1. Preparation of Liposomes

The liposomes are composed of 1,2-Distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, TopFluor cholesterol, and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000] (DSPE-PEG-maleimide). These liposomes were manufactured from Avanti Polar Lipids, based on a specified composition and certain requirements. Specifically, the liposome formulation consists of 65.8 mol% DSPC, 31.9 mol% cholesterol, 1 mol% TopFluor cholesterol, and 1.3 mol% DSPE-PEG2000-maleimide in a 1x PBS pH 7.4 solution at a concentration of 30 mg of liposomes per mL. The liposomal solution was prepared using the extrusion method through a 0.1 um pore-size membrane yielding a final liposome diameter of 157.3 nm based on dynamic light scattering (DLS). The liposome forms the basis of the drug delivery vesicle, with the DSPE-PEG-maleimide ligand available for conjugation reactions with a highly reactive maleimide group at the distal end. For the formulation of our dual-ligand liposome approach, the DSPE-PEG-maleimide embedded within the liposome forms the basis of the shorter ligand in the system that is exposed after binding and cleavage of the pH-sensitive longer ligand.

Les Liposomes,, 2013. [Online]. Available: [Accessed: 21- Oct- 2016].

In the production of the liposomes, a desired size of 100 nm was not able to be achieved due to a DSPC transition temperature of 55°C, which required significant heat to push through the extruder. Another issue mentioned by Avanti was that, based on the formulation, precipitation of the liposome product occurred. Since the desired diameter of the original batch of liposomes was expected to be lower than 150 nm to be able to pass through the blood brain barrier, a second batch of liposomes was prepared using a micro-emulsion technique instead of a extrusion method to reduce the overall liposome size. The sample prep of the second batch of liposomes was performed at a lower concentration of 18.9 mg/mL and the calculations were adjusted accordingly to ensure similar protein loading amounts per liposome.

2. Conjugation of EGF ligand onto liposomes

A. Reconstitution of EGF

Epidermal growth factor (EGF) was purchased in powdered form from Sigma-Aldrich in a 500ug bottle. A 23.87 uM EGF stock solution was prepared by reconstituting 500ug of EGF 1x phosphate buffered saline (PBS) pH 7.4 supplemented with 5 mM ethylenediaminetetraacetic acid (EDTA). The addition of EDTA prevents the formation of disulfide bonds between two reactive sulfhydryl groups during downstream crosslinking reactions with 2-iminothiolane. The EGF stock solution was aliquoted into 1mL portions and stored at -20C to prevent denaturation and degradation of the protein.

B. Reconstitution of 2-iminothiolane (Traut’s reagent)

Pierce Traut's Reagent (2-iminothiolane) - Thermo Fisher Scientific,, 2016. [Online]. Available: [Accessed: 21- Oct- 2016].

Powdered Traut’s reagent was purchased from Sigma-Aldrich in 1 mg quantities. A 368.41uM stock solution of Traut’s reagent was prepared by resuspending the desired amount of reagent in 5mM EDTA in 1x PBS pH 7.4. Traut’s reagent is a small thiolation compound with a length of 8.1 angstroms, that reacts with primary amines and functionalizes the target with a free sulfhydryl group.

C. Thiolation of EGF

EGF was thiolated using Traut’s reagent to provide a sulfhydryl group that could be conjugated onto a reactive maleimide group on the distal end of the ligand on the liposome1. The thiolation reaction uses crosslinker chemistry to react primary amines on the amino acid sequence and functionalize the protein with a free sulfhydryl group. To perform the thiolation, previously prepared stock solutions of EGF and Traut’s reagent in PBS were thawed via centrifugation. Complete reaction was performed with a 4x molar excess of Traut’s reagent to 1 mL of EGF stock solution under room temperature conditions. The thiolation reaction was incubated for one hour, which ensures that all EGF molecules are sufficiently modified with at least 3-7 sulfhydryl groups, based on the manufacture protocol2.

D. Dialysis of Thiolated EGF

Due to the molar concentrations of Traut’s reagent selected for the reaction, the removal of the unbound reagent is necessary to both stop the reaction and purify the EGF before further processing3. To remove the Traut’s reagent from solution, a buffer exchange was achieved through dialysis against a PBS solution. Based on the molecular weight (MW) of EGF, 6.2 kilodaltons (kDa), and the MW of Traut’s reagent, 137.63 Da, the molecular weight cut off (MWCO) of the dialysis tubing is required to be lower than our desired retained protein to ensure a size exclusion separation. For separation using size exclusion and selection of a pore size, a general rule of thumb used is “a MWCO that is half the size of the MW of the species to be retained and at least 10 times larger than twice the size of the MW of the species intended to pass through”4. For separation of EGF from Traut’s reagent, a dialysis tubing with a MWCO of 1000 Da was sufficient and satisfied this heuristic.

Removal efficiency of Traut’s reagent was improved through the implementation of two buffer replacements of the dialysate. The dialysis procedure was performed using a 300x volume excess for 2 hours, a dialysate buffer replacement, followed by an additional 300x volume excess processed overnight. Under standard conditions, a 100x volume excess is used, but to ensure complete submersion of the sample in solution, a 300x volume excess provided better surface coverage. After dialysis, a sample was taken for analysis of protein concentration and thiol content to quantify losses in the process as well as confirm the amount of sulfhydryl reactive groups on the surface of the EGF, respectively.

E. Thioether bonding of Thiolated EGF with Liposome

The attachment of the thiolated EGF unto the liposome ligand was performed via a thioether conjugation reaction between the sulfhydryl functionalized EGF to the reactive maleimide groups on the distal end of the DSPE-PEG2000 chain length. Thiolated EGF was combined with the 30 mg/mL stock liposome solution in a molar ratio calculated based on 3.5 mol% of EGF on the total available maleimide ligands. With this formulation, each liposome contains approximately 200 molecule of EGF attached to the ligands, with the maleimide ligands in excess. Due to the light sensitivity of the TopFluor cholesterol in the liposomes to photobleaching, the tubes were incubated at room temperature overnight, covered in foil away from light.

F. Dialysis of Liposomes

To separate unbound EGF from the EGF-Liposome conjugates, dialysis was performed. Based on the previously mentioned heuristic, a tubing size of 300 kDa was selected to provide sufficient removal of EGF (6.2 kDa * 2 * 10 = 124 kDa minimum MWCO). The upper limit of the tubing size was determined by calculating the MW of the liposome based on its composition, which was determined to have a MW of 2.96*108 kDa (refer to calculations). The dialysis procedure was performed using a 300x volume excess for 2 hours, a dialysate buffer replacement, followed by an additional 300x volume excess processed overnight. After dialysis, a sample was taken for protein analysis to confirm presence of EGF on liposomes and a maleimide assay was performed to determine the total amount of free maleimide ligands that remain unreacted.

G. Collecting finished product

After dialysis, our final product, EGF-conjugated liposomes was recovered. The production of EGF conjugated onto the liposome is an intermediary step towards making the dual ligand liposome with both EGF and a pH sensitive APOE ligand. The samples collected were analyzed using maleimide, thiol, and protein assays to characterize the samples. Maleimide assays were used to verify the presence of maleimide on the liposomes, thiol assays were used to determine actual vs. theoretical thiolation of EGF on liposomes, and protein assays were used to quantify the amount of EGF. The remaining sample of the EGF-conjugated liposomes were handed over to the cell culture team to run endocytosis tests using u251 glioblastoma cells.

H. Quenching the Thioether Reaction and Capping Reactive Maleimide Groups

In theory, after the thioether conjugation reaction between the sulfhydryl functionalized EGF and the maleimide ligand on the liposome, free reactive maleimide groups are still present on the liposome. For in vivo applications, the free reactive maleimide groups have the potential to react with any component containing a sulfhydryl group, which must be avoided. To prevent this from occurring, the reactive maleimide groups must be capped with a non-reactive termination. For our liposomes, the reactive maleimide groups were capped with 2-mercaptoethanol (BME). The reaction was performed with 14.26 M stock BME solution and added to the liposomes at a 100x molar excess. The sulfhydryl group on the BME reacts with the maleimide groups on the liposome and provides a less reactive hydroxyl termination. Due to the toxicity of BME to cells, a buffer exchange via dialysis is performed to remove it from the final product.

3. Dynamic Light Scattering Methodology

Dynamic Light Scattering (DLS) is an analytical method used to determine the size of particles suspended in solution. By shining a laser through a sample of particles, DLS utilizes Rayleigh scattering and the Brownian motion to receive constructive and deconstructive interference to size the particle. Based on the calculation, irregular shaped particles are analyzed and approximated by a sphere. The Z-Average Diameter (ZAvg) is the intensity weighted mean hydrodynamic size, which best approximates the mean diameter of the suspended particles. The Polydispersity Index (PDI) is a measure of the distribution of particle size, and varies between 0 and 1, with lower values corresponding to a sample closer to monodispersity 5.

4. Assays

A. Maleimide Assay Methodology

Maleimide assays were performed to quantify the relative amounts of maleimide on liposomal samples (batch liposomes and liposomes conjugated with EGF). In typical measurements of maleimide components via spectrophotometric analysis at 302 nm, the results are insensitive due to the small extinction coefficient of 620 M-1 cm-1. The analysis at this wavelength also is complicated by protein absorbance, which makes measurement of maleimide components difficult. The Colorimetric Maleimide Assay (Catalog Number: MAK162) from Sigma-Aldrich was used and provides a measurement of the maleimide groups by reacting the maleimide samples with a known amount of thiol in the form of 2-Aminoethanethiol Hydrochloride (MEA). The MEA solution added was performed in excess to both quantify and detect the excess amounts of thiol in the reaction with 4,4 -Dithiodipyridine (DTDP). The determination through this indirect measurements provides more reliable results due to the larger DTDP extinction coefficient of 19,800 M-1 cm-1 at 324 nm. In order to verify calculated values using Beer’s Law, a thiol standard curve was constructed using L-cystine. The reaction of the maleimide samples with the MEA solution forms a thioether bond and must be incubated for 20 minutes to ensure complete reaction. Due to the diverse preparation and buffer matrix of the samples, the readings were normalized by measuring the absorbance prior to incubation with DTDP and after addition of 10 uL and 2 minute incubation of DTDP. The implementation of the procedure ensures high noise background is eliminated and that the reading is representative of only the reacted DTDP thiol samples. The incubation with DTDP was perform

B. Thiol Assay Methodology

Since the Maleimide Assay from Sigma Aldrich indirectly measures maleimide content by reacting with a known about of thiol, the detection of thiol samples in the kit can be utilized and taken advantage of. For the thiol assay protocol, the desired result is to analyse the thiolated samples of proteins and lipids as a result from the crosslinking reaction from Traut’s reagent. Since the thiol samples already contain sulfhydryl reactive groups, the samples were introduced into the Maleimide Assay kit prior to the addition of DTDP. Similar to Maleimide Assay, DTDP is reacted with thiol, and the resulting absorbance is measured at 324 nm. An initial reading before adding DTDP and after incubation with DTDP was taken to normalize the readings after adding DTDP.

C. Protein Assay Methodology

Protein assays were performed to quantify the amount of EGF and APOE in samples and conjugated onto the liposome. The assay chosen for the protein determination was a bicinchoninic acid-based (BCA) protein assay due to its compatibility with samples that contain small quantities of surfactants. The kit used was the Micro BSA Protein Assay Kit (Catalog Number: 23235) from ThermoFisher Scientific.

In the BCA assay, the protein bonds present in protein samples reduce Cu2+ ions from copper (II) sulfate to Cu1+. From the reduction reaction, Cu1+ exhibits a blue colorimetric response, and the concentration of Cu1+ can be detected through a chelation with two molecules of bicinchoninic acid, which forms a purple-coloured complex. Since the concentration of protein is directly proportional to the concentration of detected copper ions, a standard curve is constructed using bovine serum albumin (BSA).

Due to colouring components present in our liposome samples, a second standard curve was constructed with BSA but small doses of liposomes were added to determine if the addition of liposomes increased or decreased the absorbance signal of the standard curve. Samples containing proteins were reacted with the working reagent (containing copper (II) sulfate and bicinchoninic acid), incubated for 30 minutes, and the absorbance was measured using the spectrophotometer at a wavelength of 562 nm.

##D. General Assay Remarks Both the protein and maleimide assays used take advantage of colourimetric properties for detection of component concentrations. The concentration of components in the sample can be related from absorbance values and a extinction coefficient by Beer’s Law. For all of the assays, a standard curve is generated to obtain known concentration values and the associated absorbance at each of these known concentrations. This provides an additional correlation between absorbance and component concentration that is necessary for colourimetric determination.

5. Conjugation of APOE ligand onto liposomes

A. Construction of DSPE-PEG(2000)-HZ-APOE

1,2-Distearoyl-sn-glycero-3-phosphoethanolamine (DSPE) - Polyethylene Glycol (2000) - Amine provided by Avanti Polar Lipids was used as the block copolymer for APOE ligand attachment. 25 mg of the amphiphilic copolymer was reconstituted in chloroform to a final concentration of 17.9 mM (50 mg/mL), and a stock solution of 60.5 mM thiolation reagent, 2-iminothiolane, in methanol was prepared. 2-iminothiolane was added to the DSPE-PEG(2000)-Amine chloroform solution to react with the primary amine attached to the distal end of the PEG chain. This results in the addition of a free sulfhydryl group available for further crosslinking reactions. Additional methanol was added to bring the final concentration of DSPE-PEG-amine and 2-iminothiolane to 4.32 mM and 8 mM, respectively. The reaction was stirred in the fume hood for 2 hours at room temperature until chloroform and methanol was evaporated completely to a dried powder form. After drying, the powder was dissolved in PBS and dialyzed with a 1 kDa MWCO membrane to remove excess 2-iminothiolane. The resulting DSPE-PEG2000-SH dialysate was freeze dried and analyzed using H-NMR using Varian 500 mHz and a 4,4′-Dithiodipyridine thiol assay.

B. Synthesis of DSPE-PEG(2000)-CO(NH2)2 from DSPE-PEG(2000)-SH6

4-(4-N-maleimidophenyl)butyric acid hydrazide (MPBH) crosslinker was added to DSPE-PEG2000-SH in a 40X molar excess of crosslinker at room temperature for 4 hours. This allowed the free sulfhydryl group on the copolymer to form a thioether bond with the maleimide group on the MPBH cross linker, thus forming DSPE-PEG(2000)-CO(NH2)2. CO(NH2)2 is a carbohydrazine group that reacts with reactive aldehyde groups to form an acid-labile hydrazone (HZ) bond, an important component in our design. Excess and unreacted crosslinkers were dialyzed through a 1 kDa MWCO membrane, and the final product was sized using dynamic light scattering (DLS). A 2,4,6-Trinitrobenzene Sulfonic Acid Picrylsulfonic Acid Assay (TNBSA) was performed to quantify the amount of hydrazine groups attached.

C. Carbohydrazide Verification using TNBS 2,4,6-Trinitrobenzene Sulfonic Acid Picrylsulfonic Acid Assay

In order to quantify amino groups and ensure correct synthesis, a 2,4,6-Trinitrobenzene Sulfonic Acid Picrylsulfonic Acid colorimetric assay (TNB) was conducted. TNBSA is a well-established assay7 for quantifying free primary amine (NH2) containing molecules, such as our DSPE-PEG(2000)-CO(NH2)2 lipid linker. 0.1M Sodium Bicarbonate was added to DSPE-PEG(2000)-CO(NH2)2 samples and allowed to react with picrylsulfonic acid solution (Sigma Aldrich No. P2297) at 37ºC for two hours. To generate a standard curve for free primary amine quantification, known concentrations of glycine in 0.1M sodium bicarbonate (0mg/mL -20 mg/mL) were similarly prepared and incubated. After incubation, sodium dodecyl sulfate (SDS) solution and hydrochloric acid (HCl) were added to all the incubated samples. 250 µL each of the standards, samples and blanks (containing sodium bicarbonate, picrylsulfonic acid, SDS and HCl) were added in duplicate to a flat-bottomed 96-well plate, and read at 335 nm using a SkanIt Varioskan Flash microplate reader.

D. Periodate Oxidation of APOE to form reactive aldehyde

For successful conjugation of APOE to the ligand, oxidation of a sialic acid moiety on the APOE must occur in order to create a reactive aldehyde group available for linkage with the hydrazine group (NH2)2 of the DSPE-PEG-CONHNH2 lipid ligand. Sodium metaperiodate (20 mM, Sigma Aldrich No. 311448) was added to human serum purified APOE suspended in sodium acetate oxidation buffer. Sodium borohydride (40 mM) was used to quench the reaction and reduce any remaining free aldehyde groups. The final oxidized APOE was purified using gel filtration chromatography (10kDa cellulose membranes).

E. Acid-cleave test on DSPE-PEG(2000)-HZ-APOE

APOE-conjugated lipid ligand product was incubated for 1 hour at room temperature in acetic acid at pH 4.0 and 5.0. After incubation, the product was separated by gel filtration chromatography using a 10 kDa column. To ensure maximum recovery of the product, the top fraction is reverse centrifuged and collected in a new tube immediately. This top fraction, as well as the bottom fractions were collected and analyzed using Matrix Assisted Laser Desorption Ionization Mass Spectrometry (MALDI-MS). This technique involves ionization and fragmentation of molecules by a laser, the fragments are separated by their unique mass to charge (m/z) ratios and thus can be identified. Unconjugated DSPE-PEG(2000)-CO(NH2)2 lipid and purified APOE only were used as controls.

Matrix Assisted Laser Desorption Ionization Mass Spectrometry (MALDI-MS) involves ionization and fragmentation of molecules by a laser the fragments are separated by their unique mass to charge (m/z) ratios and thus can be identified.8

F. H-NMR Analysis Using Varian 500 mHz spectroscopy

Proton Nuclear Magnetic Resonance (H-NMR) is a spectroscopic technique that uses the magnetic properties of atomic nuclei to determine the structure of molecules. H-NMR analysis was done on DSPE-PEG(2000)-SH and DSPE-PEG(2000)-CO(NH2)2 samples.

G. APOE Enzyme Linked Immunoassay (ELISA)

A sandwich ELISA kit (ab108813) was used to verify the successful conjugation of APOE to DSPE-PEG(2000)-CO(NH2)2 and quantify the amount in samples before and after incubation in acid. A standard ELISA protocol was adapted from Abcam.

H. Micelle Transfer Method

To increase specificity, the second ligand was formulated separate from the liposomes containing the primary ligand. This second ligand must be incorporated into the liposome. Incorporation of micelles into liposomes was used to transfer lipid ligands to the liposomes. This involved stirring and mixing micellar lipids to liposomal lipids at 60 degrees C. This increase in temperature causes an increase in kinetic energy that allows for lipids to incorporate into the outer portion of the liposomal bilayer.

6. Process Controls

The dual ligand pH-acid sensitive cleave liposome is an extremely modular system that can be modified and varied based on the desired target of choice and the applications. The novelty in this system is that the component conjugated onto the distal end of each ligand can be either a protein, antibody, aptamer, or other biological component with a high specific binding affinity to a desired target. Because of the vast applications and various modifications that can be performed on this dual-ligand acid-sensitive liposome, controls are implement to provide reproducible results within the same system and maintain consistency between the different systems.

To characterize the final product, important parameters that must be considered and developed for a final acceptance criteria are liposome size, protein concentration, maleimide content on the liposomes, amine content remaining on the lipids, and cell culture performance testing. Liposome size can be determined by dynamic light scattering (DLS). Colorimetric assay approaches can verified through protein concentration by BCA assay; maleimide content can be calculated by reacting with a known amount of thiol and quantified by 4,4’-dithiodipyridine (DTDP); and amine content can be determined using the 2,4,6-trinitrobenzene sulfonic acid assay (TNBSA). In addition to the characterization testing, performance testing with U-251 and bEND.3 cells and fluorescent imaging will ensure successful formulation of the dual liposome.

Liposome Reference Values:

  Liposome Batch #1 Liposome Batch #2
Liposome Concentration (mg/mL) 30 18.9
Production Method Extrusion Micro-emulsification
Buffer 1x PBS pH 7.4 1x PBS pH 7.4
Liposome Size (nm) 157.3 100
MW Liposome (g/mol) 2.955 x 108 1.154 x 108
Moles/mL 1.015 x 10-10 1.638 x 10-10
Liposomes/mL 6.113 x 1013 9.866 x 1013

Liposome Composition Values:

  DSPC Cholesterol DSPE-PEG-maleimide TopFluor Cholesterol
MW (g/mol) 790.145 386.654 2981.605 576.406
Liposome Composition (mol%) 65.8 31.9 1.3 1.0
Number of Lipids/Molecules in Batch #1 283537 137459 5671 4310
Number of Lipids/Molecules in Batch #2 110683 53659 2214 1682

EGF and APOE Ligand Target Values:

  EGF Ligand APOE Ligand
Protein Concentration 23.87 uM EGF in 1x PBS pH 7.4 50 ug APOE
Desired Number of Proteins per Liposome >200 (equivalent to ~3.5 mol% of ligand for Batch #1) n/a
2-iminothiolane Stock Solution 368.41 mM (4x molar excess) 60.56 mM
Thiolation Reaction Time 1 hr at RT 2 hrs at RT
Dialysis Tubing Pore Size 1000 MWCO 1000 MWCO
Dialysis Wash Time Overnight (15+ hrs) Overnight (15+ hrs)
Dialysis Wash Volume ~400mL ~400mL
Liposome Concentration 30 mg/mL Liposome 50mg/mL DSPE-PEG(2000)-amine
EGF to Liposome Molar Ratio ~4 mol% n/a
Thioether Reaction Time 1hr at room temperature n/a
Concentration of MPBH n/a 10 mM
MPBH Reaction Time with DSPE-PEG-SH n/a 4 hrs
Na-Periodate Concentration: n/a 20mM in oxidation buffer
Reaction Time with Na-Periodate n/a 10 min
Concentration and Volume of B-mercaptoethanol (EGF)/ Sodium Borohydride (APOE) 14.21 M, 86 uL (150x molar excess) 40mM, 5uL

Micelle Transfer Method Operating Conditions:

  Micelle Transfer Method
Incubation Time 60 mins
Incubation Temperature 60°C
Concentrations 1:30 (APOE:Liposome)

7. Cell culture

b.End3 Cells as the BBB Model

To test the viability of the designed drug delivery system, the constituents of the blood brain barrier is studied in vitro.The BBB model is established by growing a monolayer of endothelial cells on a porous membrane, which is submerged in the wells of a multi-welled plate.

An immortalized mouse brain endothelial cell line, b.End3, is chosen for this purpose because of the properties it shares with the BBB. b.End3 cells are an attractive candidate as a model of the blood-brain barrier due to their rapid growth, maintenance of blood-brain barrier characteristics over repeated passages, formation of functional barriers and amenability to numerous molecular interventions8.

U251 Cells as the Glioblastoma Model

U-251 cells are used as a glioblastoma model very commonly. Their original source was from the glioma of a male patient with malignant astrocytoma9, which is a very similar disease to glioblastoma.10

Cell Culture Maintenance

U-251 Cells:

U-251 cells are kindly provided by Dr. Naus, the University of British Columbia, Canada. U-251 cells are cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich) Cells (early passages) were maintained in a humidified cell culture Incubator at 37 °C and with 5% CO2/95% air.

b.End3 Cells:

b.End3 cells are kindly provided by Dr. Naus, the University of British Columbia, Canada. b.End3 cells are cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich) Cells (late passages) were maintained in a humidified cell culture Incubator at 37 °C and with 5% CO2/95% air.

Analysis of EGF targeting ligands on U251 cellular uptake

For all experiments, 70,000 U251 cells were seeded in each well and left overnight to adhere before experiments were conducted. A 2 hour incubation time was used for concentration titration experiments as well as experiments that compared the effect of EGF on endocytosis. Fixation of cells was achieved through incubating the cells with 500 ng/mL of Hoechst nuclear stain in 4% paraformaldehyde for 10 minutes at 37°C.
Plates were imaged using a Cellomics Arrayscan VTI HCS reader. 25 pictures spanning the middle area of each well were taken. A high-content screening (HCS) analysis was performed after imaging, which identified a ring area around each cell nucleus and calculated the average fluorescence intensity for this area per well.

Figure 1. Sample raw photo and the HCS analysis of the same photo.

For each well, cellomics reported the mean of the average fluorescence intensity per ring area. The average between each well for a single condition was used in our data analysis.

  1. Concentration Titration
    • 0.1 µM, 1µM, 10µM and 100µM concentrations of liposomes and liposomes with EGF (concentrations in terms of liposome phospholipid content) were prepared in the same media used to grow the cells, and filtered using 0.4µm syringe filters. 500 µL of each of these concentrations were added to cells in replicates.
  2. Time Titration
    • The time points used for this experiment were t=0.5 hr, t=2 hr, t=4 hr and t=6 hr. 100 µM liposomes and liposomes with EGF were prepared in the same media as cells. For each of the time points, 500 µL of liposomes with EGF was added to cells in duplicates except for t=2hr, where four replicates of liposomes with EGF and four replicates of liposomes were used.

Transcytosis Assay

In preparation for all experiments, 150,000 b.END3 cells were allowed to grow on 0.4µm PET membrane cell culture inserts for two days. The media above and below the transwells was aspirated and the media below was replaced with 600 µL of 5 µg/mL BSA in PBS mixture. 100 µM liposomes were prepared in cell media and syringe filtered and added to the transwells to start the experiments. The plate was allowed to incubate, and after 8 minutes, the PBS/BSA mixture from the bottom of the plate was sampled and replaced with fresh PBS/BSA mixture. After 35 minutes, samples were taken from the PBS/BSA mixture and the experiment was terminated. 8 mins and 35 minutes were chosen in order to differentiate between the time it takes for our system to transcytose, and the time it takes for any smaller liposomes to go through the cell membrane. Based on Wagner, Zensi, Tschickardt et al11, the transcytosis process takes 10-15 minutes, while going through the pores would happen faster.
To find the amount of liposomes transcytosed, the sample’s fluorescence was measured using a SkanIt Varioskan Flash microplate reader at an excitation/emission wavelength of 495nm/511nm corresponding to the embedded cholesterol fluorophore.

Run 1

The conditions used in this run were: Batch 1 liposomes (old liposomes), liposomes with ApoE and EGF (due to experimental error this was a mixture of liposomes with ApoE and liposomes with EGF), liposomes with ApoE bonded through a hydrazone bond and with EGF (liposomes+ApoE-HZ+EGF) and Batch 2 liposomes (new liposomes).

Run 2

The conditions used in this run were: Batch 2 liposomes conjugated with ApoE and EGF (liposomes+ApoE+EGF), batch 2 liposomes with ApoE (liposomes+ApoE) and batch 2 liposomes.

Whole Cell System

In preparation of this experiment, 150,000 b.END3 cells were seeded onto 6 transwell filters on day 0. On day 1, 70,000 U251 cells were seeded under the transwells and in 4 extra wells. On day 2, 100 µM liposomes+ApoE+EGF, liposomes+ApoE and liposomes were prepared in media. For the wells with transwells, the media below them was replaced with 600 µL of fresh media and the media above them was replaced with 500 µL of the corresponding liposomes. For the wells without transwells, their media was replaced with 500 µL of the corresponding liposomes. All conditions were conducted in duplicates. After a three hour incubation period at 37°C, the transwells were removed and the cells were fixed with 500 ng/mL of DAPI nuclear stain in 4% paraformaldehyde for 10 minutes at 37°C. The fluorescence intensity of the plate was analyzed with a Cellomics ArrayScan HCS VTI.

  1. E. Bohl Kullberg, N. Bergstrand, J. Carlsson, K. Edwards, M. Johnsson, S. Sjöberg and L. Gedda, “Development of EGF-Conjugated Liposomes for Targeted Delivery of Boronated DNA-Binding Agents”, Bioconjugate Chem., vol. 13, no. 4, pp. 737-743, 2002. 

  2. “Traut’s Reagent”, Thermo Fisher, 2016. [Online]. Available: [Accessed: 19- Oct- 2016]. 

  3. E. Mamot C, “Epidermal growth factor receptor (EGFR)-targeted immunoliposomes mediate specific and efficient drug delivery to EGFR- and EGFRvIII-overexpressing … - PubMed - NCBI”,, 2016. [Online]. Available: [Accessed: 30- Sep- 2016]. 

  4. “Fundamental of Dialysis”,, 2016. [Online]. Available: [Accessed: 19- Oct- 2016]. 

  5. R. Shaw. (nd). Dynamic Light Scattering Training Achieving reliable nano particle sizing [Online]. Available: 

  6. S. Biswas, N. Dodwadkar, R. Sawant and V. Torchilin, “Development of the Novel PEG-PE-Based Polymer for the Reversible Attachment of Specific Ligands to Liposomes: Synthesis and in Vitro Characterization”, Bioconjugate Chem., vol. 22, no. 10, pp. 2005-2013, 2011. 

  7. Hermanson, G. (1996). Bioconjugate Techniques, p.112-113. Academic Press, San Diego, California. This book is available from Pierce as Prod. No. 20002. 

  8. R. C. Brown, A. P. Morris, and R. G. O’neil, “Tight junction protein expression and barrier properties of immortalized mouse brain microvessel endothelial cells,” Brain Research, vol. 1130, pp. 17–30, 2007. 

  9. A. Torsvik, D. Stieber, P. Ø. Enger, A. Golebiewska, A. Molven, A. Svendsen, B. Westermark, S. P. Niclou, T. K. Olsen, M. C. Enger, and R. Bjerkvig, “U-251 revisited: genetic drift and phenotypic consequences of long-term cultures of glioblastoma cells,” Cancer Medicine, vol. 3, no. 4, pp. 812–824, Aug. 2014. 

  10. “Glioblastoma and Malignant Astrocytoma,” Glioblastoma and Malignant Astrocytoma. American Brain Tumor Association, Chicago, Illinois. 

  11. S. Wagner, A. Zensi, S. Wien, S. Tschickardt, W. Maier, T. Vogel, F. Worek, C. Pietrzik, J. Kreuter and H. von Briesen, “Uptake Mechanism of ApoE-Modified Nanoparticles on Brain Capillary Endothelial Cells as a Blood-Brain Barrier Model”, PLoS ONE, vol. 7, no. 3, p. e32568, 2012. 

  12. A. Clark, E. Kaleta, A. Arora and D. Wolk, “Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry: a Fundamental Shift in the Routine Practice of Clinical Microbiology”,, 2016. [Online]. Available: [Accessed: 22- Oct- 2016]. 

  13. H. Gan, A. Cvrljevic and T. Johns, “The epidermal growth factor receptor variant III (EGFRvIII): where wild things are altered”, FEBS J, vol. 280, no. 21, pp. 5350-5370, 2013.