 30 cells using a cooled CCD camera
BRET² assays were first performed with serially diluted total proteins isolated from 293T cells 24 h after being transiently transfected with pBRET² and phRluc plasmids along with 1/10th amount of CMV-Fluc plasmid as transfection control. For BRET² expressing cells, GFP² signal shows a linear increase with the total protein quantity added in each well. Next we imaged component light signals at each alternate minute for 15 min directly from a fixed number of 293T cells in well plates transfected with pBRET², phRluc plasmid, or nontransfected cells with added DBC. Using GFP² filters, the average radiance at 1 min on the pBRET² plasmid transfected wells is 98.16 ± 2.2 x 10⁴, which gradually decreases to a value of 51.05 ± 0.8 x 10⁴ over 15 min. While using DBC filter, the same cells at min 1 yield an average radiance of 74.47 ± 1.89 x 10⁴, gradually decreasing to 42.7 ± 1.9 x 10⁴ over 15 min. The background average radiance was determined as 1.96 ± 0.2 x 10⁴ and 2.02 ± 0.7 x 10⁴ at minute one using GFP² and DBC filter respectively on the wells with nontransfected 293T cells with added DBC. Further, to determine the least number of cells that can be imaged using the CCD camera approach we used serially diluted pBRET² and phRluc transfected 293T cells in individual wells of a 96 well plate and imaged using a zoom lens. Our results show that individual cells from well plates can be spectrally resolved by the cooled CCD camera and as few as 30 cells per well can be resolved spectrally.

2. CCD camera imaging allows in vivo monitoring of the interactions between FKBP12 fused to N terminus of hRluc and FRB fused to C terminus of GFP² with changes in the small molecule mediator drug rapamycin
Donor (hRluc) and acceptor (GFP²) DNA with and without interacting proteins cloned in different combinations were transfected in 293T cells and checked for expression of the intact fusions by Western blot (not shown) using Renilla or living color peptide antibody for GFP² expression. Cells cotransfected with pFKBP12-hRluc and pGFP²-FRB in presence of rapamycin shows the highest signal with average radiance of 39.12 ± 0.66 x 10⁴ on the GFP² filter and 26.4 ± 1.71 x 10⁴ on the DBC filter, whereas wells not incubated with rapamycin imaged simultaneously using the GFP² filter yields average radiance of 8.73 ± 1.52 x 10⁴ (Fig. 1 B). In the presence of rapamycin, RLUC-DBC photon emission is lower than in cells not incubated with rapamycin (Fig. 1A ). The use of 320 nM rapamycin shows saturation of the BRET² signal (Fig. 1C ). We determine a concentration of 160 nM to be optimum for obtaining significant BRET signal and this was used as the optimum dose in all cell culture experiments. We repeatedly imaged 293T cells cotransfected with pFKBP12-hRluc and pGFP²-FRB by adding and withdrawing rapamycin at different time points from the incubating medium. As a control, the same transfected cells were maintained in rapamycin-containing media. The result shows that a BRET² ratio shift from 0.29 at 24 h to 1.16 at 30 h occurred in 6 h of rapamycin incubation. Once rapamycin is added in the medium, it takes 40 h to have the BRET² ratio come down to 0.17 in rapamycin-free condition. At 70 h, cells treated with rapamycin for 8 h showed a gain in the BRET² ratio of 0.79 (Fig. 1D ) signifying the occurrence of BRET.

View larger version (53K): [in this window] [in a new window]   Figure 1. BRET2 signal from rapamycin-mediated FKBP12 and FRB interaction in live cells. A) CCD camera image of 293T cells either transfected with pFKBP12-hRluc alone or cotransfected with pFKBP12-hRluc and pGFP2-FRB along with CMV-Fluc control plasmid, added with or without rapamycin (+R or –R) showing that specific GFP2 signal can be obtained only when the two proteins interact forming heterodimers in presence of rapamycin. On rapamycin-added wells, DBC photon emission yield is lower than otherwise. B) Chart showing significant BRET2 ratio (bar) and GFP2 emission light (line) as obtained only by specific interactions of FKBP12 and FRB in presence of rapamycin. 293T cells transfected with phRluc alone (RL), pFKBP12-hRluc, and pGFP2-FRB cotransfected (F-RL: GFP-F), phRluc, and pGFP2 cotransfected (RL:GFP), and mock in the presence (+R) or absence (–R) of rapamycin are shown. Error bars for GFP2 emission signal represents standard error of mean. C) CCD image of the well plate using different filters showing BRET2 signal change with varying rapamycin doses. The experiment was done by cotransfecting 293T cells with pFKBP12-hRluc and pGFP2-FRB in 24-well plate. After transfection the wells were incubated overnight with culture media containing specified (nM) dose of rapamycin. 24 h later, DBC was added (1 µg/well) after removing the media and the plate was imaged at GFP2 filter first and DBC filter for the next minute. D) Chart showing dynamic nature of BRET signal. 293T cells transfected with pFKBP12-hRluc and pGFP2-FRB plasmids in 1:1 ratio were distributed in equal numbers in a 24-well plate. The pink line represents BRET2 ratio over time, where transfected cells were maintained throughout with rapamycin (160 nM). The blue line represents BRET2 ratio, where rapamycin was added and withdrawn at different times from the incubating media of the transfected cells. As shown by the colored zone on the X-axis marked as +R or –R, rapamycin was added or withdrawn from the experimental cells.

3. Quantitative photon outputs from CCD imaging can be used to determine the BRET² ratio from live cells
Based on the average radiance (photon/s/cm²/sr) obtained by drawing a ROI over appropriate wells and applying background subtracted photon values to the standard formula for a BRET ratio calculation as [{Emission @ 500-570 – Emission @ 360-460 x (Emission of RLUC @500-570/Emission of RLUC @360-460)}/Emission @ 360-460], we established the BRET² ratio to be on average 1.5 at each protein concentration used. We also determined the average BRET² ratio as 1.3 for pBRET², which remains unchanged at each point of CCD measurement. BRET² ratio from FKBP12-FRB interaction is estimated as 1.04, indicating a strong interaction between the two proteins. The same experiments in N2a and A375 cells lines lead to similar results (Fig. 1B ).

4. A CCD camera can noninvasively monitor BRET² specific GFP² signal from living mice implanted with cells overexpressing control BRET² plasmid at subcutaneous as well as from deeper tissue depths
Implanting 3 x 10⁶ 293T cells transiently transfected with pBRET² plasmids at subcutaneous tissue depths show both GFP² and DBC component light signals. The same number of cells when delivered to lungs by i.v. injection show only BRET specific GFP² signal with an increased DBC dose of 100 µg.

5. CCD imaging allows BRET signal detection from specific protein-protein interactions from small animals with transient and stable cellular expression
For the protein-protein interaction experiment, cooled CCD camera imaging of mice (n=4) that were implanted with transiently transfected 293T cells expressing BRET², FKBP12-RLUC, and both FKBP12-RLUC and GFP²-FRB, and also received rapamycin (5 mg/kg), systemically show an average radiance on the respective sites as 24.6 ± 4.1 x 10⁴, 1.22 ± 3.2 x 10⁴, and 7.84 ± 1.9 x 10⁴ on GFP² filter (Fig. 2 ). The other set of mice (n=3) that were implanted with the same cells but did not receive rapamycin show an average radiance of 79.82 ± 11.2 x 10³, 5.81 ± 1.77 x 10³, and 13.46 ± 4.9 x 10³, respectively, using GFP² filter. These results clearly indicate that mice that receive rapamycin in vivo produce a specific GFP² signal in the appropriate locations, indicating the occurrence of BRET.

View larger version (55K): [in this window] [in a new window]   Figure 2. Detection of in vivo BRET2 signal from specific protein-protein interaction from living mice. Dorsal view of a nude mouse implanted s.c. with 5 x 106 293T cells transiently transfected with pBRET2 (L) or with pFKBP12-hRluc (LL) alone or cotransfected with pFKBP12-hRluc and pGFP2-FRB (LR) in presence (upper panel) or absence (lower panel) of rapamycin. Mice that received the small molecule mediator drug rapamycin (5 mg/kg) were injected i.p. immediately after cell implantation. Scan was performed 7 h after drug administration. Mice were scanned for 5 min integration time using either GFP2 or DBC filters in succession after injecting with 25 µg DBC i.v..

CONCLUSIONS AND SIGNIFICANCE

The interactions of specific cellular proteins form the basis for many important biological processes, including various signal transduction and hormone activation pathways involved in maintaining important biological functions. Several techniques have been developed for studying protein-protein interactions in cell culture and small living subjects. This is the first demonstration we know of where a cooled CCD has been used for spectral measurements of BRET² component light signals in determining protein-protein interactions in intact cells or living small subjects. This study shows that quantitative BRET imaging is possible from cell lysates, live cells in culture, and semiquantitatively from small animals by performing relatively simple modifications to the CCD imaging device (Fig. 3 ). The CCD imaging approach of BRET² signal is particularly appealing due to its capacity to seamlessly bridge the gap between in vitro and in vivo studies.

View larger version (56K): [in this window] [in a new window]   Figure 3. Schematic showing small molecule-mediated protein-protein interaction leading to bioluminescence resonance energy transfer (BRET). A) FKBP12 is fused to the N terminus of RLUC donor protein and FRB is fused to the C terminus of GFP2 acceptor protein. When the genes encoding for both fusion proteins are expressed in cells and rapamycin is present to mediate FKBP12-FRB interaction, resonance energy transfer occurs. This BRET signal can be detected by using the deep blue coelenterazine (DBC) substrate for RLUC. B) Diagram explaining the emission spectral properties of two distinct BRET systems. The blue line represents the YFP and RLUC emission curve showing that the spectral resolution between donor and acceptor emission wavelength is 50 nm. The black line represents the GFP2 and RLUC emission curve showing that the improved spectral resolution is 100 nm. The rectangular color zones represents the wavelength range covered by the BRET2 specific band pass emission filters as marked. C) Diagram of a black box cooled CCD imaging apparatus (Xenogen) used to measure BRET2 signal from both live cells as well as cells from within living mice. The 6 position filter wheel was equipped with BRET2 specific DBC donor emission and GFP2 acceptor emission filters. Cells in culture plate or implanted in living mice can be placed on the height adjustable stage and the donor and acceptor component lights can be captured using the highly sensitive CCD camera located on the top of the box. The white light sources allow for obtaining a gray scale photograph of the subject on which to superimpose light signal from within the subject due to BRET2.

This work also validates BRET as a powerful tool for interrogating and observing protein-protein interactions directly at tumor sites of living mice. Although BRET was demonstrated in the current work by using GFP² and RLUC, other acceptors including quantum dots should allow more optimal wavelengths (red-shifted) for use in living subjects. Based on the experimental results, we propose that CCD camera imaging is a novel tool for noninvasive monitoring of BRET signal from cell lysates, from single cells and cells from living small animals. By using specific filter sets, other BRET or FRET systems should also be able to be adapted for cooled CCD imaging.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-4628fje;

This article has been cited by other articles:

				G. Rosenfeldt, R. M. Viana, H. D. Mootz, A. G. von Arnim, and A. Batschauer Chemically Induced and Light-Independent Cryptochrome Photoreceptor Activation Mol Plant, January 1, 2008; 1(1): 4 - 14. [Abstract] [Full Text] [PDF]

				A. De, A. M. Loening, and S. S. Gambhir An Improved Bioluminescence Resonance Energy Transfer Strategy for Imaging Intracellular Events in Single Cells and Living Subjects Cancer Res., August 1, 2007; 67(15): 7175 - 7183. [Abstract] [Full Text] [PDF]

				X. Xu, M. Soutto, Q. Xie, S. Servick, C. Subramanian, A. G. von Arnim, and C. H. Johnson From the Cover: Imaging protein interactions with bioluminescence resonance energy transfer (BRET) in plant and mammalian cells and tissues PNAS, June 12, 2007; 104(24): 10264 - 10269. [Abstract] [Full Text] [PDF]

				R. Paulmurugan and S. S. Gambhir An intramolecular folding sensor for imaging estrogen receptor-ligand interactions PNAS, October 24, 2006; 103(43): 15883 - 15888. [Abstract] [Full Text] [PDF]

				W. S. El-Deiry, C. C. Sigman, and G. J. Kelloff Imaging and Oncologic Drug Development J. Clin. Oncol., July 10, 2006; 24(20): 3261 - 3273. [Abstract] [Full Text] [PDF]

				W. S. El-Deiry The Promise of Optical Technologies for Measuring Tumor Response in Cancer Therapy Trials Am. Assoc. Cancer Res. Educ. Book, April 1, 2006; 2006(1): 143 - 146. [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 19/14/2017 05-4628fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by De, A. Articles by Gambhir, S. S. Search for Related Content PubMed PubMed Citation Articles by De, A. Articles by Gambhir, S. S.