Subscriber access provided by UNIV OF DURHAM
Quantifying Cellular Internalization with a Fluorescent Click Sensor Laura Ina Selby, Luigi Aurelio, Daniel Yuen, Bim Graham, and Angus P. R. Johnston ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b00219 • Publication Date (Web): 20 Apr 2018 Downloaded from http://pubs.acs.org on April 20, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sensors
Quantifying Cellular Internalization with a Fluorescent Click Sensor Laura I. Selby†,‡, Luigi Aurelio†, Daniel Yuen†, Bim Graham† and Angus P. R. Johnston*†,‡ †Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Victoria, Australia, ‡ARC Centre of Excellence in Convergent Bio-Nano Science and Technology, Monash University, Parkville, Australia ABSTRACT: The ability to determine the amount of material endocytosed by a cell is important for our understanding of cell biology and in the design of effective carriers for drug delivery. To quantify internalization by fluorescence, the signal from material remaining on the cell surface must be differentiated from endocytosed material. Sensors for internalization offer advantages over traditional methods for achieving this as they exhibit improved sensitivity, allow for multiple fluorescent markers to be used simultaneously, and are amenable to high throughput analysis. We have developed a small fluorescent internalization sensor, similar in size to a standard fluorescent dye that can be conjugated to proteins and uses the rapid and highly specific bio-orthogonal reaction between a tetrazine and a trans-cyclooctene group to switch off the surface signal. The sensor can be attached to a variety of materials using simple chemistry and is compatible with flow cytometry and fluorescence microscopy, making it a useful tool to study the uptake of material into cells. KEYWORDS: endocytosis, internalization, click chemistry, fluorescence, flow cytometry, live cell microscopy Endocytosis is a critical cellular process that governs the entry of a variety of materials into the cell. It plays a role in numerous cell functions, including the uptake of nutrients,1 signal transduction2 and in immune responses.3 In addition, endocytosis is the main route of entry for many nanomaterials,4 highlighting its importance in the development of drug delivery systems.5–7 Fluorescent probes are widely used to investigate interactions of proteins and nanoparticles with cells, however directly measuring fluorescence only gives information about association, rather than internalization.8 Quantifying the amount of material internalized requires the ability to distinguish between material inside the cell from that remaining bound to the surface. The majority of current techniques for determining internalization rely either on confocal microscopy to identify material inside the cell membrane,9 or on non-specific removal of surface fluorescence using trypan blue quenching10,11 or acid-washing.12,13 Microscopy limits the number of cells that can be analysed while the other methods remove all surface fluorescence, making them incompatible with immunophenotyping and other fluorescent cell surface markers. Dyes that change their fluorescence in response to pH have also been used to quantify internalization, by exploiting the acidification of endosomes that occurs during endocytosis to sense material inside the cell.14 However, these dyes can remain measurably fluorescent at the neutral pH of the extracellular environment, resulting in high background levels and a low signal-to-noise ratio.
Recently, we developed a DNA based internalization sensor (Specific Hybridization Internalization Probe; SHIP) that can be specifically quenched through hybridization.15–18 In this assay, a fluorescent internalization probe (FIP) is conjugated to the material of interest and surface bound material is quenched using a complementary DNA strand coupled to a quencher. This method has previously been demonstrated to be superior to both trypan blue and acid wash methods for removing cell surface fluorescence without interfering with other fluorescent signals. However, labeling of proteins and particles with large tags has been shown perturb intracellular trafficking pathways.19,20 Although it was demonstrated that FIP oligonucleotides did not interfere with the trafficking of the proteins or nanoparticles studied, the size and negative charge of these probes can become important when used to study the internalization of smaller materials. The need to quantify the internalization of materials of all sizes prompted us to design an equivalent sensor but with a reduced molecular weight and with a net charge closer to zero. Here, we present a fluorescent sensor that is similar in size to a conventional fluorophore and that exploits the Inverse Electron Demand Diels-Alder (IEDDA) click-type reaction between tetrazine and trans-cyclooctene groups to quantify the internalization of material into cells (Figure 1a). We demonstrate its ability to visualize the internalization of protein through fluorescence microscopy
ACS Paragon Plus Environment
ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 10
EXPERIMENTAL Oligonucleotide Sensor. The previously published oligonucleotide internalization sensor15 was purchased as the following custom sequences: 5’-Cy5-TCA GTT CAG GAC CCT CGG CTazide-3’ (FIP) and 5’-AGC CGA GGG TCC TGA ACT GA-BHQ23’ (QComp) from Integrated DNA Technologies (IDT). Oligonucleotides were reconstituted in Milli-Q water to prepare 150 µM or 600 µM stock solutions. Nanobody Expression. Histidine-tagged anti-human epidermal growth factor receptor 2 (anti-HER2) nanobody clone 2D321 was expressed in E. coli BL21(DE3) and purified using immobilized metal affinity chromatography followed by size exclusion chromatography. Nanobody protein concentration was determined via the UV/Vis absorbance at 280 nm using an extinction coefficient of 35,075 M-1 cm-1, calculated using ProtParam.22 DNA Sensor Bioconjugation. Labeling reactions used 50 µL of 5.0 mg mL-1 transferrin in phosphate buffered saline (PBS). An excess of 1 mg mL-1 dibenzocyclooctyne N-hydroxysuccinimide ester (DIBO-NHS) dissolved in dimethyl sulfoxide (DMSO) was added (20 equivs) and allowed to react for 2 h at 4˚C. Unreacted DIBO-NHS was removed using a Zeba Spin Desalting column, 7 kDa molecular weight cutoff (MWCO). The column was first equilibrated with PBS by washing with 300 µL and spinning at 1500 g for 1 minute, repeating 3 times. After adding the sample and spinning at 1500 g, an excess of FIP-azide was added (1.5 equivs) and allowed to react overnight at 4˚C. Excess FIP-azide was then removed using an Amicon Ultra – 0.5 mL 30 kDa MWCO centrifugal filter. The filter was rinsed with 500 µL PBS by spinning at 13,000 g for 0.5 min. After adding the sample, the unit was topped up to reach a total volume of 0.5 mL and spun at 13,000 g for 3 min and repeated until the filtrate was clear. After washing, the unit was placed upside and spun at 1000 g for 2 min to collect the product. Concentration of protein and label were calculated using an extinction coefficient and molecular weight for transferrin of 87,000 cm-1 M-1 and 81 kDa respectively,23 and an extinction coefficient of 250,000 cm-1 M-1 for Cy5.24 Absorbance was obtained via a NanoDrop Spectrophotometer ND-1000 and the degree of labelling was determined as described previously.25
Figure 1. Scheme of click internalization sensor. (a) The fluorescent component of the sensor is attached to a material of interest and contains a click group compatible with that on the quencher unit. Chemical structures of the (b) fluorescent component (sCY5TCO) and (c) quencher component sQSY-Tet. (d) Following binding and internalization of the protein or material into cells at 37oC, uptake is halted by bringing the temperature down to 4oC. The fluorescence of any material on the surface is removed selectively by adding the quencher component of the sensor, leaving the intracellular signal unaffected.
and quantify the amount of material internalized via flow cytometry.
pHrodo Red Bioconjugation. pHrodo red NHS ester (1.5 equivs, 3.3 µg, 5 nmol) at 1 mg mL-1 was added to 5 mg mL-1 human holo transferrin (0.25 mg, 3.1 nmol) and incubated at 4oC overnight before purifying using a Zeba Spin Desalting column, 7 kDa MWCO. The concentration and degree of labeling were obtained via a NanoDrop Spectrophotometer ND-1000 ND by measuring the absorbance at 560 nm and using an extinction coefficient of 65,000 M-1cm-1 obtained from the supplier. Click Sensor Bioconjugation. Labeling reactions used 50 µL of 5.0 mg mL-1 transferrin (0.25 mg), 38 µL of 0.9 mg mL-1 2D3 (0.035 mg) or 14 µL of 14 mg mL-1 trastuzumab (0.2 mg) in PBS. An excess of the click sensor (sCY5-TCO) (2 – 4 equivs to the amount of protein) was added to the solution followed by N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC) and NHS dissolved in PBS to bring the final concentration of those reagents to 0.1 M and 5 mM respectively. In the case of trastuzumab, the sensor was pre-incubated with 10, 50 or 100 mM EDC/5mM NHS for 15 min before adding this volume to the antibody in addition to Alexa Fluor 488-NHS dissolved in DMSO (2.5 equivs). Sulfo-Cy5 was also conjugated to
ACS Paragon Plus Environment
Page 3 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sensors
trastuzumab with sCy5-NHS (2.5 equivs) as a comparison to samples using EDC/NHS to assess cross-linking. All reactions were incubated overnight at 4˚C. Unreacted dye was removed using a Zeba Spin Desalting column, 7 kDa MWCO. The concentration and degree of labeling were obtained via a NanoDrop Spectrophotometer ND-1000 ND using an extinction coefficient of 271,000 cm-1 M-1 for sCy5-NHS at 646 nm and 73,000 M-1 cm-1 for Alexa Fluor 488 at 495 nm obtained from the suppliers. An extinction coefficient of 250,000 M-1 cm-1 was used for sCy5TCO.26 For trastuzumab, an extinction coefficient of 225,000 M-1 cm-1 at 280 nm and an approximate molecular weight of 150 kDa were used.27 Cell Culture. The non-adherent human lymphoblast cell line C1R (ATCC CRL-1993) was maintained at 37°C and 5% CO2 in DMEM with GlutaMAX supplemented with 10% FBS, 100 U mL1 penicillin and 100 µg mL-1 streptomycin. The adherent human ovarian epithelial cell line SKOV-3 (ATCC HTB-77) was maintained at 37 °C and 5% CO2 in McCoy’s 5a modified medium with 10% FBS. Cell Surface Quenching and Internalization of Transferrin. C1R cells at a concentration of 2×106 cells mL-1 were split into tubes with 4×105 cells per sample and cooled on ice for 10 min. Transferrin labeled with sensor was added at a final concentration of 15 µg mL-1 for 15 min and left on ice for 15 min to allow binding to the cell surface. Cells were washed twice in cold PBS before incubating the cells at either 4oC or 37oC for 15 min. In experiments involving wheat germ agglutinin- Alexa Flour 488 (WGAAF488), the lectin was added at 1 µg mL-1 for the final 5 min of binding time. The cells were split into a 96-well V-bottom plate and spun at 300 g for 5 min. The cells were resuspended in 200 µL of PBS with or without quencher (4 equiv for the click sensor, 5 equiv for the DNA sensor) to the total amount of sensor added, calculated using the degree of labelling. The samples were mixed and then left on ice for approximately 15 min before analysis by flow cytometry.
For microscopy, cooled C1R cells (4×105 cells per sample) in tubes were incubated with transferrin at a final concentration of 50 µg mL-1 at 4oC or 37oC for 25 minutes with 1 µg mL-1 WGAAF488 added for the final 5 min of binding. Cells washed twice in cold PBS by spinning at 300 g for 5 min. Samples were split in two and resuspended in PBS with or without 4 equivs of QSYTet, based on the degree of labelling of Tf with sCy5-TCO before imagining by fluorescence microscopy. Trastuzumab and 2D3 Binding and Internalization. SKOV-3 cells were seeded at 1×105 cells in 400 µL per well in 24-well plates one day prior to the experiment. One plate of cells was cooled to 4oC on ice for 15 min while the other was maintained at 37oC. 4 nM antibody or 8 nM nanobody labelled with sCy5-TCO was added per well at 0.5, 1, 2 and 4 h. The cells were washed twice in cold PBS and detached with 200 µL TrypLE for 10 min. 100 µL 1% bovine serum albumin in PBS was added to each sample and the entire contents transferred to a 96-well, V-bottom plate. The cells were spun at 400 g for 5 min and resuspended in 200 µL PBS with or without 0.5 µM sQSY-Tet before analysis by flow cytometry. Quenching Efficiency and Internalization. To calculate internalization and quenching efficiency, the average mean fluorescence intensity (MFI) of the background (PBS for solution-phase, control cells for flow cytometry experiments) was subtracted from
each sample. Quenching efficiency was calculated as a percentage from the following equation using the MFI: Q) η" = $1 − ' +, × 100 (1) N) Where, hQ = quenching efficiency, N0 = MFI of the unquenched sample at 4oC, Q0 = MFI of the unquenched sample at 4oC. The percentage of material internalized, compensating for incomplete surface quenching was calculated from the follow equation:28 I3 = 41 −
N5 − Q5 6 × 100 (2) N Q N5 − 5 ) N)
Where, If = fraction internalized, N1 = MFI of the unquenched sample at 37oC, Q1 = MFI of the quenched sample 37oC For the trastuzumab and 2D3 experiment, the average quenching efficiency determined at 30 min and 4 h was used to compensate for incomplete surface quenching.
RESULTS AND DISCUSSION Internalization Sensor Design. When designing an internalization sensor, two key attributes are important: specificity for the target and the rate at which the sensor can detect the event. For flow cytometry and microscopy applications using live cells, the quenching of surface fluorescence must proceed to completion over a time scale of seconds to several minutes. In the SHIP assay, hybridization of complementary DNA sequences is extremely rapid. The second-order rate constant is approximately 104 – 105 M-1 s-1 for oligonucleotides containing 15 – 20 base pairs.29,30 To achieve similar reaction kinetics while maintaining a low molecular weight, we looked to a sensor design based on the bio-orthogonal Inverse Electron Demand Diels-Alder (IEDDA), reaction between tetrazine and trans-cyclooctene groups, which is renowned for its high specificity and rapid reaction rate.31 Reaction rates for click chemistry pairs are highly variable and depend on the particular groups used. Strainpromoted azide-alkyne cycloadditions (SPAAC), such as the reaction between an azide and dibenzocyclooctyne (DIBO) have a reaction rate in the order of 0.2 – 0.5 M-1 s-1 and were considered too slow to be practical for this purpose, especially as the concentration of the sensor expected to be present on the cell surface is very low. To carry out reactions quickly at low concentrations requires a reaction pair with a high rate constant. The reaction of tetrazine (Tet) with trans-cyclooctene (TCO) is rapid, with reaction rates up to 2x106-fold higher than azide-cyclooctyne reactions. TCO has been used in numerous in vitro labeling applications, demonstrating its suitability for bio-orthogonal labelling.32–34 Based on these attributes, we chose the TCO-Tet click chemistry reaction pair as the basis for the sensor.
ACS Paragon Plus Environment
ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 2. Sensor response to quencher components in solution. (a) Kinetics of quenching 0.6 µM sCy5-TCO with 5 equivalents of sQSY-Tet, BHQ2-mTet and BHQ2-Tet or DNA sensor with complementary quencher strand (QComp) in solution. (b) Quenching efficiency at equilibirum: 0.5 min for sQSY-Tet and QComp, 6.5 min for BHQ2-Tet and 60 min for BHQ2-mTet. The mean of n = 3 or n = 2 (sQSY-Tet) data points is plotted with error bars representing the standard deviation.
The internalization sensor is composed of two components (Figure 1a). The first component contains a fluorescent dye (sCy5), a click group (TCO) and anchor point for conjugation to a molecule or particle of interest (sCy5TCO, Figure 1b). The distance between each of these functional groups was extended with short polyethylene glycol (PEG) linkers to minimize steric hindrance so reactivity was maintained once conjugated to the material of interest.35 The molecular weight of sCy5-TCO is ~ 1390 g mol-1, slightly larger than the commonly used dye Alexa Fluor 647 (~1250 g mol-1) but significantly smaller than FIP (~ 7 kDa). The second component contains a quencher dye with absorbance matched to the emission of sCy5 (BHQ2 or sQSY21) and a compatible click group. The reaction rates of 1,2,4,5-tetrazines are tunable by modification of the attached functional groups at the 3- and 6- positions on the ring, but stability is compromised in exchange for speed.36 Substitution of a methyl group at the 6- position (mTet) exhibits higher stability under biological conditions, but is over 30-fold less reactive than the non-methylated counterpart (Tet).37 We synthesized quencher components with mTet and Tet modifications to determine the compromise between optimal stability and reaction kinetics. It was anticipated that degradation over the timescale needed for the quencher to react to the fluorescent component was likely to be insignificant. To quantify internalization, the fluorescent component was first attached to the material of interest and incubated with cells for a desired length of time. Cooling the cells to 4oC arrests endocytosis,15,38,39 allowing the quencher to interact with remaining surface-accessible sensor, but not internalized material (Figure 1d). For this reason, it is crit-
Page 4 of 10
ical that the quencher is not membrane permeable and remains restricted to the extracellular environment. To investigate membrane permeability, we developed two quenchers. The first, BHQ2, is a hydrophobic molecule and when conjugated to mTet (BHQ2-mTet, Figure S-1) or Tet (BHQ2-Tet, Figure S-2), retains its small size and neutral charge. To minimize membrane permeability a second quencher was synthesized based on a sulfonated QSY21 (sQSY-Tet, Figure 1c). The introduction of charged sulfonate groups limits the ability of the quencher to diffuse across the cell membrane. The quencher was also modified to include two PEG spacers between the sQSY and Tet group, as well as an additional PEG linker terminated with a carboxylic acid. These modifications resulted in a 2-fold increase in the molecular weight of the quencher component, which together with the introduction of several charged groups assists in preventing diffusion across the cell membrane.
Quenching Efficiency in Solution. The quenching kinetics and the quenching efficiency of the Tet quencher was compared with mTet quencher and the DNA internalization sensor. Addition of BHQ2-Tet or sQSY-Tet to sCy5-TCO in solution caused an immediate reduction in the intensity (Figure 2a). Both Tet quenchers and the DNA sensor reached their maximum signal reduction rapidly (
