Author: George Hwang
Institution: University of Connecticut
Date: October 2006
Abstract:
De novo designed peptide nanoparticles can have the size of a virus and carry antigenic information on their surface. Through genetic engineering, it is possible to mount almost any peptide-based epitope on the surface of the nanoparticles. In theory, the nanoparticles can elicit an immune response when used as a vaccine. Burkhard et. al. have designed such a peptide nanoparticle that consists of 60 peptide monomers and which has a diameter of about 20 nm. Our goal was to develop a novel Human Immunodeficiency Virus (HIV) vaccine by mounting HIV epitopes on the surface of the nanoparticles. In this project, the forward and reverse strands of DNA oligomers that coded for a highly conserved portion of the HIV protein gp41 were annealed. The annealed oligomers were successfully ligated into the expression vector pP_3a that coded for the peptide monomer of the core particle. The recombinant DNA was transformed into the Escherichia coli strain BL21(DE3)pLysS and the bacteria were cloned. The protein was expressed and purified successfully. The dialysis and refolding step did not work successfully and we propose a possible improvement.
Introduction
The Human Immunodeficiency Virus (HIV), the causative agent of the Acquired Immunodeficiency Syndrome (AIDS,) is marked as one of the most devastating viruses of our time. Since the first diagnosis of AIDS in 1981, HIV has claimed more than 20 million lives, and infected an estimated 38 millions people around the world (Armstrong et al. 2004).
A major challenge in the control of HIV stems from the difficulty of developing an effective vaccine against the virus. The virus dynamically mutates and generates quasi-species to fool the immune system. The virus has also evolved a heavily glycosylated antigenic determinant that antibodies usually bind ineffectively (Smith 2003). The traditional approaches of vaccine design using live-attenuated or inactivated viral particles have been proven to be limited in eliciting neutralizing antibodies (Letvin 2002), while other novel designs are being actively researched.
Virus-like particles (VLPs) are particles that contain viral components and authentic viral proteins that are readily recognized by the immune system. The ability of VLPs to promptly elicit an immune response has been demonstrated (Noad and Roy, 2003; Bachmann and Dyer, 2004). Burkhard et al. have recently designed peptide-based nanoparticles that carry authentic viral epitopes on the surface to elicit an immune response (Raman et al. 2005). These particles are easier to handle and to produce than VLPs, and can play versatile roles because of their structure. In this research, we attached part of a highly conserved glycoprotein of HIV (N-peptide of gp41) onto the nanoparticles. In theory, this gives the particles the ability to repeatedly display antigenic information to the host and therefore elicit an immune response against HIV.
Methods and Materials:
Annealing of the DNA oligomers:
Figure 1: The oligo sequence of the combination of epitope HIV_common and HIV-ep2, along with the corresponding restriction enzymes and amino acid information. The length of the combined epitope is 134 base pairs.
Varying the length of the surface epitopes elicits varying immune responses. In an attempt to produce a vaccine that would elicit an optimized response, we designed constructs of nanoparticles with HIV gp41 epitopes of three different lengths. The DNA oligomers coding for the epitope consisted of a common oligomer (named HIV-common) that had the size of 96 and 97 bp (forward and reversed strand), respectively and a varied-length oligomer. The three varied length oligomers had the size of 38 and 37 bp (HIV-ep2), 59 and 58 bp (HIV-ep3), and 80 and 79 bp (HIV-p4), respectively. The oligomers ends were constructed as restriction sites such that the overhanging 5' end of the HIV-common oligomers would anneal to the 3' end of the vector cut by the restriction enzyme Xho I, and the 3' end of the HIV-ep2, HIV-ep3 and HIV-ep4 to the 5' end of the vector cut by the restriction enzyme EcoR I. The 3' end of the HIV-common and the 5' ends of the HIV-ep2, HIV-ep3 and HIV-ep4 were designed as 5 bp overlapping complementary ends of the sequence TTAAA. See Figure 1 for the epitope oligomer sequence and the location of the restriction sites and overhanging ends. In this work I produced the construct containing HIV-ep2.
The forward and reverse strands of HIV-ep2 DNA fragments were annealed by heating them to 90º C for 1 minute and then cooling them slowly to room temperature. The annealed oligomers were desalted using the Bio-Rad Micro Bio-Spin 30 column (catalog# 732-6223). The forward and reverse strands of HIV-common were annealed in the same way.
Digestion of the expression vector (modified pPEP-T vector):
The modified pPEP-T vector pP_3a that coded for the peptide monomer of the core particle was cut with restriction enzymes EcoR I (Fisher Scientific #BP3362) and Xho I (Fisher Scientific #BP3450): 2 µg of plasmids were mixed with 4 µl of MilliQ H2O, 2 µl of 10 x EcoR I buffer, 2 µl of 10 x BSA, 1 µl of EcoR I enzyme, and 1 µl Xho I. The mixture was incubated at 37º C for 3 hours followed by 65º C for 15 min to deactivate the enzymes. The final volume was 20 µl. The digested vectors were dephosphorylized with Antartic phosphatase (New England Biolabs #M0289S) using the following procedure: 20 µl of digested vector mixture was mixed with 2 µl of 1 x phosphatase buffer, 2 µl of MilliQ H2O, and 0.5 µl of phosphatase at incubated at 37º C for 30 min.
Ligation of the digested vector and annealed HIV-ep2 and HIV-common oligomers:
13.4 ng of the annealed DNA oligomer inserts (HIV-ep2 and HIV-common) was added to 100 ng of digested vector pP3_A and mixed with MilliQ H2O at 45ºC for 5 min followed by cooling on ice. Then, 1 µl of 1 x ligation buffer and 1 µl of T4 DNA ligase (Fisher Scientific #BP3210) were added to make a final volume of 10 µl. The ligation reaction was carried out at 16º C overnight. The new plasmid was designated pH2_3b.
Transformation with recombinant plasmids pH2_3b:
Competent E. coli cells of the strain DH5α were heat-shock transformed with 5 µl of the novel pH2_3b plasmids (Inoue, H. et al. 1990). The transformed cells were concentrated by centrifugation and plated onto a 100 µg/ml ampicillin LB plate and incubated overnight at 37º C. The colonies counts were compared with a negative control, which was transformed with cut and dephosphorylated pP_3a plasmids without adding DNA oligomers, to confirm a positive insertion. The plasmids of the colonies carrying a positive insert were extracted using a Qiagen QIAPrep Spin MiniPrep Kit (catalog #27104) and subsequently sequenced to ensure the correctness of the DNA sequence. This plasmid stock was used for the subsequent expression of the peptides.
Expression and purification of the H2_3b construct: E. coli BL21(DE3)pLysS cells were transformed with the recombinant plasmids pH2_3b using the same method as mentioned above. This is an optimized E. coli strain for protein overexpression. The cells were cloned and grown overnight at 37º C in the presence of Ampicillin (200 µg/ml) and Chloramphenicol (30 µg/ml). The expression was induced using 1 mM Isopropyl--D-thiogalactopyranoside (IPTG) for 4 hours. The cells were harvested by centrifugation at 4000 x g for 15 min. The pellets were collected and frozen at -20ºC overnight.
For purification, the pellets were resuspended in lysis buffer (9 M urea, 100 mM NaH2PO4, 10 mM Tris, and 10 mM β-Mercaptoethanol, pH 8.0) and sonicated. The cell components were removed by centrifugation (40 min at 35000 x g). The supernatant was mixed with Ni-NTA Superflow nickel-charged resin (Qiagen #30410)) for 1 hour. The impurities were washed using 4 wash buffers (same composition as the lysis buffer, the wash sequence was pH 8.0 -> pH 6.3 -> pH 5.9 -> pH 5.0). The peptide monomers were fractionally eluted with two elution buffers (a) 9 M urea, 100 mM NaH2PO4, 10 mM Tris, 10 mM ß-Mercaptoethanol, pH 8.0 and 500 mM imidazole and b) 9 M urea, 100 mM NaH2PO4, 10 mM Tris, 10 mM ß-Mercaptoethanol, pH 8.0 and 1 M imidazole. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS PAGE) was performed to see in which fractions the peptides were eluted.
Refolding of the H2_3b construct:
The peptides were refolded by a step-wise removal of urea from the refolding buffer by dialysis. The order of dialysis was 8 M urea, 20 mM Tris, 150 mM NaCl, 1 mM DTT, pH 7.5, room temperature -> 6 M urea, 20 mM Tris, 150 mM NaCl, 1 mM DTT, pH 7.5, room temperature -> 3 M urea, 20 mM Tris, 150 mM NaCl, 1 mM DTT, pH 7.5, 4º C -> 1 M urea, 20 mM Tris, 150 mM NaCl, 1mM DTT, pH 7.5, 4º C-> 0M urea, 20 mM Tris, 150 mM NaCl, 1 mM DTT, pH 7.5, 4º C. The last step was repeated twice to remove the urea completely. The protein concentration was kept at 0.1 mg/ml. Chelex 100 Resin beads (Bio-Rad #142-1253) were added to the dialysis buffers to remove nickel ions. This is to prevent aggregation of the peptides caused by their histidine tag. After dialysis the peptides were first filtered through a 0.1 µm polyvinylidene fluoride membrane filter to remove aggregates (Millipore #SLVV 033 RS). The peptides were subsequently concentrated using an Amicon Ultra-15 centrifugal filter with a molecular weight cut-off of 5000 Da (Millipore #UFC9 005 08).
Results:
Annealing of the DNA oligomers HIV-ep2:
Figure 2: Agarose gel of the annealing of HIV DNA oligomers. The gel shows that the annealed oligomers traveled slower than the single forward or reversed oligomers, which was expected. The molecular weight marker on the first lane did not fully reveal the actual sizes of the oligomers due to the short run time of this particular gel.
The annealing of DNA oligomers was analyzed by a 1.5 % agrose gel. The result is shown in Figure 2. The gel shows that the annealed oligomers (center, bright band) travel slower than the individual forward and reverse oligomers. The heavier band represents the annealed oligomer.
Transformation of the DH5α E. coli strain with the Novel Recombinant Expression Vector pH2_3b:
Figure 3: Agarose gel of the extracted pH2_3b plasmids. The 3000 bp fragments represent the vector. The 134 bp fragments correspond to the HIV inserts.
We counted 422 colonies containing pH2_3b vector with the HIV DNA oligomer inserts and 177 colonies of the negative control (transformed with the cut and dephosphorylated pP_3a vector only). The DH5α; E. coli cells with the pH2_3b plasmids outgrew the negative control by a significant margin of 245 colonies.
Extraction of Positive Recombinant DNA Inserts:
The DNA plasmids in the transformed E. coli DH5α strain were extracted and digested with EcoR I and Xho I restriction enzymes to see which colonies contained positive HIV insertions (Figure 3). The properly transformed samples would result two segments, the vector p_P3A DNA (approx. 3000bp) and the combined HIV epitope (134bp). Figure 3 shows that 6 out of the 13 tested colonies contain the positive inserts.
Sequencing of Positive Inserts:
The DNA material extracted from the bacteria colonies showing positive HIV DNA insertions was sent to the Biotechnology facility for sequencing. The sequencing result, in both forward and reverse strands, showed that the insertions contained exactly the same sequence as found in the HIV DNA oligomers.
Figure 4 shows the result of forward strand.
Figure 4: Result of DNA sequencing in forward (5' to 3') direction. The shaded region corresponds to the DNA sequence on the HIV oligomers, which matched the original viral DNA perfectly.
Expression and Purification:
Figure 5: SDS PAGE showing the purification result. The 14.4 kDa bands indicated presences of the monomers (MW 14.2 kDa). However, close examination showed 3 visible bands around the 14.4 kDa region indicating that some degradation occurred. The bands at 42 kDa may be trimers, which also appeared in the purification results of the P_3a core peptides (without viral extensions).
12-14 g of cell pellets were harvested from 3 liters of LB medium. SDS PAGE was performed to determine the sizes of the eluted proteins. The first elution step eluted proteins with sizes greater than 100, 40, 35, 28, and little less than 14.4 kDa, which was comparable to our construct (Figure 5). The second elution step eluted little protein. The SDS PAGE of the used Ni-NTA resin sample revealed no visible protein traces. The UV spectroscopy of the purified proteins pre-dialysis showed a peak absorbance at 260 nm rather than 280 nm. (Figure not shown).
Dialysis and Refolding:
The SDS PAGE result of the dialyzed sample showed very little protein. The absence of proteins was confirmed by circular dichroism analysis, which showed no absorption difference between left and right polarized light (Figure not shown). Despite the absence of proteins, the dialyzed sample did show an absorbance peak at around 255 nm under UV spectroscopy (Figure not shown.)
There was no sign of protein precipitation on the dialysis bag.
Discussion:
The annealing of the forward and reverse HIV-ep2 oligomers was successful as indicated by the DNA gel (Figure 2). The sizes of the markers should only be used to give a relative measurement since the sample oligomer fragments are lightweight. The DNA gel separates the DNA oligomers by weight; the slower moving band represents therefore the annealed DNA oligomers because its weight is twice that of the monomer.
The pH2_3b plasmid contains an ampicillin-resistance gene. The transformed DH5α E. colicells should therefore able to grow on an agar plate containing ampicillin. The significant higher counts in the group transformed with the novel pH2_3b than in the negative control group suggests that the insertion of the HIV DNA oligomers into the pP_3a vector was successful. Theoretically there should be no count in the negative control group since these bacteria should lack the antibiotic resistance. However some growth is usually observed because not all control plasmids are fully digested and dephosphorylated, thus some may self-ligate and confer ampicilline resistance to the host.
Only 6 out of 13 sampled transformed colonies show the insertion on the gel (Figure 3) , which was expected. Previous experience indicated that the restriction enzymes do not always cut or cut precisely. Many other factors may negatively affect recombination, transformation or the integrity of the insert. However, the result of the DNA sequencing assures the DH5α E. coli contains the correct novel recombinant DNA.
The purification scheme removed much of the unwanted proteins even though some impurities were still found in the elution fractions (see Figure 5). Previous attempts with P_3ahad shown that it binds very tightly on the Ni-NTA resin and is not eluted unless the elution buffer has very high ionic strength. We believe the 14 kDa bands actually represented the H2_3b construct. This result suggests that the purification scheme cleans out most of the impurities, and the constructs come out during the first elution step as expected. However, the 260 nm peak from the UV spectroscopy suggests there might be DNA materials present in the eluted sample.
The dialysis and refolding did not yield folded constructs as expected. The size of the dialysis membrane was carefully examined during the step. We believe the constructs may have precipitated out. The concentration used for the dialysis may need to be lowered to avoid precipitation and improve the yield.
Subsequent experiments with the other HIV constructs revealed that in the first steps of dialysis the protein precipitates into colorless spherical particles of 5 µm diameter which resemble the Chelex 100 Resin beads. However, they can be stained blue using Protein Assay Dye Reagent (Bio-Rad #500-0006) which confirms that they consist of protein. This indicates that the HIV epitope interferes with folding and that the design of the peptides must be revised for this type of epitope.
Conclusion:
We have successfully produced the recombinant DNA plasmid which codes for the de novodesigned nanoparticles carrying an HIV epitope. Gel electrophoresis suggests that the peptide monomers were expressed successfully and that the purification scheme removed most of the impurities as expected. The dialysis step did not yield a satisfactory result and using less-concentrated dialysis bags may improve the refolding outcome.
Acknowledgements
We would like to thank Eric Mann, Lecturer, Department of Microbiology, University of California, Davis; Thomas A. P. Seery, Associate Professor, Department of Chemistry, University of Connecticut; Michelle Ross, Graduate Program Coordinator, Institute of Material Science, University of Connecticut; Kristen McBreairty, Institute of Material Science, University of Connecticut and the Biotechnology facility, University of Connecticut for their help and support. This work was facilitated by a Research Experience for Undergraduates Sponsorship, National Science Foundation, 2005 to George Hwang.
References
Armstrong, S. et al. (2004) 2004 report on the global HIV/AIDS epidemic: 4th global report. Joint United Nations Program on HIV/AIDS (UNAIDS).
Smith, K. A. (2003) The HIV vaccine saga. Medical Immunology 2 (1), 1-1.
Letvin, N. L. (2002) Strategies for an HIV vaccine. Journal of Clinical Investigation 110 (1), 15-20.
Noad, R. and P. Roy (2003) Virus-like particles as immunogens. Trends in Microbiology 11, 438-444.
Bachmann, M. F. and M. R. Dyer (2004) Therapeutic vaccination for chronic diseases: a new class of drugs in sight. Nature Reviews: Drug Discovery, 3, 81-88.
Raman S. K. et al. (2005) Technical Proceedings of the 2005 Nanotechnology Conference and Trade Show, NSTI-Nanotech 2005 1, 47-50
Inoue, H. et al. (1990) High efficiency transformation of Escherichia coli with plasmids. Gene 96 (1), 23-28.