Light‐Activatable MBD‐Readers of 5‐Methylcytosine Reveal Domain‐Dependent Chromatin Association Kinetics In Vivo

Abstract 5‐Methylcytosine (5mC) is the central epigenetic mark of mammalian DNA, and plays fundamental roles in chromatin regulation. 5mC is dynamically read and translated into regulatory outputs by methyl‐CpG‐binding domain (MBD) proteins. These multidomain readers recognize 5mC via an MBD domain, and undergo additional domain‐dependent interactions with multiple additional chromatin components. However, studying this dynamic process is limited by a lack of methods to conditionally control the 5mC affinity of MBD readers in cells. Light‐control of MBD association to chromatin by genetically encoding a photocaged serine at the MBD‐DNA interface is reported. The authors study the association of MBD1 to mouse pericentromeres, dependent on its CxxC3 and transcriptional repressor domains (TRD) which interact with unmethylated CpG and heterochromatin‐associated complexes, respectively. Both domains significantly modulate association kinetics, arguing for a model in which the CxxC3 delays methylation responses of MBD1 by holding it at unmethylated loci, whereas the TRD promotes responses by aiding heterochromatin association is studied. Their approach offers otherwise inaccessible kinetic insights into the domain‐specific regulation of a central MBD reader, and sets the basis for further unravelling how the integration of MBDs into complex heterochromatin interaction networks control the kinetics of 5mC reading and translation into altered chromatin states.


Vector Construction
All vectors were derived from pTzL1745 which is based on pcDNA3.1-GoldenGate-VP64(Addgene 47389) with removed VP64 and lacZα gene as described previously [1,2] To construct plasmids encoding EGFP-tagged hMBD1, a Myc tag was first introduced into pTzL1745 by quick change SDM using primers o3167/o3257, resulting in pTzL1746.The human full length MBD1 coding sequence was amplified from a human prostate cDNA library (BiocCt 10108-A-GVO-EB) using primers o3292/o3293, then MBD1 and EGFP (amplified with primers o3294/o3295) were assembled with pTzL1746 (amplified by primers o3290/o3291) via Gibson assembly, yielding pTzL1747.Finally, remaining unwanted sequences were either removed by quick change using primers o3642/o3643 to afford pTzL1836, or replaced with GGGGS linker by ligating o3387/o3388 using SacII/AscI to yield pTzL1833 (containing N-terminal Flag tags).
Mouse embryonic fibroblast NIH/3T3 cells (ATCC, CRL-1658) were maintained in the same conditions described above.The plasmid transfection of NIH/3T3 was done either by PEI as described above, or by electroporation using the 100 µL Neon Transfection System (Invitrogen, Thermo Fisher Scientific Inc.).Briefly, 3,000,000 cells were resuspended in 100 µL resuspension buffer R (Neon-transfection 100 μL kit, Invitrogen, MPK10096) with 10-30 µg of plasmid and electroporated at a pulse voltage of 1350 volts, pulse width of 20 ms, and pulse number of 2. The cells were subsequently seeded in either 10 cm culture dishes or 96-well plates containing growth media (DMEM supplemented with 10% FBS and 2 mM L-glutamine as described above, w/o penicillin and streptomycin), then incubated in a humidified 37°C incubator with 5% CO2.

Fluorescence-activated cell sorting (FACS) and light-activation of hMBD1
NIH/3T3 cells grown in 10 cm cell culture plate (Sarstedt) were transfected with plasmids encoding indicated MBD1-S45 TAG constructs (wt, R22C+R44C, C338A+C341A, or ΔTRD) and LeuRS/tRNA Leu using either PEI or Neon-transfection system.For PEI transfection, growth media was exchanged with media supplemented with 0.05 mM 1 after 3 h of transfection and allowed further expression for 21 h.For Neon-transfection, cells were directly seeded in growth media containing 0.05 mM 1 after transfection and allowed expression for 24 h.Following transfection, cells successfully expressed caged hMBD1 (and its domain mutants) were distinguished and sorted by FACS.The cells were trypsinized from the culture dish, washed once with DPBS, pelleted by centrifugation, resuspended in 500 µL warm DPBS containing 1% BSA, and subjected to cell strainer before sorting.Cell sorting was performed with Sony Cell Sorter model LE-SH800SFP using 488 and 561 nm laser coupled with 525/50 and 617/30 nm filter to detect EGFP and mCherry, respectively.To afford the desired population expressing caged hMBD1, cells similarly transfected but grown in the absence of 1 were used as the negative control to determine fluorescence intensity thresholds in cell sorting.During cell sorting, cells were kept at 37 o C in the sample loading chamber before subjected to flow system (sample pressure 4, flow rate 21 µL/min).Sorted cell population was collected at RT into a 15 mL tube containing 6 mL warm (or RT) imaging media (DMEM containing 4.5 g/L glucose, stable glutamine, sodium pyruvate, 0.5g/L NaHCO3, and 25 mM HEPES from PAN Biotech, P04-01163, supplemented with 10% FBS).The collected cells were pelleted and resuspended in warm conditioned media, then approximately 500-2,000 cells per well were seeded into the black 96-well plate with flat polymer coverslip bottom (ibidi, 89626) pre-treated with 0.01% poly-L-lysine (CAS 25988-63-0, Sigma-Aldrich, P1274).Finally, cells were incubated in a humidified 37°C incubator with 5% CO2 for 4-5 hours to allow adherence.
Live cell imaging was performed after cells have adhered and recovered from cell sorting.Before imaging experiments, DNA staining was performed by incubating cells with 1 µM SiR-DNA (SiR-DNA kit from Spirochrome AG, SC007) and 5 µM Verapamil (supplemented by the SiR-DNA kit) in growth media for 1 h in trypsinization and centrifugation, the cell pellet was washed two times with ice-cold DPBS and incubated with ice-cold lysis buffer (2.5 mM Tris-HCl pH 8.0, 500 mM NaCl, 50 mM glucose, 10 mM EDTA, 0.2% NP40, 0.2% Tween 20) supplemented with protease inhibitor cocktail (cOmplete mini easy pack, Roche, 4693124001) for 10 min on ice.The lysed suspension was passed through a AE0.8 needle for 5 times, followed by a AE0.6 needle for 5 times and incubated for another 20 min on ice.Following centrifugation at 13,400 g for 20 min at 4 ℃, the supernatant was snapped-frozen and stored at -80 ℃ before use.
The biotinylated DNA probes were generated by PCR using primers o4681 (5'-biotinylated)/o4666 to amplify the 201-bp long partial sequence of the human VEGF-A (vascular endothelial growth factor A) promoter sequence.
The PCR reaction was performed with either the normal dNTP mix (NEB, N0447L) or the 5mC dNTP mix (Biozol, ZYM-D1030) to generate the non-methylated and methylated VEGF-A probes, respectively.
The VEGF-A probes were subsequently immobilized on the streptavidin-modified magnetic beads (Dynabeads™ MyOne™ Streptavidin T1, Invitrogen, 65601).50 μg beads was washed and resuspended in 2x binding and washing buffer (10 mM Tris-HCl pH 7.5, 2 M NaCl, 1 mM EDTA).Following addition of equal volume DNA solution (containing 500 ng methylated or non-methylated VEGF-A), the beads were incubated for 1 h at room temperature with rotation.Then the beads were washed 3 times with 1x binding and washing buffer, followed by 2 times of protein binding buffer (50 μM Tris-HCl pH 8.0, 150 mM NaCl, 0.25% NP40).The desired amount of cell lysate was diluted in protein binding buffer (supplemented with 1 mM DTT and 1 mM PMSF) to 300 μL and incubated with beads at 4 ℃ for 16 h with rotation.Following washing one time with protein binding buffer, two times with TBS-T (20 mM Tris-HCl pH 7.6, 150 mM NaCl, 0.02% Tween 20), and two times with DPBS, the beads were resuspended in 15 μL DPBS and 5 μL 4x SDS sample buffer (200 mM Tris-HCl pH 6.8, 8% SDS, 40% glycerol, 0.08% bromophenol blue, 4% β-mercaptoethanol) and subsequently denatured at 95 ℃ for 5 min.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot were performed to quantify the interaction between hMBD1 proteins and the VEGF-A probes.In short, the samples were loaded onto the 8% SDS-PAGE gel and the electrophoresis was conducted with 120 V for 70 min.After blotting onto the 0.2 μM PVDF membrane with the Trans-Blot Turbo transfer kit (Bio-rad, #1704272), the C-terminally Myc-tagged hMBD1 constructs were detected by the rabbit monoclonal Myc-Tag antibody 71D10 (1:1000 dilution, Cell Signaling, #2278) and the Alexa Fluor 750-conjugated goat anti-rabbit secondary antibody (1:1000 dilution, Invitrogen, # A-21039).Additionally, beta-tubulin from the cell lysate was detected by the mouse anti-beta tubulin primary antibody (1:1000 dilution, Cell Signaling, #86298) and the Dylight680-conjugated goat anti-mouse secondary antibody (1:5000 dilution, Invitrogen, #10797775).The Odyssey® DLx imaging system (LI-COR) was used to detect immunofluorescence.

Uncaging 1 in solution
Deprotection of 1 was carried out independently in simple buffer to illustrate the uncaging efficiency.100 μL of 0.5 mM 1 in DPBS was irradiated in black 96-well plate with flat polymer coverslip bottom (ibidi, 89626) by the microscope light source mentioned above (violet LED with 395/25 nm bandpass filter at 100% intensity, Lumencor SPECTRA X light engine®).Irradiation of each well was performed by multiple image alignment (MIA) function to acquire panoramic well picture using a 10x objective.Every image in the MIA scanning area was acquired with the desired exposure time ranging from 10 to 500 ms.The UV-Vis absorption spectrum was then obtained by a NanoDrop 2000 (Thermo Fisher Scientific, v1.6).

S3§
Figure S1.Representative FCM density plots showing the incorporation of 1 for indicated hMBD1 constructs and sorting strategies.The EGFP fluorescence of HEK293T cells co-transfected with vectors encoding LRS/tRNA Leu and a) hMBD1-S45 TAG b) hMBD1-C338A+C341A-S45 TAG c) hMBD1-ΔTRD-S45 TAG , or d) hMBD1-T27 TAG and grown in the presence (left) or absence (right) of 0.05 mM 1 were analyzed by FCM 24 h after transfection.Red lines indicate the intensity thresholds determined by the fluorescence intensity of cells grown in the absence of 1 (right).The determined thresholds were used for sorting cell populations that successfully incorporated 1.Two independent biological replicates were performed.

Figure S3 .
Figure S3.Decaging efficiency of 1 (0.5 mM in DPBS).In vitro decaging experiment was conducted with the same light source used in all cell experiments.Different exposure time was applied to irradiate 100 µL sample in the flat-bottom cell culture plate (same as other cell experiments) to obtain the decaging efficiency.

Figure S4 .
Figure S4.Titrating the light dose for activating caged hMBD1 in live NIH/3T3 cells.NIH/3T3 cells co-expressing hMBD1-S45→1 and hMBD1-R22C+R44C-S45→1 were repeatedly irradiated for 5 ms with light (395/25 nm, 100% intensity, LED) in a 10 min interval, and images were acquired at 10 minutes after each light pulse to allow sufficient time for the chromocenter binding of uncaged hMBD1.Fluorescence saturation at the chromocenters was observed after 50 minutes, indicating that at duration of at least 30 ms of a full intensity (100%) light pulse is necessary to fully uncage hMBD1-S45→1.Notably, significant photobleaching or decreased cell viability due to phototoxicity were not observed after a total of 50 ms full intensity light pulse.Scale bar: 5 µm.

Figure S5 .
Figure S5.In vitro binding affinity of caged and uncaged hMBD1 to methylated and nonmethylated DNA.a) The overexpressed hMBD1 and S45 → 1 mutant (either irradiated or not irradiated) in HEK293T cell lysates were use in pull-down assays with methylated and non-methylated VEGF-A probe and blotted for the C-terminal Myc-tag (hMBD1-EGFP-Myc construct: 97 KDa).b) Beta-tubulin in the pulldown supernatant was blotted as the loading control.c) The following HEK293T cell lysates were blotted with c) anti-Myc antibody and d) anti-betatubulin antibody as input control.1: HEK293T cells expressing hMBD1; 2: HEK293T cells expressing hMBD1 S45 → 1 and irradiated with light; 3: HEK293T cells expressing hMBD1 S45 → 1; 4: Non-transfected HEK293T cells.The experiment was independently repeated three times.

Figure S16 .
Figure S16.Kinetic analysis of the chromocenter association of the EGFP-tagged hMBD1 wt, ∆TRD mutant, and the CXXC3 mutant following light activation (single transfections).

Figure S17 .
Figure S17.Schematic illustration of the image analysis workflow.Firstly, chromocenter and nucleus masks were defined by images acquired from the SiR-DNA nucleus staining channel and the EGFP/mCherry fluorescence channel, respectively.The nucleus mask was used to confine the selection of chromocenters in order to exclude false positive chromocenter detection outside the nucleus.An additional "nucleoplasma" mask was created by subtracting the chromocenter areas from the nucleus area.The mean fluorescence intensities (MFI) of chromocenters or the nucleoplasma was measured from the images acquired by the EGFP and mCherry channels.Finally, the MFI of each chromocenter was normalized to the nucleoplasma MFI background of the corresponding nucleus to distinguish the chromocenter-specific and -nonspecific localization.In kinetic measurements, these background-corrected MFI of each chromocenter were normalized to the average of normalized MFI value at t = 0, i.e. time point 20 min before irradiation.
Colocalization analysis with Pearson's correlation coefficient for Figure 2.