Biomod/2013/Aarhus/Results And Discussion/System In Action
System in action
Test of photosensitizers and cholesterols
Hybridization of the chemically modified DNA strands
In order to test the ability of the photosensitizers to induce apoptosis in cells, a test system was designed. This system consisted of two complementary DNA strands, of which one was modified with the photosensitizer, In(PPa)Cl, and the other with cholesterol. In order to achive the highest yield of annealing of the modified DNA strands, the annealing of three different designs was tested. The chosen design was design DII (see structure on Figure 42). Gel analysis of this test system showed that the two modified DNA strands had annealed well (see Figure 43 below).
The gel was scanned at 670 nm, where the photosensitizer absorb light, and subsequently stained for the nucleic acids with ethidium bromide and scanned again.
On the gel in Figure 43, the single stranded DNA modified with In(PPa)Cl (lane 5) gives rise to a band below the band for the two modified DNA strands together (lanes 6 and 7). This band shift is due to the annealing of the two strands. The gel, both stained and unstained, confirms that the modified DNA strands of DII anneal. On this gel it is also possible to see the two unmodified DNA strands (lane 8), which also anneal, although the annealing is not complete.
Cell experiments with DII
As the gel in Figure 43 showed a sufficient annealing of the two strands in DII, the system could consequently be tested on cells, to see if the cholesterols were able to bring the photosensitizers close enough to the cells to mediate photoinduced cellular death. First, the localization of the cholesterols to the membrane was assayed. HeLa cells were incubated with DII in a concentration of 5µM for 30 min.
As seen from Figure 44A, no bubble formation or flattening of the cells was observed, indicating that the cells were still viable after the treatment. This indicates that no dark toxicity was induced with the DII system. In Figure 44B, a very weak fluorescence is observed. The cellular debris is seen as a bright spot at the center of the image. In order to obtain a higher degree of cellular interactions, the experiment was repeated with DII in a 5 µM concentration for a period of 2 hours. An increase in cellular interactions would also yield a higher fluorescence signal, that might help locating the position of the DII testsystem on the cell. The cells were again visualized with brightfield microscopy and fluorescence microscopy (Figure 45).
The cells were examined after 2 hours of incubation with DII (Figure 45A), and appeared healthy. In order to locate the DII system, the flourescence was imaged (Figure 45B). All parts of the cells seems stained, and therefore it was not possible to conclude anything about the localization of DII. The flourescens is nonetheless better than after 30 minutes incubation, thus higher incubation time has lead to higher cell interaction.
The DII system was then tested for the ability the photoinduce death in cells. Cells were incubated with 5 µM DII, and subsequently irradiated at 670 nm for 15 minutes.
After irradiation, the formation of vacuoles (seen as bubbles, and marked as red arrows) was observed, as well as flattened cell bodies and pronounced nuclei (black arrows). As the image in Figure 46, show clear signs of cell death, it can be concluded that the photosensitizer was able to emit singlet oxygen upon light exposure, and induce cell death. The drug delivery system designed in this project and also the designed test system DII was aiming for the tight controlled self-induced apoptosis pathway for cell death. In many cases photodynamic therapy (PDT) is highly efficient in inducing apoptosis, though high doses have been shown to result in the target cells becoming necrotic, while the surrounding tissue become apoptotic.  Thus DII might still induce apoptosis, but the applied dose is too high, and the cells have become necrotic. The brightfield image in Figure 47 shows that only cells in the irradiated area are affected, and thereby demonstrate the selective cytotoxicity of the system.
Since DII showed a very weak fluorescence signals, a new design, DII*, was made. In this system, In(PPa)Cl was replaced with PPa, which is known to be a very good flourophor. This system was then tested in cells, as described above.
Conjugation of 3’amine modified DNA strand with PPa-NHS ester (design DII*)
Design DII* was made using, the automated oligonucleotide synthesizer. The product was obtained as a black solid (12 nmol, 52 %, 239.6 μM). ε=287300 for the DNA strand (found on ) MALDI-TOF linear possitive mode (appendix): m/z calcd. for 5’- GGATGCTTCACCTAGAATGTCGGTCGATCG-3-Amino Mod C7-C33H33N4O2 [M+H]+: 9967.47 found 9962.36.
Cell experiments with DII*
In the experiments with DII, it was not possible to localize DII on the cells. To further test, if it is the cholesterol part of the construct that is inserted into the cell membrane, HELA cells were exposed to only PPa-DNA. This was done to verify, that the photosensitizer itself was note able to attach itself to the membrane, thereby indirectly establishing the role of the cholesterols in the attachment. The cells were incubated with a 1µM PPa-DNA for 30 min.
The cells in Figure 49A appear healthy, thereby indicating that the PPa-DNA strands alone have no dark toxicity. Subsequently, the cells were irradiated to examine the flourescence and the cellular interactions. From Figure 49B it can be seen that the cells are only weakly fluorescent. This indicates that the cellular interactions of the PPa-DNA strand alone is very low. This could be explained by the lack of ability of PPa to bind to the cells without the cholesterols. Figure 50 shows the comparison between untreated cells and irradiated cells, incubated with 1µM PPa-DNA for 30 min.
From Figure 50, it is further demonstrated that no cells had died upon irradiation with PPa-DNA alone. Thus PPa-DNA has no effect on the cells alone.
Test of PPa-DNA along with cholesterol
Cells were then incubated with a 1µM sample of PPa-DNA annealed with DNA-cholesterol strand (Figure 51).
As seen from Figure 51A, the cells were healthy before the irradiation, demonstrating that DII* has no dark toxicity. After irradiation, an enhanced fluorescence signal is observed, compared to cells with PPa-DNA alone, (Figure 49B), showing a higher degree of cellular interactions. This establishes the role of the cholesterols in the construct, to mediate a contact with the cell membrane and enable the cytotoxic function of the photosensitizer. To test the combined ability of the PPa-cholesterol duplex to induce apoptosis, cells were again treated with a 1µM sample of PPa-DNA annealed with DNA-cholesterol strand.
In Figure 52B a prominent vacuole formation is prominent, which indicated that the cells were necrotic. Due to the high dose, that leads to necrosis, it is not possible from these experiments to determine if this construct primarily causes apoptosis or necrosis.
Attachment of the chemical modifications to the plate
Upon showing that the origamis were able to assemble correctly with non-modified staples, the functional components were added to make sure that the origamis could fold, even when the 3’ ends of several staple strands were modified with varying modifications.
First, fluorophores were added to the staple modules that were intended to carry the peptide lock and the photosensitizers. This was done to test the ability of the cholesterols to attach the plate to the cell membrane, using confocal microscopy. The peptide lock module was labeled with Cy5 and the photosensitizer module with Cy3. A co-localization of the two flourophores on the concfocal images would indicate the presence of an intact origami. In order to label the modules with the desired Cydyes, a terminal transferase reaction was employed to elongate the staple strand with a Cy-ddUTP, which is a ddUTP that has a fluorophore, attached to it. Using ddUTP instead of dUTP or dNTP will ensure that the terminal transferase will stop the elongation process. To test if the fluorophores had been attached to the staples, a PAGE gel with the modified strands was run. It showed a Cy3 and Cy5 signal colored red and green respectively at a slightly higher position than for the control in lane 2 (Figure 53), showing that the two modules had been successfully labelled with the CyDyes. The higher bands that can be seen for both terminal transferase reactions are most likely Cy-ddUTPs bound to the terminal transferase enzyme or Cy-ddUMPs/UMPs.
Having shown that Cy3/5-ddUTP had been attached to the staples using terminal transferase, a new batch of terminal transferase reaction was made. Three different plates were made for confocal microscopy: One labeled with Cy3 and Cy5 for a Lipofectamine transfected positive control, showing cellular uptake, one with Cy3, Cy5 and cholesterols and lastly one with only Cy3 and Cy5, which is another control to show that the cell-origami interaction is less efficient in this sample compared to the one with cholesterol. The three diffent plates were run on an agarose gel, both to see if they had assembled correctly and to purify them using freeze n’ squeeze. As seen in Figure 54, the bands is clearly shifted up and shows limited aggregates. Overlay emissions from Cy3 and Cy5 makes the band appear orange. The band containing the plate was cut and purified with Freeze N’ Squeeze Spin Columns. Half of the sample with only Cy3 and Cy5 was used for Lipofectamine transfections.
KB-EGFPluc-Wagner cells were seeded out in an eight well plate, with 80,000 cells per well, and was 24 hours later treated with the differently modified plates. A negative control was included, containing untreated cells. After transfection the cells were incubated at 37°C for two hours, after which they were stained and fixated on the slide using Wheat Germ Agglutinin, Alexa Fluor® 488 Conjugate (WGA-A488), paraformaldehyde (PFM) and Prolong Gold with DAPI. No fluorescence signal was seen during confocal microscopy, (Figure 55) probably due to a too small molar amount of origami in the transfections and too short incubation time. Hence, it was not confirmed that cholesterol plays an important factor in the origami-cell interaction. Due to time constraints, the experiment could not be performed with a higher amount of origamis and a longer incubation time.
Attachment of photosensitizers to the origami plate
A plate was folded with photosensitizers linked to a staple module through an amide modification. In order to confirm its attachment, a denaturing gel was run. Three different origami samples were made. The first reaction had photosensitizer-modified staples mixed in, but was not processed with an annealing ramp (negative control). The second reaction had the photosensitizer module and was processed with a non-linear annealing ramp. The third reaction had ModLeft (staple strands without photosensitizers) and was also processed with a non-linear annealing ramp (positive control) In the gel in Figure 56, an orange band is seen in lane 4. This indicates that the photosensitizer had been successfully attached to the plate. In lane 3 and 5 green bands are visible which indicates that these plates do not have photosensitizers on them. The lower red bands which are staples with the photosensitizer module, indicate the same thing.
Melittin modified sisiRNA attached to the origami plate
To investigate if the melittin modified sisiRNA duplexes could be attached to the origami plate, the guide strand, W376, was radioactively labeled and annealed to the two segments of the passenger strand, W004 and W179. A sample of W376 was labeled using a T4 polynucleotide kinase kit and purified on a spin column. To determine the concentration of the purified, labeled W376 a sample of the labeled product was run on a 16 % denaturing PAGE gel alongside a series of samples of W376 in different, known concentrations (see supplementary data).
The labeled W376 was annealed with a 1.3 times excess of the two segments of the passenger strand, W004 and W179 and run on a 4 % agarose gel (Figure 57).
The radiolabeled sisiRNA annealed well with a yield above 90 %, and could consequently be used for attachement to the origami plate without further purification.
Attachment of sisiRNA to the origami plate
An origami plate was folded and purified and subsequently checked on a 1 % agarose gel to determine if it had folded properly (Figure 58).
From the gel it is seen that a plate had been folded, and that this plate could be used for attachment of the sisiRNA. The labeled sisiRNA duplex was annealed to the plate by mixing the plate with a 3 times excess of sisiRNA per binding site, i.e. a 30 times excess per plate, as each plate contains 10 binding sites. A sample of this reaction was run on a 1 % agarose gel and visualized using a phosphor storage screen.
To verify that the uppermost bond did correspond to a folded plate, the gel was stained with SYBR Gold and scanned (Figure 59B). When comparing the two gels, it is seen that the band containing plate and annealed sisiRNA on the storage phosphor screen corresponds to the folded plate, showing that the labeled sisiRNA did anneal to the plate.
Transfection with melittin conjugated sisiRNA attached to the origami plate
A sisiRNA duplex was annealed with W004-melittin and W179-melittin and run on a 4 % agarose gel to estimate the yield of the annealing (Figure 60).
The yield of the annealing was estimated to 95 %, which would be sufficiently high to allow annealing to the origami plate without further purification. A larger batch of the plate was prepared and purified on a spin filter and subsequently checked on a 1 % agarose gel (Figure 61).
The purified plate was mixed a 30 times excess (three times excess per binding site) of the annealed sisiRNA and incubated for 20 minutes at 37°C to allow the overhang of the sisiRNA to anneal to the staple strands of the plate.
The cells were transfected with the doubly melittin-modified sisiRNA attached to the origami plate in two different concentrations of sisiRNA, 50 nM and 10 nM. As each plate contains 10 binding sites for the sisiRNA, the concentration of the origami plate was 5 nM and 1 nM in these samples. For each concentration two transfections were performed, one with Lipofectamine and one without it. The Lipofectamine control was included to be able to compare and see if the plate caused the construct to be endocytosed by the cell. The luciferase activity was measured and corrected to the cell viability.
From the results in Figure 62, it appears that treating the cells with the plate with sisiRNAs attached did not induce a knockdown of luciferase. An apparent increase in luciferase activity was seen, for both concentrations, with or without the plate, corresponding to the results seen for the previous transfections without Lipofectamine. The two controls that were transfected with Lipofectamine, a knockdown was observed. This retained knockdown abilities of the transfected construct further indicates, that the disulfide bridges in the sisiRNA design functioned as intended, and released the sisiRNAs intracellularly.
Different testsystems were made by annealing DNA strands modified with cholesterols and photosensitizers. From the cell experiments with the different test systems, it can be concluded that the In(PPa)Cl-DNA-DNA-Cholesterol has no dark toxicity, i.e. no cytotoxic effect is observed without irridiation. Furthermore, it is the cholesterol that binds to the cells, and only upon binding to the cells, the photosensitizer can exert their cytotoxic effects. To conclude whether the cells die following the necrotic or the apoptotic pathway, more cell experiments will be needed. Both the cholesterols and the photosensitizers were succesfully attached to the plate. The cholesterol-modified plate was flourescently labeled and tested on cells. From the confocal images, it was not clear if the plates had located to the cell membrane, although this had been indirectly shown with photosensitizer test systems. For the plate with attached sisiRNA-CPP conjugates, annealing of the constructs onto the plate was achieved with a doubly CPP-modified sisiRNA through staple strand overhangs. The plate was tested on cells, both with and without a transfection agent. It was not concluded if the CPPs enabled a cellular uptake, as the sisiRNA could not be liberated outside the cells without a functional peptide lock. However, the disulfide bridges in the sisiRNA design appeared to function as intended, and liberate the sisiRNA-CPP conjugates inside the cell, where it was shown to be functional.
- Oleinick, N. L. et al. The role of apoptosis in response to photodynamic therapy: what, where, why, and how. Photochem. Photobiol. Sci. 1, 1-21 (2002).
SITEMAP | BIOMOD 2013 NANO CREATORS | Aarhus University
Copyright (C) 2013 | BIOMOD Team Nano Creators @ Aarhus University | Programming by: Peter Vium Skaarup.