Facilitating Intracellular Drug Delivery by Ultrasound-Activated Microbubbles

The goal of this thesis was to investigate the combination of ultrasound and microbubbles
(USMB) for intracellular delivery of (model) drugs in vitro. We have focused on clinically
approved drugs, i.e. cisplatin, and microbubbles, i.e. SonoVue™, to facilitate clinical
translation. In addition, model drugs, predominantly SYTOX Green, were used to increase
the understanding of the mechanisms involved in USMB-induced cellular uptake, in
particular during cellular recovery after membrane permeabilization.
Chemotherapy is one of the major treatment modalities in cancer, next to radiotherapy and
surgery. However, in certain types of cancer, chemotherapy shows low efficacy while
inducing serious adverse effects. Much research has focused on increasing the efficacy or
decreasing the toxicity of these chemotherapeutic drugs, with limited success. The
combination of ultrasound and microbubbles is a promising new technique to increase the
efficacy of chemotherapeutic drugs, by increasing drug extravasation in the tumor and
enhancing cellular uptake of drugs.
The variety in treatment strategies for USMB-induced drug delivery were discussed in
Chapter 2, focusing on drugs and microbubbles, route of administration, ultrasound
equipment and treatment schedules. Chemotherapeutic drugs can be loaded into the shell of
the microbubble, resulting in a microbubble that not only acts as a cavitation-enhancing
agent, but also as a drug carrier. In this case, ultrasound-induced destruction of the
microbubble results not only in vascular- and membrane permeabilization, but also in
concurrent local drug release. This may lead to higher efficacy and lower toxicity compared
to the co-administration approach, where conventional microbubbles and free drugs are
administered simultaneously. However, the co-administration approach is likely the fastest
way towards the clinic, as both microbubbles and drugs are clinically approved, in contrast
to the drug-loaded microbubbles. While drugs and microbubbles can be injected directly
into the tumor to ensure direct contact between drugs, microbubbles and tumor cells, most
in vivo studies opt for intravenous administration. As microbubbles are confined to the
vascular space due to their micrometer size, they will be primarily in contact with
endothelial cells, rather than tumor cells. Nevertheless, enhanced chemotherapeutic efficacy
when combined with USMB has been reported for both routes of administration.
Although pre-clinical success has been demonstrated in vitro and in vivo, the translation
towards the clinic has been limited thus far. One clinical case study has been published,
where pancreatic cancer patients were treated with gemcitabine and concurrent USMB,
using the clinically approved SonoVue™ microbubbles. Patients who received gemcitabine
in combination with USMB showed enhanced treatment efficacy compared to a historical
control group of patients that received only gemcitabine [1].
As discussed in Chapter 1 of this thesis, chemotherapy is often combined with surgery or
radiotherapy in an (neo)adjuvant setting. Nevertheless, practically all studies on
chemotherapeutic drug delivery by USMB focus on chemotherapy as a single modality
treatment. Cisplatin is one of the most widely used chemotherapeutics in combination with
radiotherapy, where both treatments act synergistically to enhance treatment efficacy. In
advanced staged (stage III and IV) head and neck cancer patients with high-risk features,
treatment often includes concurrent cisplatin and radiotherapy treatment.
Therefore, Chapter 3 addressed the potential of USMB to enhance the efficacy of
chemoradiation by increasing the intracellular uptake of cisplatin in a head and neck cancer
model in vitro. Measurements with inductively coupled plasma mass spectrometry
demonstrated that intracellular cisplatin levels were 2.7-fold higher when cisplatin
treatment was combined with USMB, compared to cisplatin only. These enhanced
intracellular cisplatin levels were associated with increased DNA damage, which was 82%
higher after cisplatin + USMB treatment compared to cisplatin only treatment, while
USMB alone did not enhance DNA damage. Subsequently, USMB significantly enhanced
the efficacy of the combination cisplatin and radiotherapy, without affecting radiotherapy
efficacy in the absence of cisplatin. While these results demonstrate the potential of USMB
to enhance the efficacy of the combination cisplatin and radiotherapy in vitro, future
research should confirm if similar results can be obtained in vivo.
The next chapters aimed to increase the understanding of the biological mechanisms
involved in USMB-induced membrane permeabilization and drug uptake. The majority of
research into the mechanism of USMB focused on the direct effects on the cell membrane,
e.g. by investigating the pore formation or the upregulation of endocytosis [2], [3].
However, less attention has been paid to cellular recovery after USMB exposure and how
this relates to prolonged membrane permeability and (model) drug uptake. In Chapter 4,
we studied the duration of cell membrane permeability after a single USMB treatment by
assessing the intracellular uptake of model drug SYTOX Green as a function of acoustic
pressure and cell line. Fluorescence microscopy demonstrated that the number of cells with
SYTOX Green uptake was highest when cells were sonicated in the presence of the dye.
The number of SYTOX Green positive cells decreased with increasing time interval
between USMB treatment and addition of the model drug, suggesting that more cells
recovered from USMB-induced damage with increasing time after sonication. The duration
of membrane permeability in the breast cancer cell line MDA-MB-468 was only slightly
affected by the acoustic pressures used in the study, and returned to the level of nonsonicated
cells within 1 – 2 hours after USMB treatment. Larger differences in duration of
membrane permeability after USMB exposure were found between cell lines. As mentioned
before, the membrane permeability of MDA-MB-468 cells returned to control level within
2 hours, while this lasted more than 3 hours for the breast cancer cell line 4T1, and less than
1 hour for the endothelial cell line HUVEC. This difference in time window of drug uptake
following sonication may reflect the heterogeneity between cell lines in their ability to
recover from the biophysical stress induced by USMB.
Intracellular drug uptake by USMB has been associated with disruption of the cellular
membrane. Direct microscopic observations in addition to transmembrane current measurements have demonstrated that membrane pores can be created when cells are
exposed to USMB, which facilitate the intracellular uptake of (model) drugs. However, it
recently has been suggested that USMB may also disrupt the tightly regulated lipid
composition of the cell membrane. Therefore, the effect of USMB on the membrane lipid
phosphatidylserine (PS) was studied and presented in Chapter 5. Fluorescence microscopy
demonstrated that USMB induced a transient PS translocation from the inner leaflet of the
plasma membrane to the outer leaflet. The duration of PS externalization was positively
correlated with model drug uptake, suggesting that it is related with USMB-induced
membrane permeability.
While PS externalization is normally regarded as a precursor of apoptosis, USMB-induced
PS externalization was found to be transient and not associated with loss of cell viability. In
our experiments, we did not find evidence that USMB-induced PS externalization occurred
via calcium-stimulated activation of the phospholipid scramblase. Therefore, it was
hypothesized that the membrane pores could provide a new pathway for PS translocation
between the inner and outer leaflet of the plasma membrane. During membrane recovery,
external PS may be internalized by the transmembrane lipid transporter protein flippase or
by endocytosis and exocytosis.
SYTOX Green and propidium iodide (PI) are membrane impermeant dyes that exhibit 500-
fold and 30-fold fluorescence intensity enhancement, respectively, upon binding to nucleic
acids. They are widely used as model drugs to study USMB-induced membrane
permeabilization and cellular uptake, as was shown in Chapters 4 and 5. The fluorescence
intensity enhancement after USMB-mediated cellular internalization has been associated
with cellular uptake kinetics [4], where others related the duration of fluorescence intensity
enhancement to pore size [5]. However, SYTOX Green fluorescence intensity enhancement
is induced by nucleic acid binding, which is an indirect result of cellular uptake.
To increase the understanding and interpretation of the fluorescence intensity enhancement
of model drugs like SYTOX Green after cellular internalization, Chapter 6 characterized
the fluorescence intensity enhancement upon intracellular SYTOX Green uptake in vitro,
and how this is influenced by experimental parameters. Swept-field confocal microscopy of
single cells showed that a membrane pore was created upon exposure to USMB. SYTOX
Green fluorescence intensity increased around the pore, suggesting that the model drug
entered the cell through this pore. Next, the SYTOX Green fluorescence signal spread
throughout the cell and accumulated in the nucleus during the 9-minute acquisition,
whereas the cytosolic signal reached a maximum fluorescence intensity already after 15
seconds. These results underlined that SYTOX Green fluorescence intensity depends
primarily on nuclear accumulation and binding DNA and not just on cellular uptake.
The prolonged signal intensity enhancement of SYTOX Green after USMB exposure was
confirmed in populations of cells on a fibered confocal fluorescence microscope. The
cellular fluorescence signal of SYTOX Green increased at least for 10 minutes, and often
more than 30 minutes, after USMB exposure, and was substantially affected by the
experimental conditions. It was found that the duty cycle of the fluorescent laser largely influenced the signal enhancement of SYTOX Green upon USMB-induced cellular uptake,
due to a 6.4-fold enhancement of the photobleaching rate. In addition, the rate of
fluorescence enhancement upon cellular internalization was demonstrated to be dependent
on the extracellular dye concentration, as we found a positive linear relation between the
dye concentration and fluorescence rate constant. While it has been postulated in the
literature that increased acoustic pressure results in larger pore sizes, we did not find a
relation between the fluorescence rate constants and the acoustic range tested in this study
(350 kPa – 850 kPa). Furthermore, we found that the rate of fluorescence enhancement
differs between the glioma cell line C6 and the head and neck squamous cell carcinoma cell
line FaDu in both ultrasound- and chemically permeabilized cells. This suggests that the
differences in dynamics of fluorescence enhancement cannot be solely attributed to
different susceptibility to ultrasound exposure. The data in this chapter demonstrated that
intercalating model drugs like SYTOX Green are useful as a biomarker for membrane
permeability, but the dynamics of signal enhancement upon cellular internalization should
be carefully interpreted before drawing conclusions on the underlying biology, such as pore
resealing.