Biomod/2013/Aarhus/Results And Discussion/Peptide lock
Introduction to peptide lock
In order to keep the plate and the dome origami closely attached to each other a lock is needed. This lock should not only keep the construct assembled, but also open as response to a specific stimulus. In this case, the presence of the endoproteolytic enzyme matrix metalloprotease 2 (MMP2) which is overexpressed by some types of metastasizing cancers.
Matrix metalloprotease 2
MMP2, also known as gelatinase A, is a zinc-dependent 72 kDa endoprotease.  It is known to participate in the degradation of the extracellular matrix (ECM), which has countless biological consequences.  Oncological studies have shown that some cancer types e.g. breast cancer, show increased MMP2 expression at the cell surface of metastasizing cancer cells.  As the enzyme degrades the ECM, it is easier for malignant cells to invade noncancerous tissue, and as ECM degrades, more space for the metastasizing cell is liberated. 
The lock was designed to contain the recognition and cleaving sequence of the enzyme, thus unlocking the construct in the vicinity of the cancer cell and exposing its functionalized interior. The protease is shown on Figure 22.
The peptide sequence which was attempted to be incorporated was H-Gly-Pro-Leu-Gly-Ile-Ala-Gly-Gln-OH.  The sequence was chosen based on its chemical simplicity, which theoretically should lead to less unwanted side reactions in the processing of the molecule. The two approaches for the synthesis of the lock are elaborated in the following sections.
The peptideThe peptide used for incorporation of the sequence into DNA is shown on Figure 23.
The glutamine was converted into its methyl ester derivative and the C-terminal residue was glycinol in the gradual build-up approach. For the fragment conjugation approach the native sequence was used and C-terminal residue was glycine. The Fmoc-protected methyl ester derivative was succesfully synthesized (see Scheme 1).
For the N-terminal linker of the peptide, pent-4-ynoic acid, has been covalently coupled onto the peptide using standard peptide coupling reagents. For the C-terminal linker, either glycinol or glycin was used. These were derived from the solid support used in the solid phase peptide synthesis.
The DNA sequence
The length and composition of the DNA strand was based on design by NUPACK  to have the least amount of intramolecular interactions. The length of the oligonucleotide was selected based on the design of the plate and dome DNA origami. Furthermore, the design allowed for the structure to "breathe", as one end of the lock would act as a zipper while the other end would be antiparallel, and thereby form a rigid structure. This is explained in the origami section.
The first approach in the synthesis of the lock was to utilize the fully automated oligonucleotide synthesis, in which a gradual build-up of DNA is achieved. According to literature[7, 8], the glutamine moiety exhibits nucleophilic features in the used standard phosphoramidite approach. To circumvent this problem, the amide was masked as a non-nucleophillic group, its methyl ester derivative. Once the peptide was coupled onto the DNA, a small organic molecule, 2-(2-(2-Azidoethoxy)ethoxy)ethanol, was coupled to the DNA peptide conjugate. The product of the click reaction would result in a free alcohol on which the oligo nucleotide synthesis could continue. The methyl ester derivative could after the synthesis had been completed, be converted back to the glutamine residue thus resulting in the original sequence. The approach is schematically depicted in scheme 2.
The second approach was to conjugate DNA to the peptide sequence, using standard peptide coupling conditions and click reaction. This approach does not require the glutamine residue to be masked, thus the unmodied peptide sequence can be used. For this approach azide and amine modified DNA strands were required. These were synthesized using standard conditions and reagents on the automated oligonucleotide synthesizer. Figure 24 shows the setup for the approach.
Results and discussion
The initial DNA strand was synthesized using the automated oligonucleotide synthesizer. The approach in which the peptide was activated with phosphoramidite before the addition to the automated oligonucleotide synthesizer did not yield any coupling products. The only observed mass in MALDI-TOF analysis was of the truncated DNA strand. Therefore the approach of phosphoramidite activation of DNA was preferred.
Purification after peptid coupling
The oligonucleotide was synthesized and activated prior to the addition of peptide. The coupling was attempted and the resulting mixture was cleaved off solid support, purified and analyzed. The analysis did not yield any DNA sequences according to LCMS, even though UV/Vis spectrophotometic analysis confirmed that the probe contained DNA, as it showed an absorbance in the 260 nm region. A PAGE gel analysis was performed, though no DNA was observed in the gel.
It was suspected that the possible DNA-peptide conjugate would form micelles once it had been cleaved into the solution in the purification step. To determine if aggregates had formed, Dynamic Light Scattering (DLS), measurements were performed. The DLS measurements are shown in table 1.
From the DLS measurements, it can be concluded that micelles have formed once the DNA-peptide conjugate were cleaved into solution.
Purification after total synthesis
To circumvent the micelle formation in solution, the strategy was to conjugate the small organic molecules onto the DNA-peptide conjugate using click reactions and continue the synthesis of the second DNA strand before purification.
The free 5’ alcohol of the 25 mer was then in situ activated with phosphoramidite prior to the addition of the peptide. After the peptide coupling, the small organic molecule was attempted coupled onto the molecule. On the free alchohol group of the peptide, continued oligonucleotide synthesis was attempted. The product was cleaved off solid support, purified and analyzed by MALDI-TOF and LCMS.
The mass of the DNA-peptide-DNA conjugate was not observed in either mass spectra (not shown). The observed mass by MALDI-TOF was approximately that of the DNA-peptide conjugate, using a linear positiv mode. The peak was broad, and it was therefore inconclusive if this contained the desired product (not shown).
The second approach was to synthesize each fragment separately and conjugate them afterwards. The before mentioned 20 and 25- mer were synthesized on automated oligonucleotide synthesis with a 3’ and 5’ amine modification, respectively. This modification allowed the use of standard peptide coupling conditions for the conjugation of the C-terminal carboxylic acid and the amine modified DNA strand. The other DNA strand was subsequently modified with a small molecule with an azide, thereby making the click coupling between the N-terminal alkyne and the DNA possible. The modification is shown in scheme 3.
Both amine modified DNA strands and azide modified DNA strand was successfully synthesized.
Amide bond formation
As numerous attempts (LINK til experiemntals)in conjugating the amine modified DNA strand to the peptide with standard peptide coupling chemistry failed, the click reaction experiments were initialized.
Conjugation through click reaction
The click reaction between the azide modified DNA and the peptide yielded the desired molecule, as confirmed by MALDI-TOF analysis. However, according to UV/VIS measurements, the yield was very low (0.14%).
As the DNA-peptide conjugation using click chemistry was successful, a template mediated reaction was performed. To increase the local concentration of the reactive moieties a DNA template directed approach was attempted. This approach is based on work by Liu and coworkers.  The template was designed to be complementary to both DNA strands of the lock. A fragment of multiple thymine bases inbetween the complementary strand corresponds to the length of the peptide. The setup is seen on Figure 25.
The DNA-peptide conjugate was not sufficiently purified, which is seen as multiple bands present in the corresponding lanes 1-4. The PAGE analysis (Figure 26) shows that no additional bands are formed in neither one of the tested conditions, thus concluding that the desired molecule was not formed.
The DNA-peptide conjugate was successfully obtained by the use of click reactions. The amide bond formation between the amine modified DNA strand and the DNA-peptide conjugate was unsuccesful and no further experiments were initiated due to time limitations of the BIOMOD competition.
- O. Mendes et al. MMP2 role in breast cancer brain metastasis development and its regulation by timp2 and erk1/2. Clin. Exp. Metastasis, 24, 341-351 (2007). 
- D. E. Kleiner et al. Matrix metalloproteinases and metastasis. Cancer. Chemother. Pharmacol., 43, 42-51 (1999). 
- R. Visse et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: Structure, function, and biochemistry. Circ. Res., 92, 827-839 (2003). 
- E. Morgunova et al. Structural insight into the complex formation of latent matrix metalloproteinase 2 with tissue inhibitor of metalloproteinase 2. Proc. Natl. Acad. Sci. USA, 99, 7414-7419 (2002). 
- E. I. Chen et al. Smith. A unique substrate recognition profile for matrix metalloproteinase-2. J. Biol. Chem., 277, 4485-4491 (2002). 
- J. N. Zadeh et al. Nucleic acid sequence design via efficient ensemble defect optimization. J. Comput. Chem., 32, 439-452 (2011). 
- T. Moriguchi et al. Synthesis and properties of aminoacylamido-amp chemical optimization for the construction of an n-acyl phosphoramidate linkage. J. Org. Chem., 65, 8229-8238 (2000). 
- J. Robles et al. Peptide-oligonucleotide hybrids with n-acylphosphoramidate linkages. J. Org. Chem., 60, 4856-4861 (1995). 
- Z. J. Gartner et al. Expanding the reaction scope of dna-templated synthesis. Angew. Chem. Int. Ed, 41, 1796-1800 (2002). <1796::AID-ANIE1796>3.0.CO;2-Z
SITEMAP | BIOMOD 2013 NANO CREATORS | Aarhus University
Copyright (C) 2013 | BIOMOD Team Nano Creators @ Aarhus University | Programming by: Peter Vium Skaarup.