Results and Discussion

EGF Results

Maleimide Assay Results

In the maleimide assay performed, the following liposomes samples were used: Batch 1 liposomes with a concentration of 30 mg/mL, Batch 2 liposomes at 18.9 mg/mL, and EGF conjugated to batch 1 liposomes. The results are as shown:

  Concentration (mM) Uncertainty (mM)
Batch 1 Liposomes 0.9 ± 0.1
EGF + Batch 1 Liposomes 0.6 ± 0.1
Batch 2 Liposomes 0.6 ± 0.1
Batch 1 Liposomes Theoretical 0.58 n/a
Batch 2 Liposomes Theoretical 0.36 n/a

Comparing the results relatively, they are consistent with expected trends. Since the concentration of Batch 2 Liposomes are less than Batch 1 liposomes, the total maleimide concentration detected was lower in Batch 2 liposomes at 0.6 ± 0.1 mM. Both the compositions of the liposomes were the same, with the exception of the liposome size and the concentration, so maleimide concentration is expected to be lower. Additionally, EGF conjugated to Batch 1 liposomes express less maleimide because after the conjugation of EGF to the liposome via thioether bonding of maleimide and sulfhydryl groups. The maleimide is not longer reactive with other thiolated components.

Based on the liposome size and the calculated number of maleimides per liposome, the theoretical concentration of maleimide on Batch 1 liposomes is 0.58 mM. The theoretical concentration of maleimide on Batch 2 liposomes is 0.36 mM. Comparing these values with the experimental results, it shows that the experimental results provide larger than expected values by a minimum of 50% of the theoretical. One of the reasons for this is that due to the addition of TopFluor in the liposomal formulation, the samples exhibited a pink colour, which could potentially influence the absorbance reading during analysis. In the protein assay, a standard curve with BSA, and a separate curve with BSA and liposomes, was performed and the signal of the liposome standard curve exhibited a negative signal increase. Due to the fact that the maleimide assay utilized a standard curve with L-cysteine without the addition of liposomes to match the matrix, the results can only be analyzed relatively to each other.

Thiol Assay Results

Samples of EGF, thiolated EGF with Traut’s reagent, and DSPE-PEG(2000)-SH were analyzed to determine the thiol content and the binding efficiency of the crosslinking reaction with Traut’s reagent. Results from the thiol assay were inconclusive. This is likely due to over incubation of the assay samples. Since the assay performed was not directly meant to be performed as a thiol assay, there are possible intrinsic specification that were not met to be able to reliably analyze our samples. In theory, the methodology of modifying the Maleimide Assay into a Thiol Assay should work, but due to proprietary information in the formation of kit reagents and concentrations, inconclusive results were obtained from this assay.

Protein Assay Results

As aforementioned, the coloured components of the liposome may have contributed to absorbance thus yielding inaccurate results. To verify this hypothesis, a protein standard curve with BSA and a standard curve with BSA and liposomes were prepared. Results show the presence of liposomes increase the absorbance and had a negative signal increase. At low concentrations of proteins, absorbance from the liposome greatly affects the absorbance by proteins at 562 nm. At higher concentrations the total absorbance is slightly higher because of the interference of the liposomes.

  Concentration (ug/mL) Error (ug/mL)
EGF First Batch Stock 310 ±10
EGF Second Batch Stock 340 ±10
Thiolated EGF First Batch 240 ±10
Thiolated EGF Second Batch 140 ±10
EGF + Liposome First Batch 650 ±20
EGF + Liposome Second Batch 600 ±20
  Total Protein Amount (umol) Error (umol)
EGF First Batch 0.0500 ±0.0001
EGF Second Batch 0.0137 ±0.0001
Thiolated EGF First Batch 0.0488 ±0.0001
Thiolated EGF Second Batch 0.00711 ±0.00003
EGF + Liposome First Batch 0.2474 ±0.0002
EGF + Liposome Second Batch 0.0571 ±0.0001

Samples of EGF dissolved in PBS were analyzed to verify that EGF was not lost through dialysis and to confirm the concentration of the EGF stock solutions. In both cases, EGF concentration is lower than the expected values, likely due to reagent loss during sample preparation. In the first batch, there is a minimal (4%) change in the amount of EGF, which confirms that the dialysis tubing was sized properly. A similar analysis can be made for the second batch. The dialyzed EGF+Traut’s sample was smaller in volume, and thus more buffer entered the same size tubing, which likely led to the large (47%) decrease in concentration.

DLS Results

The machine used by this team is a Malvern Zetasizer machine. Samples taken for DLS were diluted by a factor of 500 in 1x PBS pH 7.4.

DLS was primarily used to verify the integrity of liposomal structures after the conjugation and dialysis processes. DLS was also used to confirm the manufacturer’s specification size. From internal DLS sizing of the liposomes purchased from Avanti Polar Lipids, the size for Batch 1 liposomes had a mean size of 157.3 nm. Based on the information obtained from our own sizing results, the mean size was below this value. Based on the figure above, after conjugation with a small molecular weight protein (EGF) as well as with a large molecular weight protein (ApoE), the size of the liposome samples do not very much after conjugation. This is reasonable because the protein moieties are orders of magnitude smaller in size than the liposomes. The PDI of the samples are all less than 0.350, which indicates a small to moderate dispersion. Due to the low variability between samples, the low PDI values also suggests that the experiments did not damage the integrity of the liposomal samples or cause aggregation.

APOE Ligand Results

1. Assay Results

Analysis of hydrazone bond content in DSPE-PEG(2000)-CO(NH2)2 was done through a 2,4,6-Trinitrobenzene Sulfonic Acid Assay (TNBSA). From a standard curve, the concentration of free amino groups was determined to be 11.07 µg/mL. The free amino groups can be related to the terminus end of the hydrazone bond and thus this relatively high concentration of free amino groups confirms correct synthesis of the hydrazone bond was achieved.

Analysis of thiol content in DSPE-PEG(1000)-SH was also conducted using a thiol assay; however, results were inconclusive. This may have been due to the limiting reagent, 2-iminothiolane, preventing complete thiolation of all DSPE-PEG(1000) terminal groups, and thus reducing the signal. However, the highly sensitive TNBSA assay results indicated that there was some degree of thiolation which allowed for the subsequent steps to be carried out correctly. As well, H-NMR analysis also shows successful thiolation. This synthesis process may be improved by increasing the concentration of 2-iminothiolane to allow for all DSPE-PEG(2000) terminal groups to become thiolated, thus making the process more efficient.

2. Analysis of DSPE-PEG(2000)-SH and DSPE-PEG(2000)-CO(NH2)2 Using Nuclear Magnetic Resonance and Dynamic Light Scattering

Analysis using H-NMR indicates success in the synthesis reactions. The NMR spectra show aromatic protons, signifying that DPSE-PEG(1000)-SH was converted to DSPE-PEG(2000)-CO(NH2)2.

Figure 1: H-NMR Spectra of DSPE-PEG(2000)-SH and DSPE-PEG(2000)-CO(NH2)2. Aromatic protons present in DSPE-PEG(2000)-CO(NH2)2 are indicated by the peaks marked in the box and are not present in DSPE-PEG(2000)-SH indicating that crosslinker addition was successful.

Dynamic light scattering (DLS) was performed and yielded a z-average particle size of 1347.67 nm and a polydispersity index (PDI) of 0.937. The large size may be attributed to aggregation of the lipid while the wide polydispersity index due to variability in size of aggregates.

Figure 2: Dynamic light scattering intensity distribution for three samples of DSPE-PEG(2000)-CO(NH2)2 indicate some variance in average particle size.

3. MALDI-MS Acid Cleave Tests

Matrix Assisted Laser Desorption Ionization Mass Spectrometry (MALDI-MS) was used to evaluate the presence of APOE and DSPE-PEG(2000)-CO(NH2)2 and the successful cleavage of the hydrazone bond. MALDI analysis of APOE in 2,5-dihydrozybenzoic acid showed a peak at 34.606 kDa. DSPE-PEG(2000)-CO(NH2)2 analysis showed a great deal of variation due to the polydispersity of the PEG2000, as each of the peaks differs by 40 Da, equivalent to one monomer of PEG.

Results of the acid cleave test were inconclusive due to inability to obtain signal from the samples. Following incubation of the samples in acetic acid at pH 4.0 and 5.0, the samples were filtered using a 10 kDa gel filtration chromatography column and the top and bottom fractions were collected. Upon successful cleavage of the hydrazone bond, the previously conjugated APOE would have been freed and collected in the top fraction while DSPE-PEG would be found in the bottom fraction. Causes of erroneous results or loss of signal may be due to use of different MALDI systems, loss of product during intermediate purification steps, or disintegration of the cellulose membrane due to the acidic buffer. Further studies may be conducted to evaluate successful cleavage of APOE from the system.

4. APOE Human Enzyme Linked Immunoassay (ELISA)

APOE conjugation to DSPE-PEG(2000)-CO(NH2)2 was quantified through an APOE immunoassay (ab108813). The average concentration of APOE present was found to be 0.38 ug/mL. Additionally, the ELISA was used as another measure for success of acid cleave. After incubation of DSPE-PEG(2000)-CO(NH2)2-conjugated APOE in pH 4.0 and pH 5.0 dilute acetic acid, the contents were separated in a 10 kDa centrifuge tube and top fractions (TF) and bottom tractions (BF) were collected. The APOE content in each fraction was quantified. The top fractions contained an average of 0.36 ug/mL while the bottom fractions contained an average of 0.28 ug/mL.

Figure 3: Concentration of APOE determined by ELISA after incubation of DSPE-PEG(2000)-CO(NH2)2-conjugated APOE in acetic acid. After filtration, top fractions collected show a higher amount of APOE than bottom fractions. APOE would be expected to be present in the top fractions while the cleaved DSPE-PEG(2000)-CO(NH2)2 would be expected in the bottom fractions.

The presence of APOE in the bottom fractions may be due to acetic acid damaging the cellulose membrane of the filtration tube. Additional experiments using a more compatible acid, such as lactic acid, may be conducted for improved results.

Cell Culture Results

Analysis of EGF targeting ligands on U251 cellular uptake

This experiment was developed to analyze the effect of EGF targeting ligands on U251 cellular uptake. In addition, the concentration of liposomes added and the time of incubation with cells were titrated in order to find the optimal conditions.

1. Concentration titration

Figure 1. A) Average fluorescence intensity of replicates for each concentration of EGF+ liposomes is presented. The uncertainties are determined using t-values for 5% significance. The large error bars apparent in the lower concentrations of liposomes are representative of the low signal to noise ratio and high background due to decreased cellular uptake. B) Comparison of the signal intensity between liposomes with EGF (EGF+) and liposomes without EGF (EGF-). The concentration of liposomes for the data represented is 100 µM.

Figure 1A shows that 100 µM EGF+ liposomes have the highest average fluorescence intensity, indicating that this is the ideal concentration to use for the U251 cellular uptake experiments. Figure 1B shows that at 100 µM, a considerable amount of liposomes with EGF is endocytosed inside the cells in comparison to liposomes without EGF. This is confirmed in Figure 2, which reflect the increased U251 cellular uptake of liposomes with EGF ligands, represented by the green fluorescence (Top Fluor cholesterol ex/em 495/507). The positive effect of EGF on endocytosis can be analysed statistically using hypothesis testing. The t value for the difference in the means is 8.11 and the critical t value at 5% significance level is 4.303. Since the t value is greater than critical t value, this shows that EGF has a positive effect on the endocytosis of liposomes into cells. However, in order to verify this, a second run was conducted.

Figure 2. Sample photos from wells with treatments of liposomes with EGF (EGF+) and of liposomes without EGF (EGF-). There is an apparent difference in the green fluorescence signal (Top Fluor cholesterol ex/em 495/507) of EGF+liposomes (left) versus EGF- liposomes (right). Hoescht (ex/em 369/457) was used as the blue nuclei stain. The background seen in EGF- photo is due to amplification of background signal by the software. This amplification does not affect quantitative results.

2. Time titration

The purpose of this experiment was to verify the results of the concentration titration experiment, and to determine the optimum time it takes for cellular uptake of EGF containing liposomes.

Figure 2. A) Time titration experiment for endocytosis of liposomes. For all these time points, 100 µM EGF+ liposomes were used. All conditions were run in replicates, except for the two hour condition, where four replicates were used. B) Comparison of effect of EGF on endocytosis of liposomes. For the EGF- condition, four replicates were used and they were incubated for four hours. The high error bars for EGF- liposomes are due to the high background and low signal to noise ratio due to low cellular uptake

Figure 2A shows that signal from the cells is at its highest at 6 hours and it has a drop at 4 hours. The cause of this is unknown. It could potentially be due to measurement error from cellomics and a repeat of the experiment would be needed to verify this result. Figure 3A shows the images of the time titration experiment taken by Cellomics. Figure 2B verifies the results from the concentration titration experiment.The t-test from these results showed a value of 43.87, which is much higher than the critical t value for 5% significance. Figure 3B confirms the results of this visually, where an apparent difference in green fluorescence signal is observed in the EGF+ liposomes.

Figure 3A. Sample photos of the time titration experiment for endocytosis of liposomes. Green fluorescence represents Top Fluor cholesterol (ex/em 495/507), while Hoescht (ex/em 369/467) is a nuclei stain in blue.

Figure 3B. Comparison of images from wells treated with liposomes with EGF (left) and without EGF (right). There is no amplification of signals present in these photos. Green fluorescence represents Top Fluor cholesterol (ex/em 495/507), while Hoescht (ex/em 369/467) is a nuclei stain in blue.

Sizing of liposomes by dynamic light scattering (DLS)

Liposomes only (old batch), liposomes only (new batch), liposomes with HZ-APOE and EGF, and a mixture of liposomes with APOE and liposomes with EGF were sized using a Malvern Zetasizer DLS. The old batch of liposomes represent the first batch of liposomes received, which are suspected to have aggregated over an increased storage time. The new batch of liposomes represent the second batch of liposomes used. Any ligand conjugated liposomes reported in Figure 4 made use of the old batch of liposomes.

Figure 4. Size distribution curves obtained from a Malvern Zetasizer DLS. Three measurements per sample were conducted on A. Old batch of liposomes only, B. New batch of liposomes, C. mixture of liposomes with APOE and liposomes with EGF, and D. liposomes with HZ-APOE + EGF

Figure 4B shows a uniform distribution of the new liposomes sized with DLS. In contrast with this, Figure 4A, 4C and 4D have a disperse and non-uniform distribution, indicating that some liposomes have either aggregated or not all of liposomes were uniformly conjugated with the ligands. These figures also all have a small peak indicating the presence of some smaller sized liposomes. Since liposomes were filtered using a 0.22 µm syringe filter, some liposomes could have been broken down into smaller units during this step. Furthermore, it is hypothesized that these liposomes would be able to go through the pores in the cell layer without the need for transcytosis through the cells.

Analysis of APOE targeting ligands on b.END3 receptor mediated transcytosis

Run 1: The purpose of this experiment was to determine if ApoE ligand facilitates receptor mediated transcytosis through b.END3 cells, and if the hydrazone modification affects the transcytosis in any way.

Figure 5. The t=0 data is from a sample taken from the liposomes before they were added to the cells. The evident difference observed with the initial fluorescence of liposomes with APOE-HZ+EGF is a result of a miscalculation of its initial concentration.

In this experiment, old liposomes without any ligands had the highest fluorescence intensity at both t=8 and 35 minutes, indicating that more liposomes have passed through the membrane filter with b.END3 cells. This is in contrast with the low fluorescence intensity of the new liposomes, indicating that less new liposomes have passed through at t=8 and 35 minutes. However, because the conjugated liposomes were made using the old liposomes and because their size distributions greatly vary (see the homogenous size distribution of Figure 4B versus Figure 4A), these results are incomparable. Furthermore, these results had very high background signals, which largely decreased their reliability. Thus, this run was repeated with conjugated liposomes made from new liposomes.

Run 2: For this experiment, the fluorescence intensity of samples that have passed through the cell culture inserts were measured using a SkanIt Varioskan Flash microplate reader at an excitation/emission wavelength of 495nm/511nm. The samples used were liposomes conjugated with EGF and ApoE, new liposomes conjugated with only ApoE, and new liposomes without any ligands.

Figure 6. Results from transcytosis assay with new liposomes prepared. The t=0 samples were taken from a single source and therefore the uncertainty is unknown. The t=8 and t=35 samples both showed negative numbers relative to the blank sample.

Figure 6 shows a negative fluorescence intensity for all samples incubated for t=8 and t=35 minutes, relative to the blank samples containing only 1% BSA in PBS. In addition to this, there are relatively high error bars observed. This indicates that no conclusion can be drawn from this experiment, and that the transcytosis of liposomes and conjugated liposomes through a BBB cell culture insert model is indeterminate. These results may be due to the unreliability of the fluorescence assay, or the unsuccessful shearing. Since the fluorophore was embedded in the lipid bilayer of liposomes, it was necessary to shear the liposomes by vortexing for more accurate fluorescence analysis.

Full Trojan Bull drug delivery system analysis

This experiment was conducted using the complete BBB cell culture configuration, consisting of b.END3 cells grown on cell culture inserts and U251 cells growing on the bottom of the wells. The purpose of this experiment was to determine if the full system can transcytose across the b.END3 cells grown on cell culture inserts, and to investigate any cellular uptake of U251 cells on liposomes that have passed through the endothelial cell layer (b.END3 cells).

Figure 7. Average fluorescence intensity of Trojan Bull system after passing through the b.END cell culture inserts. The samples labeled with “T” were the full Trojan Bull systems that were incubated using the complete BBB cell culture configuration. A sample with just new liposomes, and liposomes with APOE and EGF were incubated with just U251 cells on 24-well plates (no cell culture inserts included) as a comparison between the two configurations.

Figure 7 shows that there was high U251 cellular uptake for liposomes and liposomes with APOE and EGF, when no b.END3 cell layer was available. This configuration represented a system where no brain endothelial layer (no BBB) was present. A minimal difference between the liposomes with APOE and EGF ligands, and just the liposomes was observed. This shows a somewhat positive effect of the targeting ligands, though it does not differentiate between which targeting ligand, APOE or EGF, contributed to this increased cellular uptake. The low intensity of all conjugated liposomes incubated using the complete BBB cell culture configuration (labeled with “T” in Figure 7) can be attributed to the U251 cells adhesion to the underside of the cell culture inserts. Because of this unwanted adhesion, minimal cells were grown on the bottom of the wells and a negligible number of cellular nuclei was detected by the Cellomics. These images were not reported here. Furthermore, it is uncertain if the transcytosis of liposomes and conjugated liposomes through culture inserts worked successfully (see Figure 6) enough to have been taken up by U251 cells grown on the bottom of each well.