Method of Targeting Circulating Tumor Cells

Cancer metastasis accounts for 90% for all cancer-related deaths. Thus, treatment and prevention of secondary tumor formation is vital. A liposome coated with an adhesion molecule and death ligand is proposed in order to target the cancer cells in the bloodstream.


Results & Discussions


Circulating tumor cells (CTCs) and cancer stem cells (CSCs) are cells that shed from a tumor mass and circulate in peripheral blood. These cells have been recognized as the origin of metastasis in epithelial and non-hematologic cancers1,2, such as breast, prostate, and colorectal cancer. Properties have recently been discovered through the isolation and purification of CTCs, including its adhesion to E-selectin and anti-EpCAM (anti-Epithelial Cell Adhesion Molecule)3. When the CTCs travel in the bloodstream during metastasis, it must have the capability of attaching to the endothelium of the vessel in order to penetrate the body tissue. The known adhesion molecules on the endothelial surface are termed selectin. After preliminary experimentation in this project, it was noted that E-selectin was the only of these ligands that possessed an adhesion to CTCs. Anti-EpCAM has previously been used in literature for the capture and isolation of CTCs from the whole blood. The cancer’s ability to take advantage of blood and immune cells to further the metastatic progression was noted, as well. Current research has been directed to the complete elimination and killing of CTCs withminor rebounding effect 4, 5, 6. However, the cell’s rarity and speed makes it very difficult to target and eradicate. No current techniques have yielded clinical potential. Therefore, improvement of CTC capture and killing is a pressing issue. However, complete removal of all circulating tumor cells is not always necessary. Immune cells, such as macrophages and neutrophils, are able to phagocytize approximately < 5% of all CTCs. 

In new studies, CTC clusters, or the fused interaction of multiple cancer cells, are believed to have a higher potential of surviving in the bloodstream during the course of metastasis than individual circulating tumor cells or cancer stem cells7. Thus, these clusters’ properties need to be fully understood in order to target them. Furthermore, in terms of achieving cancer death, the tumor necrosis factor related apoptosis-inducing ligand (TRAIL) has been discovered to induce apoptosis, or programmed cell death, in tumor cells by binding to certain death receptors8,9. A remarkable finding concerning TRAIL was its ability to non-specifically interact with leukocytes’ receptor. This nonspecific binding ensured no apoptosis would occur in normal blood cells, eliminating any rebounding effect10. Subsequently, a surge in TRAIL research began. However, no current treatments or tests have been clinically proven to effectively reduce widespread cancer cell populations in their home environment.

TRAIL alone cannot pose significant advancements in metastatic prevention, as attempted in previous literature. In order to achieve cellular death by TRAIL, three aims have been proposed:

By conjugating the TRAIL protein with an adhesion molecule such as E-selectin, a microfluidic chip has been proposed. In order to allow for this conjugation, a vector must be used. Plasmids and liposomes are potential carriers. Plasmids, however, are targeted by the immune system due to its foreign appearance, thereby reducing its lifetime. Liposomes are composed solely of lipids; none of which are foreign to the body.

The first test to assess the efficiency of this technique was measuring the bonding force between the liposome, leukocytes, and circulating tumor cells. By testing the strength of this interaction, an extremely efficient method can be used to kill CTCs in the blood without harming other cells/structures. The liposome consists of the E-selectin on one end and the TRAIL protein on the other. The liposome will first attach to either the endothelium or leukocytes in the blood. The attachment will occur due to the E-selectin in the liposome. The adhesion force between the E-selectin and leukocyte/endothelium was tested and mimicked in a microfluidic device to determine its efficiency in the first step of the process of killing CTCs. Based on these preliminary results, a more effective method can be composed by adding different adhesion molecules to the liposome instead of E-selectin, such as P-selectin or anti-EpCAM. In order to quantify this measurement and preliminary trial, one group of cells needs to be monitored, or fluorescent dyes need to be released when the bonds break, indicating the maximum limit of the bond’s force.

The second process involves the interaction between the endothelium/leukocyte and CTC. Leukocytes are used because they are plenty in number. Therefore, there is a very high probability that they will come in contact with CTCs due to the laminar flow of blood. Endothelium is used because in order to cause metastasis, CTCs need to adhere to the endothelium at the site of common secondary tumor formation; therefore, if the endothelium contains the TRAIL ligand, many CTCs that try to metastasize will be killed. In this process, a microfluidic device will house the endothelium with liposomes attached, or leukocytes with liposomes attached. CTCs will then flow, at different rates, through the device. The device may contain some geometrical parameters to mimic various conditions of blood vessels. Analysis will include measurements of efficiency, showing how many CTCs were tagged with the liposome and attached to either the leukocyte or endothelium, measurements of bond force, showing the efficiency in terms of the adhesion strength for individual cells based on the shear stress, measurements of how long CTCs will live once attached to the liposome, and measurements of leukocyte or red blood cell deaths, showing the extent of its harm on the body. Based on these analyses, improvements can be made in the future to ensure maximum efficiency in all circumstances and cases. Specifics can also be made for each type of cancer during treatment (i.e. breast cancer, prostate cancer, colorectal cancer, etc.).  

The design of concept for experimental validation has been pursued to measure the efficiency of this method. The last aim is achieved through this test, where there will be validation of TRAIL’s and E-selectin’s (and anti-EpCAM’s) actions. The TRAIL protein was tested to ensure it induces apoptosis in CTCs. The TRAIL and E-selectin proteins were conjugated to create a liposome. Additionally, preliminary testing was performed to confirm a bond formation between leukocytes and liposome.


Preparation of Liposomes

Detergent removal ensures the spontaneous formation of unilamellar liposomes with sporadic,infrequent development of multilamellar liposomes. For membrane protein incorporation, no previous precedents could be referenced for this particular situation. Different synthetic vesicle carriers could be created using varying concentrations of components. Three components used to generate the backbone structure of the liposome were cholesterol (lyophilized), phosphatidylglycerol (PG), and L alpha phosphatidylcholine (PC).PG is used because it allows the liposome to become negatively charged. This is needed for the Atomic Force Microscopy because a negatively charged liposome needs to attach to a positively charged Poly-L-lysine coated cantilever.PC is used because its transition temperature is -15 C, meaning that it can be manipulated at room temperature. It is also a vital and abundant phospholipid used in many instances of liposome preparation.Cholesterol is necessary, as well. The presence of cholesterol exerts a profound influence on the properties of the lipid bilayers of the liposomes. It has been known for several decades that the addition of cholesterol to a fluid filled bilayer (mainly unsaturated lipids) decreases its permeability to water. A liposome that is made from 100% unsaturated lipid in fluid phase cannot hold its encapsulated or incorporated content and the proteins will leak out over time. Therefore, the addition of cholesterol is necessary in order to prevent the destruction of the incorporated membrane protein from the liposomes. 

The detergent chosen to dissolve and support the phospholipids was n-octylglucoside (OG). N-octylglucoside possesses a relatively high critical micellar concentration (CMC), at 25 mM. This allows it to have a better efficiency for detergent removal from the mixed micelles thus resulting in vesicles less contaminated by the residual detergent. In terms of the mechanism of detergent removal, the addition of all three phospholipids with OG with slight (< 180 rpm) vortexing resulted in mixed micelles, as shown in Figure 1 (11). In micelles, hydrophobic tails occupy the encapsulated center, whereas hydrophilic heads rest on the outer layer. This implies that oil-based substances are the only molecules that can be encapsulated in the center and among the tails (hydrophobic repels water and attracts oil). Many membrane proteins are aqueous-based. Two problems thus arise: [1] the center needs to be aqueous, so hydrophilic heads must be on both the outer and inner layer, and [2] the micelle is too small (< 75 nm ø). Based on known approaches, if the detergent is removed, all detergent molecules in the center are extruded through the hydrophilic heads, eliminating the oil-based center and forming a bilayer system, as shown in Figure 2 (12). This enlarged micelle, or liposome, provides solutions to the two listed problems. Its size is > 200 nm ø. Detergent removal, though, is unstable, and can produce liposomes as large as 1000-3000 nm ø. In this case, this instability is essential, for multilamellar (> 1000 nm ø) in conjunction with unilamellar liposomes need to be produced. Unfortunately, with detergent removal, protein orientation is unpredictable (it will either be right side up or inside out). Therefore, multiple experiments were performed to find the best preparation method.

Considering that bothmultilamellar and unilamellar liposomes were necessary for the experiment, size-based cytometry was utilized to separate the two liposomal sizes. Approximately 0.2 of the liposomes were multilamellar, the remaining unilamellar, in every 10 µL of concentrated liposome solution (the concentration of liposomes for every 1 mL was calculated to be 8.95e+15).Multiple solutions were produced in order to increase both respective liposomal concentration. Unilamellar liposomes can be characterized in two classifications: small unilamellar vesicles (SUVs) and large unilamellar vesicles (LUVs). Using the above-mentioned approach, no SUVs were formed (< 100 nm ø). For later experimentation, only LUVs (200 nm ø) and multilamellar liposomes were used for comparison.

Phosphatidylcholine and phosphatidylglycerolwere reconstituted at initial concentrations of 25 mg/mL and 30 mg/mL, respectively, in n-octylglucoside. Cholesterol had little solubility in OG, so it was added to acetone at 50 mg/mL. All lipids were subsequently added together at a volume ratio of 60%, 30%, 10%(PC; Chol; PG) in a predetermined quantity, with resulting concentrations of 15 mg/mL, 15 mg/mL, and 3 mg/mL, respectively. Two situations were tested during this procedure: adding the membrane proteins before and after lipids. Adding the proteins (ES, TRAIL, anti-EpCAM, PS, stabilizing proteins) before the lipids was more efficient. This was proven by the color of the resultant solution. Due to high liposome density, the color of the solution should be white and opaque; when adding proteins after lipids, no white color was seen.

All proteins were dissolved according to the supplier’s instructions, and final concentrations of ES, PS, and anti-EpCAM were all 70-90 nM, or approximately 5-15 µg/mL. Stabilizing proteins and TRAIL had concentrations of 20 µg/mL and 4 µg/mL, respectively. Concentrations were evaluated using liposomal surface area, protein structure, and experimentation. If results were low, densities were reevaluated.Due to low concentrations, volume ratios of proteins were not taken into account. Combinations of proteins were used according to the experimental plan.

Detergent removal was completed by Pierce Detergent Removal Spin Columns. Other methods (i.e. dialysis, filtration, ultrasound)were expensive, time-consuming, and required extra materials. Pierce Detergent Removal was as efficient as these other methods, with protein recovery at 95% and detergent removal efficiency at 99%for octylglucoside.

  1. Using one Pierce Detergent Removal Spin Column, remove the bottom closure from column and loosen cap
  2. Place column into the 2mL collection tube.
  3. Centrifuge for 1 minute at 1500 x G(4126 rpm) to remove storage buffer.
  4. Add 0.4 mL wash/equilibration buffer and centrifuge at 1500 x Gfor 1 minute and discard the buffer. Repeat this step two additional times.
  5. Place column in a new 1.5mL collection tube.
  6. Slowly apply detergent containing the sample (25-100 µL) to the top of the compact resin bed and incubate for 2 minutes at room temperature.
  7. Centrifuge at 1500 x G (4126 rpm) for 2 minutes to collect the detergent-free sample. Discard the used column.

The main component in the column was the removal resin (0.5 mL), which attracted to detergent molecules and allowed all other substances to be collected.

Atomic Force Microscopy – Single Molecule Force Spectroscopy

The electrostatic charge of multilamellar liposomes was important in order for strong adhesion to the AFM cantilever. For unilamellar liposomes, due to the abundance and size, adding a high pH buffer, sodium bicarbonate, would accomplish adhesion, so no electrostatic force was needed. The cantilever tips, as seen under 400x magnification in Figure 3, were washed with acetone 3 times for 5 minutes each. After drying in an UV/ozone chamber, 50 µL of sodium biocarbonate and 2 µL of liposomal solution is added to the cantilever. To keep the added solutions around the cantilever, parafilm is used. The cantilever-solution mixture was incubated for 30 minutes; afterwards, the cantilevers were attached to the slot in the AFM machine. 30 force scans were taken of 10 cells for reduced error. The scan prints a graph plotting voltage (Figure 4). To converts volts into force, the string constant and base unit of a flat surfaceneeded to be calculated.




Cell Culture and Preparation

PC3 (prostate cancer cells) were selected as appropriate platforms for optimization study of separation of CTCs, as a higher concentration of E-selectin ligands were found on the PC3 cells. PC3 cells and HL60 cells (leukocytes) were cultured at 37 ⁰C in 5% CO2 in F-12K growth medium containing 1.5 mM L-glutamine supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin with media change every 2–3 days. Cells were then released through incubation in 0.05% Trypsin–0.53 mM EDTA at 37 ⁰C for around 5 minutes. A hemocytometer was used to count cells and appropriate dilution was subsequently carried out to reach a cell concentration of around 105 mL-1. The cell concentration was selected also for ensuring the yield of a large result data pool to reach a reliable conclusion. Cell tracker, a fluorescent stain, was used for easier identification of the PC3 cells. During the culturing of the tumor cells, a fluorescent stain is added. Afterwards, it is either mixed with PBS or whole blood. When PBS is used, alginate is added to mimic the viscosity of the blood. Alginate also reduces the cell settling effect described earlier. During testing, a syringe pump pushes the blood inside the device. The syringe pump also controls the infusion rate.

Flow Test Measurements

During the test, the blood will flow using a syringe pump to monitor the shear rates/stresses. The device will be observed under a microscope to note the interactions of the different factors involved in the attachment. The exact quantity of cells remaining with the liposome on its surface will be determined using a hemocytometer. Efficiency for bonding is the final shear stress needed to break the bond, and the efficiency for the cancer cells is the number of cancer cells divided by the total cells inserted for each level or rate of flow.

Wave Designing

Three methods were used to create the wave designs. All but one was post-fabrication. The first technique is preparing the initial pattern directly onto wax. The wax is poured over the silicon wafer and the pattern is ingrained. Wax is very sensitive to temperature changes; therefore, if it is heated to a certain temperature, the grooves will slightly melt, creating a smoother, wave-like surface. The second method is the reflow process. This uses the negative photoresist process more than once. Repeated UV exposure damages the sharp edges of the groove, creating curved grooves, or waves. The final method was producing a wave-pattern using a 3D printer. The wave pattern is programmed into the 3D printer and the PDMS is poured on the printed surface rather than a silicon wafer. The most effective, inexpensive, and easiest was the double reflow process.

Device Parameters for the Wave & Mutli-Groove Designs

Widths of channel: 1 mm and 5 mm. These were used because they are ideal widths for microfluidic devices (< 8 mm). Specifically, they were chosen to have two completely different widths for experimentation. The optimal width is determined by observing the isolation efficiency.

Inclination Angle: 30 and 45⁰. These are based on a computer simulation. The exact path of the cells was calculated by noting the velocities from the preliminary results.

Wave Widths: 35 and 100 µm. The width has to allow the cells to roll for an extended period of time. The diameter of the CTC is 17-30 µm. Therefore, all widths have to be greater than 30. 

Both wave and groove designs were tested. Groove designs were eliminated from the device due to the harsh deformation from the sharp edges.

In the wave design, the height of the channel, h, is approximately 41 µm. The depth of the waves, d, is ≈ 32 µm. The height of the channel and depth of waves also changed from 25 to 40/50 µm.

Results and Discussion


Lifetime Studies

Lifetime studies were used to determine the efficiency of liposomes in the patient’s blood based on time. Ten microliters of the liposome solution was added in 2.8 x  PC3 cells. Viscometers induces constant pressure throughout the fluid; however, the shear stress (forces) experienced does not mimic those in the blood. In order to accurately measure the lifetime, slight, continuous vortexing action (~1034 rpm) was induced throughout the time period, mimicking blood stress. The liposomal solution was incubated at 37°C (temperature of body). The lifetime was determined by measuring how many cells were attached to liposomes. This measures the time of peak liposomal activity.


Anytime further will mark a stagnancy in activity of liposomes, allowing the remaining CTCs to induce metastasis. Time recorded was every 2 hours from 0 to 48 hours. Based on the graph, the efficiency of EP liposomes continues to rise and only begins to level off at 30 hours This implies that not only the lifetime, but also the efficiency of liposomes, compared to ES, for every hour was higher. ES liposomes starts to level off and, surprisingly, slightly decline after 26 hours. The highest efficiency that the ES liposome reached was 0.51 at 26 hours. As time progressed afterwards, the efficiency slightly declined. This implies that there will be a time in the blood vascular system when the liposomes break their bonds with cells due to fluid force, or when fewer liposomes have the ability to selectivelytarget cells. This demonstrates the long-lasting effects and high selectivity of EP liposomes.


Atomic Force Microscopy Studies

The AFM machine was used to measure the rupture force of the bonds between the liposome and cells. This would measure the breaking point of cellular adhesion while in the blood.

The maximum shear stress acting on a 10 µm ø cancer cell in the most stressful area of the body (aorta) during systole is 5.4 x  N, and on a 200 nm ø liposome is 6.3 x  N. The average shear stress on the cancer cell attached to the liposome is 1.2 x  N, which includes capillaries (low pressure and velocity).

The mean of TRAIL-CTC is 2.62 x N for the rupture force. This shows that TRAIL alone will not be efficient in targeting CTCs. Although size is not a problem with the liposome, TRAIL is still too weak for adhesion against blood flow. This explains why efficiency was low in the literature for TRAIL. 

The mean of ES-CTC is 1.88 x N for the rupture force. Therefore, ES-CTCs will not remain bonded to each other throughout circulation. This explains the decline in efficiency as time progresses after 26 hours in the body. The liposome is continually weakened until the bond breaks. 

The mean of EP-CTC is 8.02 x N for the rupture force. This is 6 orders more stable than ES-CTC. Thus, it will produce a strong ligand-receptor interaction despite the forces of blood flow. This explains the consistent increase in efficiency as it stays longer in the blood, as shown in the Lifetime Studies. This also explains the higher efficiency than ES-CTCs. 

The AFM works by transmitting mechanical stress from the AFM cantilever tip to an electrical charge. The laser reflects off the sample, and is transmitted from light to electrical energy using a 2-segment photodiode. The two signals put together formed a voltage graph on the computer. The schematic below correctly explains the process:

Fundamental Tests

For ES liposomes, the TRAIL and E-selectin were able to only capture a maximum of 56% of cancer cells. The best results were in the wave device in blood under an 120/s shear rate. This demonstrates that when large unilamellar ES liposomes enter the blood, it is crucial to attach first to white blood cells before targeting the cancer. Predictably, the efficiency decreased as flow rate increased, in agreement with the AFM results.

EP liposomes achieved a much higher efficiency in killing cancer cells. Device 2 was the more capable device. Since waves, however, are not found in the endothelial lining of the vessel, hemodialysis with this device seems most beneficial. EP liposomes in a pure CTC solution is shown rather than inblood because the two scenarios were very similar. The highest efficiency reached was approximately 77%. Interestingly, there was only slight variation in cell killing when shear rate increased. This proves that although the lifetime of liposomes is not very long, they are efficient carriers and are able to eliminate large populations of CTCs in a limited amount of time. In the EP liposome trial, no harm to the white blood cells occurred because EP does not adhere to normal cells. However, with ES and TRAIL, although no normal cell destruction occurs, it is unknown if there is any cell deformation.

Since flow rate in the capillaries, veins, and most arteries mimic the flow rates experienced in these trials, blood can enter the device at a normal speed and can be separated with a high efficiency. In terms of further optimization of metastatic cancer cell targeting and killing, liposomes can be pre-coated onto the surface of the device. This allows the capture of all cells in the device without the concern of fluid pressure, fluid speed, and length of the device. 


Based on the lifetime studies, one of two actions must be taken during clinical trials: (1) provide a continuous supply of liposomes; or (2) utilize hemodialysis techniques. A continuous supply of ES liposomes causes clogging with multiple white blood cells, and inactivation and immobility of healthy immune cells. Therefore, either EP liposomes must be used or hemodialysis can be performed, both of which have proven to be highly efficient. EP liposomes were also noted to have a stronger bond with cancer cells, allowing it to resist the unpredictable shear forces of the blood. Overall, the highest efficiency was noted in the wave design device with EP liposomes at approximately 77%. Further experimentation could be conducted on encapsulating chemotherapeutic drugs in EP liposomes. Since EP liposomes never specifically attach to normal cells, the body tissue will not be affected. When EP liposomes enter the blood, they will search for cancer cells and tumor tissue. Once adhered with the anti-EpCAM, they will release the encapsulated drug into the cancerous tissue, thereby destroying primary and secondary sites.


  1. S. Wicha and D. F. Hayes, J. Clin. Oncol., 2011, 29, 1508.
  2. Pantel, R. H. Brakenho ff and B. Brandt, Nat. Rev. Cancer, 2008, 8, 329.
  3. D. Pratt, C. Huang, B. G. Hawkins, J. P. Gleghorn and B. J. Kirby, Chem. Eng. Sci., 2011, 66, 1508.
  4. Gertler, R. Rosenberg, K. Fuehrer, M. Dahm, H. Nekarda and J. Siewert, in Molecular Staging of Cancer, 2003, vol. 162, p. 149.
  5. Vona, A. Sabile, M. Louha, V. Sitruk, S. Romana, K. Schutze, F. Capron, D. Franco, M. Pazzagli, M. Vekemans, B. Lacour, C. Brechot and P. Paterlini-Brechot, Am. J. Pathol., 2000, 156, 57.
  6. J. Tan, R. L. Lakshmi, P. Chen, W. T. Lim, L. Yobas and C. T. Lim, Biosens. Bioelectron., 2010, 26, 1701.
  7. L. Stott, C.-H. Hsu, D. I. Tsukrov, M. Yu, D. T. Miyamoto, B. A. Waltman, S. M. Rothenberg, A. M. Shah, M. E. Smas, G. K. Korir, F. P. Floyd, A. J. Gilman, J. B. Lord, D. Winokur, S. Springer, D. Irimia, S. Nagrath, L. V. Sequist, R. J. Lee, K. J. Isselbacher, S. Maheswaran, D. A. Haber and M. Toner, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 18392.
  8. Nagrath, L. V. Sequist, S. Maheswaran, D. W. Bell, D. Irimia, L. Ulkus, M. R. Smith, E. L. Kwak, S. Digumarthy, A. Muzikansky, P. Ryan, U. J. Balis, R. G. Tompkins, D. A. Haber and M. Toner, Nature, 2007, 450, 1235.
  9. Wan, Y.-t. Kim, N. Li, S. K. Cho, R. Bachoo, A. D. Ellington and S. M. Iqbal, Cancer Res., 2010, 70, 9371.
  10. Liu, S. Wang, Y. Song and J. Yang, Micro Nanosyst., 2012, 4, 254.
  11. Zhao, C. H. Cui, S. Bose, D. Guo, C. Shen, W. P. Wong, K. Halvorsen, O. C. Farokhzad, G. S. L. Teo and J. A. Phillips, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 19626.
  12. Myung, J. H., Launiere, C. A., Eddington, D. T., & Hong, S. Langmuir, 2010, 26, 8589.
  13. Wujin Sun, Zhen Gu, ATP-Responsive Drug Delivery Systems, Expert Opinion on Drug Delivery, 2016, 13, 3, 311
  14. Quanyin Hu, Wujin Sun, Chengen Qian, Chao Wang, Hunter N. Bomba, Zhen Gu, Anticancer Platelet-Mimicking Nanovehicles, Advanced Materials, 2015, 27, 44, 7043
  15. Chaffer CL, Weinberg RA (2011) A perspective on cancer cell metastasis. Science 331(6024):1559–1564.
  16. Goldsmith HL, Bell DN, Braovac S, Steinberg A, McIntosh F (1995) Physical and chemical effects of red cells in the shear-induced aggregation of human platelets. Biophys J 69(4):1584–1595.
  17. Allard WJ, et al. (2004) Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin Cancer Res 10(20):6897–6904
  18. Gout S, Tremblay PL, Huot J (2008) Selectins and selectin ligands in extravasation of cancer cells and organ selectivity of metastasis. Clin Exp Metastasis 25(4):335–344.

Leave a Reply

Method of Targeting Circulating Tumor Cells

Cancer metastasis accounts for 90% for all cancer-related deaths. Thus, treatment and prevention of secondary tumor formation is vital. A liposome coated with an adhesion molecule and death ligand is proposed in order to target the cancer cells in the bloodstream.

Scroll to top