| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Nuclear Signalling Laboratory, Department of Biochemistry and Molecular Biology, and
Australian Research Council Centre of Excellence for Biotechnology and Development, Monash University, Clayton, Australia; and
Chromatin Transcription and Regulation Laboratory, John Curtin School of Medical Research, Australian National University, Canberra, Australia
1Correspondence: Nuclear Signalling Laboratory, Department of Biochemistry and Molecular Biology, Monash University, Clayton, VIC 3800 Australia. E-mail: david.jans{at}med.monash.edu.au
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6-fold more than those achieved by commercial liposomal preparations. The high efficiency is due in part to DNA condensation and protection against degradation in the reconstituted chromatin, as well as its ability to interact with high affinity with the importin proteins of the cellular nuclear import machinery. "Chromofection," gene delivery by protein transduction using chromatin enhanced for nuclear targeting represents an efficient means to deliver DNA to a wide variety of cell types, with the potential to treat complex genetic disorders.—Wagstaff, K. M., Fan, J. Y., De Jesus, M. A., Tremethick, D. J., Jans, D. A. Efficient gene delivery using reconstituted chromatin enhanced for nuclear targeting.
Key Words: gene therapy • histone • protein transduction • NLS
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In recent years, a novel class of proteins known as cell penetrating proteins/peptides (CPPs) have been documented to enter intact cells in receptor- and energy-independent fashion (2
3
4)
. This is attributable to short, usually <30 amino acid, regions known as protein transduction domains (PTDs), which are usually basic or amphipathic 
-helical in nature (2
, 4
5
6)
. PTD-containing proteins have been used as gene delivery vehicles, with varying degrees of success (for reviews see refs. 2
, 4
, 5
).
One of the most significant drawbacks with the use of CPPs is that although they are efficient at mediating cellular entry, they may not facilitate nuclear entry, which represents the most rate-limiting step of the nonviral gene delivery process, in that <1% of the DNA taken up by a cell is actually expressed (7
8
9
10
11)
. Excitingly, the core histone proteins H2A, H2B, H3, and H4 have all been shown to possess protein transduction potential (12
13
14)
. With their DNA binding and compaction properties, endogenous nuclear localization signals (NLSs; ref. 15
), and PTD activity, histones are exciting prospects to facilitate nonviral gene delivery to the nucleus of eukaryotic cells. Previous work into the use of core histones as gene delivery vehicles has shown that H2A (16
17
18)
, H3 (19)
, and H4 (19)
are all capable of delivering DNA to cells. We recently showed (14)
that histone H2B proteins optimized for nuclear targeting are exciting prospects in this regard and can mediate efficient gene delivery when used as dimers with H2A.
Transport of molecules >45 kDa requires a specific NLS (20
, 21)
, which is recognized by members of the importin (Imp) superfamily of proteins (20
21
22)
and in particular by the Imp subunit of the Imp/β heterodimer in the case of "classical" NLSs, such as that of the SV40 large tumor antigen (T-ag) (23)
, or by Impβ1 alone (or its homologues), as in the case of the core histones (24
, 25)
.
In this study, we demonstrate that histone H2B proteins optimized for nuclear targeting can be used to reconstitute chromatin, where the DNA is condensed and protected from degradation by nucleases and efficient delivery of DNA to the nucleus of intact living mammalian cells is achieved, with a 6-fold higher level of transgene expression than commercial liposomal delivery approaches. This gene delivery approach of chromofection is an exciting prospect for gene therapy to treat complex genetic disorders.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. H2B-GFP and H2B-NLS-GFP were generated using the Gateway cloning technology (Invitrogen, Carlsbad, CA, USA). Briefly, DNA fragments flanked by attB recombination sites encoding H2B or H2B-NLS were generated using standard polymerase chain reaction (PCR) techniques from the template pET3a-xH2B (25)
or pET3a-xH2B-NLS, which contains the T-ag NLS inserted C terminal to H2B. These PCR products were then recombined into either the pDONR201 or pDONR207 vectors (Invitrogen) via the "BP" recombination reaction, and the DONR vectors were subsequently recombined using the "LR" recombination reaction into vector pGFP-RfA-DEST, used to express GFP-fusion proteins in bacteria. The integrity of all expression constructs was verified by DNA sequencing. The reporter plasmid pGGDsRed2-Nuclear used throughout this study to express the nuclear localizing DsRed2-Nuc fusion proteins in mammalian cells was previously generated in our laboratory (14)
.
Protein purification and Imp/β dimerization
Fusion proteins GFP-H2B, GFP-H2B-NLS, H2B-GFP, H2B-NLS-GFP, and (His)6-H2B were expressed and purified from bacteria as (His)6-tagged proteins using nickel affinity chromatography under denaturing conditions (8 M area) (14)
. Histones H2A, H2B, H3, and H4 were purified from plasmid pET3a in Escherichia coli strain BL21DE3lysS from inclusion bodies under native conditions (14
, 25)
. The Imp proteins were purified from bacteria as GST fusion proteins as described previously (14
, 26
27
28
29)
. Where required, Imp and Impβ were predimerized at 13.6 µM for 15 min at room temperature to generate the Imp/β heterodimer used in binding studies (30)
.
Cell culture
MCF-7 (human breast cancer) cells were cultured in RPMI supplemented with 10% fetal calf serum, 1 mM L-glutamate, 1 mM penicillin/streptomycin, and 20 mM HEPES, pH 7.4, at 37°C in 5% CO2. Twenty-four hours before transfection or transduction, cells were seeded onto glass coverslips (15x15 mm).
Mammalian cell transfection
Lipofectamine 2000 (Invitrogen) was used according to the manufacturer’s instructions to transfect plasmid or chromatinized DNA into mammalian cells. Cells were imaged live 24 h post-transfection by confocal laser scanning microscopy (CLSM; Bio-Rad MRC-500; Bio-Rad, Richmond, CA, USA) using a x40 water immersion lens. Digitized images were analyzed using the ImageJ v1.37 public domain software (U.S. National Institutes of Health, Bethesda, MD, USA) to determine the ratio of nuclear (Fn) to cytoplasmic (Fc) fluorescence (Fn/c) according to the formula: Fn/c = (Fn–Fb)/(Fc–Fb), where Fb is background autofluorescence.
Generation of histone octamers
Histone proteins were first denatured in unfolding buffer (UB) (7 M guanidine HCl; 20 mM Tris HCl, pH 7.5; and 10 mM DTT) for 30 min at room temperature. Equimolar amounts of each of the four histones were then mixed together, except for the (His)6-H2B and wild-type octamers, where 0.95% H3 and H4 were used to encourage their incorporation into octamers, as the tetramer peak often overlaps the octamer peak in these smaller samples. After dilution to 1 mg/ml with UB, samples were dialyzed in 6000 molecular weight cutoff (MWCO) dialysis tubing against four changes of 2 L refolding buffer [2.2 M NaCl, 20 ml 100x TE (1 M Tris and 25 mM EDTA, pH 7.8), and 1 mM β-mercaptoethanol] at 4°C. Octamers were then concentrated in a centriprep YM-10 (Millipore, Billerica, MA, USA) and purified by fast protein liquid chromatography (FPLC) using a Hi-Load 16/60 Superdex 200 column (Amersham Biosciences, Castle Hill, NSW, Australia) at 1 ml/min at 4°C. Samples were analyzed by SDS-PAGE/Coomassie blue staining, the positive fractions were pooled, and protease inhibitors were added (1 mg/ml pepstatin, 1 mg/ml leupeptin, and 40 mg/ml PMSF).
ALPHAScreen assay
An ALPHAScreen assay was performed in triplicate as described previously to quantitate the interaction between the engineered histones and Imp proteins (26)
. Thirty nanomoles per liter of (His)6 binding partner and increasing concentrations of the biotinylated GST-Imps were used. All additions and incubations were performed in subdued lighting conditions due to the photosensitivity of the beads. The assay was measured on a Fusion- plate reader (Perkin Elmer, Waltham, MA, USA), triplicate values averaged and titration curves (standard three-parameter sigmoidal fit) were plotted using the SigmaPlot program (Systat Software, San Jose, CA, USA). Values in the "hooking zone," where quenching has occurred, were excluded from the analysis, as previously (26)
.
Chromatin reconstitution
The pGGDsRed2-Nuclear plasmid used in these studies can bind 30 octamers if every octamer binds to 208 bp DNA. To reconstitute chromatin, octamers were mixed with DNA (pGGDsRed2-Nuclear) at a ratio of 30 octamers/plasmid in TE/2.2 M NaCl. Samples were then dialyzed at 4°C in 6000 MWCO dialysis tubing against four changes of dialysis buffer with decreasing amounts of NaCl: 1) TE/1 M NaCl; 2) TE/0.75 M NaCl; and 3, 4) TE.
MNase assay
Three micrograms of chromatinized DNA or DNA alone was subjected, essentially as described previously (31)
, to 30 min digestion (unless indicated) by 25 mU of micrococcal nuclease s7 (MNase; Roche; Basel, Switzerland) in 2 mM CaCl2 and 10 mM Tris HCl, pH 8.0 at 25°C. After digestion, reactions were stopped with 25 µl stop solution (2.5% sarkosyl and 100 mM EDTA) per 100 µl sample. Samples were then extracted using phenol/chloroform, whereby an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) was added to each sample, followed by centrifugation for 5 min at 13,000 rpm at room temperature. The top layer was removed to a clean tube, and the phenol layer was washed with 100 µl of 10 mM Tris, pH 8.0, which was added to the first extraction. The DNA was precipitated in 0.1 vol of 3 M NaOAc and 2.5 vol 100% ethanol on dry ice for 30 min, followed by centrifugation at 13,000 rpm at 4°C for 30 min. The pellet was washed once with ice-cold 70% ethanol and dried. Pellets were dissolved in 10 µl TE, pH 7.8, and separated on a 2% agarose gel. Bands were stained with ethidium bromide and visualized under ultraviolet light.
Velocity analytical ultracentrifugation
Sedimentation velocity studies were performed essentially as described previously (32)
, whereby 5.2 µg of chromatinized or plasmid DNA was centrifuged at 18,000 rpm at 21°C for 80 min in the XL-A analytical ultracentrifuge (Beckman Coulter, Fullerton, CA, USA), with readings taken every 3 min at 260 nm. Data were subjected to Van Holde-Weischet analysis using the UltraScan 9.0 software (UTHSCSA, San Antonio, TX, USA) to determine the sedimentation coefficients.
Cytoplasmic electroporation
MCF-7 cells grown in T-75 tissue culture flasks (BD Falcon; BD Biosciences, Franklin Lakes, NJ, USA) were washed once with PBS and trypsinized for 5 min at 37°C until cells dislodged. Five milliliters of RPMI was added to each flask to neutralize trypsin and separate cells, followed by centrifugation at 800 g for 5 min at 4°C. The cell pellet was then resuspended in PBS, and cell density was determined using a hemocytometer, after which cells were repelleted and resuspended at a density of 2 x 106 cells/200 µl. Two hundred microliters of cells was mixed with 2 µg chromatinized or plasmid DNA in a 0.04-cm gap electroporation cuvette and made up to 400 µl with PBS. Samples were then pulse electroporated in a GenePulser electroporator (Bio-Rad) at 950 µF capacitance and 300 V. After electroporation, 1 ml RPMI was added to each cuvette, and the cells were replated into 10-cm dishes containing RPMI and left to express for 24 h.
Three-color fluorescence-activated cell sorter (FACS) analysis of transfected cells
Where indicated, samples were subjected to FACS analysis using a flow cytometer (FC500; Beckman Coulter) to determine transfection efficiency (nuclear localizing DsRed-fusion protein reporter expression) or DNA uptake (Cy-5 labeled DNA). GFP was also monitored in the green channel to examine histone uptake, but in most cases this was too faint to be determined accurately, due to the low number of GFP molecules per chromatin sample.
Chromofection: transduction of chromatin for gene delivery
MCF-7 cells were incubated with 2 µg chromatinized DNA in 400 µl RPMI at 37°C for 1 h. One milliliter of fresh RPMI was added to each well (6-well plate), and cells were incubated at 37°C for 24 h to express the DsRed2-fusion protein reporter. Cells were imaged live by CLSM as for mammalian cell transfection and either scored for expression of the reporter gene or subjected to three-color FACS analysis. Cell viability was determined by a manual cell count in the presence of trypan blue as an indicator of deceased cells.
Treatment with endocytosis inhibitors
MCF-7 cells were treated as previously (14)
with either 50 µM chloroquine (Sigma, St. Louis, MO, USA), 10 µM brefeldin A (Sigma), 2 µM wortmannin (Sigma), or 0.5 M sucrose or were exposed to 4°C for 30 min before and during either chromofection or transfection utilizing Lipofectamine 2000.
Liposomal disruption assay
The liposomal disruption assay was performed as described previously (14)
. Briefly, the assay was performed in triplicate, whereby 1 µM (final concentration) chromatin and 10 µl of purified liposomes, with or without 40% cholesterol, were added to 200 µl of pH buffer [20 mM citrate buffer (from 0.1 M citrate buffer: 0.018 M citric acid, 0.082 M sodium citrate), 20 mM 2-(N-morpholino)ethane sulfonic acid, 20 mM HEPES, and 150 mM NaCl, pH adjusted to desired point]. Control samples were set up in parallel, where the 10 µl chromatin sample was substituted for either water (no protein control) or Triton X-100 (100% liposomal disruption). After 30 min incubation and pH adjustment to the fluorescence maximum of calcein, the fluorescence was measured in a CARY Eclipse fluorescence spectrophotometer (Varian, Palo Alto, CA, USA) with excitation at 488 nm and emission at 520 nm. Triplicate values were averaged, the water controls were subtracted, and the results were expressed as a percentage of the Triton X-100 control (% total liposome disruption).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(14
, 15
, 24
, 25
, 33
34
35)
. Here we utilized both new histone H2B proteins optimized for nuclear targeting, as well as those we have previously characterized (ref. 14
; see Fig. 1
A for schematic). The proteins contain GFP fused in frame at either the N or C terminus, with or without the additional, optimized T-ag NLS (op-T-NLS; SSDDEATADSQHaaPPKKKRKV) that includes the CK2 phosphorylation site (SSDDE115) that enhances NLS interaction by Imp/β 100-fold and a nonphosphorylatable inhibitory cdk2 site (Ser123/Thr124 substituted with Ala/Ala; refs. 36
37
38
).
|
Control and engineered histone H2B-containing octamers were generated and purified by FPLC as described in Materials and Methods. The samples were analyzed by SDS-PAGE (Fig. 1B
). As indicated by an equimolar ratio of the four core histone proteins comparable to that of control octamers (Fig. 1B
, lane 6), each of the engineered histone H2B derivatives was readily incorporated into octamers (Fig. 1B
, lanes 1–5; it should be noted that H4 stains poorly with Coomassie and that H2A and H2B have a similar mobility in this SDS gel system; ref. 32
).
An ALPHAScreen assay (26)
was used to quantitate the binding affinity of Imps to the histone octamers, a necessary requirement if the octamers were to be used as enhancers of gene delivery. GFP-Op-T-NLS was used as a control and showed high affinity binding to Imp/β as expected (Fig. 1C
, bottom right panel). Results indicated that octamers containing GFP-H2B bound to Impβ with extremely high affinity [Kd=2.6±0.2 nM, maximal binding (Bmax)=102,086 ±6368], with little to no binding to Imp or the Imp/β heterodimer (Fig. 1C
, top left panel; Table 1
). As each of the core histones contains an endogenous NLS that is recognized with high affinity by Impβ or its homologues (14
, 24
, 25
, 33
34
35)
, there are potentially eight binding sites for Impβ in each octamer, reflected in the high maximal binding observed. Octamers containing H2B with GFP fused to its C terminus (H2B-GFP) also bound to Impβ with high affinity (Kd=3.1±0.7 nM; Bmax=86,660±1059), demonstrating that the GFP tag can be effectively placed on either end of the histone, without affecting octamer formation or Imp binding (Fig. 1C
, top right panel; Table 1
). Octamers reconstituted with the H2B derivatives containing the op-T-NLS showed significant binding to Imp/β (Fig. 1C
, top middle and bottom left panels; Table 1
) but a slightly reduced binding affinity for Impβ (Kd=3.2±0.7 nM and 3.7±0.6 nM for GFP-H2B-NLS- and H2B-NLS-GFP-containing octamers, respectively). Octamers containing the (His)6-H2B protein had no detectable binding observed for any of the Imps (Fig. 1C
, bottom middle panel; Table 1
). Accessibility of the (His)6 tag was confirmed by dot blot analysis (not shown), implying that the signal was not masked. Thus, the NLSs within the (His)6-H2B containing octamer may be less accessible than those in octamers containing engineered H2B proteins carrying the bulky GFP moiety, which may disrupt octamer formation just enough to allow Imp access to the NLSs (see also below).
|
Engineered histone-containing octamers can condense DNA into chromatin that is recognized by Imps with high affinity
The condensation of DNA into nucleosomes by histone octamers protects genomic DNA from degradation by nucleases, which can be examined in vitro by MNase digestion, where inefficient nucleosome assembly will result in more digestion of the plasmid DNA by the enzyme. When subjected to MNase digestion for 30 min, plasmid DNA was completely degraded as expected (Fig. 2
A, compare lanes a, b). Chromatin reconstituted with control octamers (nonengineered H2B) had a protected band of 146 bp (the length of DNA associated with a nucleosome) after 30 min of digestion, indicating efficient chromatin assembly (Fig. 2A
, lane c). Chromatin reconstituted with H2B-NLS-GFP- and H2B-GFP-containing octamers (Fig. 2A
, lanes f, g) appeared to form nucleosomes with an efficiency similar to that of control chromatin. On the other hand, in addition to the 146 bp DNA fragment, GFP-H2B- and GFP-H2B-NLS-containing chromatin samples (Fig. 2A
, lanes d, e) had shorter DNA digestion products, suggesting that these octamers formed nucleosomes that were somewhat less compact than control octamers and hence less DNA was protected from MNase activity.
|
Linear template DNA containing 12 repeats of a 208 bp nucleosome positioning sequence from the sea urchin 5S RNA gene (208–12 DNA; ref. 32
) was also reconstituted into chromatin using GFP-H2B-containing octamers. Partial MNase digestion (5 min) of this chromatin template resulted in an 200 bp ladder (Fig. 2B
), demonstrating the regular positioning of each individual GFP-H2B nucleosome within the 12-mer array. Similar results were observed for the other octamer samples (not shown).
DNA compaction was also examined by velocity ultracentrifugation analysis and Van-Holde Weischet analysis to determine the integral distribution of sedimentation coefficients for the population of nucleosomal arrays in the sample (Table 2
). In each case, a single population was observed, indicating no significant difference in the levels of DNA reconstitution. Plasmid DNA alone (pGGDsRed2-Nuclear) had an Save (the 50% boundary fraction) of 22S (Table 2)
. When reconstituted into chromatin, the Save increased to 33.5S. GFP-H2B- and GFP-H2B-NLS-containing chromatin both had a significantly reduced Save (26.5S and 25S, respectively), indicating that both samples showed condensation greater than that of DNA alone but less than control chromatin, bearing in mind that the molecular weight of GFP-containing nucleosomes is significantly greater than in control. This correlates with the MNase results in which these samples display incomplete protection of the DNA from digestion. H2B-GFP-containing chromatin had a sedimentation coefficient only slightly less than that observed for the control (Save of 29.5 and 33.5, respectively; Table 2
), indicating that GFP attachment to the C-terminal tail of H2B is less disruptive to DNA compaction than attachment to the N terminus, consistent with the important role played by the N-terminal tails of core histones in nucleosome assembly. Intriguingly, H2B-NLS-GFP-containing chromatin was found to be even more condensed than H2B-GFP, with an Save of 43.
|
Imp binding to the reconstituted chromatin samples was analyzed using ALPHAScreen (Fig. 2C
; Table 1
). GFP-H2B-containing chromatin was recognized by Impβ with high affinity (Kd=2.9±0.4 nM; Bmax= 25,365±858) but not by Imp or Imp/β. The maximal binding to Impβ achieved by this chromatin sample is 25% that of the GFP-H2B octamer (compare histone H2B-containing octamers with chromatin Table 1
), implying that only the NLSs on the GFP tagged histone H2B proteins are accessible when bound to DNA. Similar results were seen for the other samples, with H2B-GFP-containing chromatin demonstrating high affinity binding to Impβ and GFP-H2B-NLS- and H2B-NLS-GFP-containing chromatin being recognized by both Impβ and Imp/β with high affinity (Fig. 2C
; Table 1
).
Chromofection delivery of DNA to intact cells by reconstituted chromatin using protein transduction
Intact MCF-7 cells were incubated with reconstituted chromatin samples containing Cy-5-labeled DNA and examined 24 h later using CLSM (Fig. 3
A). In parallel, an equal amount of Cy-5-labeled DNA was transfected into MCF-7 cells using Lipofectamine 2000, according to the manufacturer’s instructions (DNA L). This DNA-to-cell ratio (1 µg DNA to 2x106 cells) was chosen because under these conditions Lipofectamine 2000 results in expression in <5% cells, enabling the full extent of the enhancement afforded by chromofection to be observed. FACS analysis was used to determine the percentage of cells expressing the reporter gene (DsRed2 positive), with results indicating that each of the reconstituted chromatin samples efficiently delivered DNA to intact cells (Fig. 3B
), a process we termed chromofection. MCF-7 cells were chosen for this study due to the fact that they are derived from human tissue and are a naturally occurring cell line; however, near identical results were also obtained in both HeLa (Fig. 3B
) and COS-7 (data not shown) cells, indicating that chromofection is not a cell type specific phenomenon. Importantly, the GFP-H2B-NLS-containing chromatin was more efficient than the GFP-H2B-containing chromatin (Fig. 3B
), indicating that the Op-T-NLS significantly enhances transfection efficiency. H2B-GFP-containing chromatin enhanced reporter gene expression to a similar extent to that of GFP-H2B-NLS-containing chromatin, indicating that the presence of GFP on the C terminus of H2B results in more efficient delivery, presumably due to the extra DNA compaction afforded by this engineered histone (see Table 2
). H2B-NLS-GFP-containing chromatin effected the highest level of transfection. FACS analysis also revealed that the percentage of cells positive for Cy-5 was similar for all samples including the control chromatin (Fig. 3C
), indicating that the engineered histones act, not by increasing the delivery of chromatin to cells by transduction, but rather by enhancing delivery to the nucleus for expression once inside the cell. It should be noted that 70% of cells take up determinable amounts of DNA, indicating the high efficiency of protein transduction-mediated delivery. This is highlighted (Fig. 3D
) by determining the ratio of cells expressing the DsRed2 reporter gene over the percentage of cells that are positive for DNA uptake (Cy-5). The control samples (including Lipofectamine 2000-delivered DNA) have very low ratios, while up to >35% of cells that take up the DNA express the reporter in the chromatin samples, with H2B-NLS-GFP-containing chromatin being the most efficient. This is a huge enhancement over traditional nonviral gene delivery methods (7
8
9)
.
|
To confirm that the mechanism of entry of the reconstituted chromatin samples into cells was via protein transduction rather than endocytosis, various inhibitors of endocytosis were tested, with FITC-poly-L-lysine-transferrin (transferrin) as a positive control (39)
. After 5 h incubation, transferrin was found to localize to distinct cytoplasmic speckles within the MCF-7 cells (Fig. 3E
, left panel), indicative of endocytotic vesicles, along with a slight cytoplasmic staining. Because the positively charged poly-L-lysine moiety in this construct is extremely "sticky," plasma membrane staining was also evident, as described previously (14)
. On incubation with the inhibitors of endocytosis, transferrin uptake was severely impaired in all cases (Fig. 3E
), with the intensity and number of cytoplasmic speckles dramatically reduced compared with untreated cells.
MCF-7 cells were incubated with each of the reconstituted chromatin samples in the presence of each of the endocytosis treatments. After 5 h of incubation, cells were thoroughly washed to remove the inhibitors and excess DNA and incubated in fresh media for 24 h, after which they were scored for expression of the DsRed2-fusion protein reporter. Each of the reconstituted chromatin samples, including the control chromatin, showed no significant effect of the inhibition of endocytosis (Fig. 3F
), indicating that chromofection delivers DNA to cells in a nonendocytotic fashion, consistent with protein transduction. In contrast, DNA delivered using Lipofectamine 2000 (DNA L) showed a significant reduction in transfection efficiency in the presence of each of the inhibitors (Fig. 3F
, left group of bars), indicating an endocytotic mechanism of entry (14)
.
Reconstituted chromatin can enhance electroporation or lipofection
Chromofection uses protein transduction as an uptake method and hence does not require the use of chemical or physical methods for the cellular uptake of DNA. We decided to test whether the engineered histone-containing chromatin could facilitate transgene expression effected by the more conventional delivery approaches of electroporation and lipofection. In the case of electroporation, a voltage of 300 V was used to deliver material specifically into the cytoplasm of living cells (MCF-7 human breast cancer cells); this voltage has been empirically determined to be optimal for cytoplasmic delivery but leaves the nuclear envelope intact (data not shown).
Each of the reconstituted chromatin samples or DNA alone was delivered into the cytoplasm of MCF-7 cells, which were examined 24 h later for reporter gene expression (Fig. 4
A). The results (Fig. 4B
) showed that each of the reconstituted chromatin samples containing the engineered histone H2B proteins had significantly higher (P<0.001) levels of gene expression than in the case of DNA alone or the control chromatin sample, implying that reconstituted chromatin is more efficient in facilitating the transport of DNA from the cytoplasm to the nucleus than control chromatin, due to increased DNA condensation, protection against nucleases and Imp binding (see above). Reconstituted chromatin containing the additional op-T-NLS had a significantly higher level of transfection than that without, confirming that the op-T-NLS promotes the nuclear delivery of the DNA. H2B-NLS-GFP-containing chromatin showed the greatest level of transfection, consistent with its increased DNA compaction and nuclease protection as well as nuclear targeting abilities.
|
In parallel, each of the chromatin samples or DNA alone was delivered to cells using Lipofectamine 2000 (Invitrogen). For these experiments, the DNA (pGGDsRed2-Nuclear) was labeled with Cy-5 before reconstitution into chromatin. Twenty-four hours post-transfection, samples were analyzed by FACS (Fig. 4C
), with results indicating that control chromatin did not significantly enhance the level of transfection above that of DNA alone. Each of the reconstituted chromatin samples, however, did show significantly greater levels of transfection above that of DNA alone, with H2B-NLS-GFP-containing chromatin benefiting the most. In all cases, addition of the op-T-NLS to the engineered histone H2B construct resulted in increased levels of transfection efficiency of the reconstituted chromatin, consistent with that observed with electroporation. The percentage of Cy-5-positive cells (Fig. 4D
) was similar in each sample (P>0.05 in all cases), indicating that reconstituted chromatin does not affect the number of cells that take up the DNA but rather the number of cells that express protein from this DNA, presumably as a result of increased delivery of intact DNA to the nucleus.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6-fold more efficiently than commercial liposome reagents. Addition of the op-T-NLS that confers high affinity recognition by Imp/β results in increased DNA delivery to the nucleus and a further increase in transfection efficiency. Significantly, the process of chromatin-mediated gene delivery, chromofection, occurs via protein transduction, meaning that it can be simply added to intact cells to effect transgene expression. This is exciting, as previous attempts to utilize PTDs to deliver DNA to cells reported that although other PTD components themselves undergo protein transduction, DNA uptake is via traditional endocytosis (4
, 40
, 41)
. We hypothesize that the efficient delivery of reconstituted chromatin is dependent in part on the DNA condensation ability of the histones and protection against nucleases, consistent with previous observations (14
, 19)
.
In this study we utilized engineered histone H2B proteins with, in addition to the op-T-NLS, an N- or C-terminal GFP moiety (14)
. All of the engineered histone H2B proteins can readily assemble into histone octamers, which bind Impβ with nanomolar affinity. Octamers containing the op-T-NLS-containing engineered histones also show high affinity binding for Imp/β, indicating a switching of the import pathway of these proteins from Impβ to Imp/β mediated, as previously demonstrated for a GAL4 based gene delivery vehicle (14
, 42)
. Each of the engineered histone H2B-containing octamers was used to reconstitute plasmid DNA into chromatin, resulting in condensation and protection of the DNA and maintaining the ability to be recognized by Imps.
Significantly, chromofection resulted in expression in >35% of cells to which DNA has been delivered, 40-fold more efficient than that observed for traditional nonviral mediated techniques, which demonstrate expression of <1% of the delivered DNA. When taking into account the amount of DNA delivered, chromofection is 4- to 5-fold more effective than our previously utilized engineered histone H2B monomers or dimers (14)
at achieving transgene expression (20% transfected cells per µg DNA vs. 2% transfected cells per µg DNA, respectively). Our data demonstrate that this is due to enhanced DNA compaction, protection against nuclease activity, and nuclear delivery, all vital to ensuring expression of the final gene product. Although similar amounts of DNA were taken up by cells, regardless of the delivery method, the engineered histone H2B-containing chromatin samples all showed significantly higher levels of transfection than liposomal delivered DNA. Addition of the op-T-NLS to the H2B constructs resulted in significantly higher levels of reporter gene expression (2-fold), highlighting the potential of these chromatinized samples for DNA delivery. Similarly, the reconstituted chromatin samples were also able to significantly enhance DNA delivery afforded by electroporation or lipofection, suggesting that a combinatorial reconstituted chromatin/electroporation or reconstituted chromatin/lipofection approach may enable easy transfection of difficult cell lines, such as primary cells. Furthermore, it is possible that by incorporating elements for episomal replication and scaffold attachment into the plasmid (such as those found in the episomal replication plasmid pEPI) stable transfections can be produced.
In summary, chromofection is a highly efficient means to deliver DNA to eukaryotic cells. The large size of the plasmid DNA utilized here (6000 bp) indicates that chromofection is capable of delivery of more than just a few basepairs of DNA and, since there is theoretically no limit to the amount of DNA that can be chromatinized, chromofection has the potential to deliver large pieces of DNA encoding multiple genes, allowing for the treatment of complicated genetic disorders.
Received for publication November 26, 2007. Accepted for publication February 21, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
