Effects of particle architecture on in-vivo pharmacokinetics and bio-distribution of therapeutic nanostructures

PDF

Download

Open document

Flip pages

Copy asset link

Request this asset

Transcript (if available)

Content Effects of particle architecture on in-vivo
pharmacokinetics and bio-distribution of
therapeutic nanostructures
Santosh Peddi
USC ID:
A thesis presented to the Department of Pharmacology and Pharmaceutical
Sciences at University of Southern California in partial fulfillment of the
requirements for the degree MS Pharmaceutical Sciences
August 2016

2

Acknowledgements

I
would
like
to
thank
my
mentor
Dr.
J.
Andrew
Mackay
for
his
continuous
support

and
feedback
throughout
the
project.
I
would
also
like
to
thank
my
committee

members
Dr.
Curtis
Okamoto
and
Dr.
Ian
Haworth
for
their
suggestions
and
time

spent
in
reviewing
my
thesis.
I’m
indebted
to
my
lab
members
and
friends
for

their
support
throughout
the
program

3

Table
of
contents

Abstract

4

Chapter
1:
Protein
polymers
for
small
molecule
drug
delivery

5

1.1:
Small
molecule
anti-‐cancer
drugs:
Challenges
for
delivery

5

1.2:
Nanoparticle
technology
can
address
problems
with
small
molecule
drug
delivery

8

1.3:
Pharmacokinetics/Clearance
of
nanoparticle
drug
carriers

13

1.4:
Materials
for
nanoparticle
fabrication

14

1.5:
Elastin
like
polypeptides

15

Chapter
2:
FKBP-‐ELP
fusion
proteins
for
delivery
of
anti-‐cancer
drug
rapamycin

17

2.1:
Introduction

17

2.2:
Materials
and
Methods

20

2.3:
Results
and
Discussion

26

References

39

4

Abstract

Small
molecule
chemotherapeutics,
although
routinely
used
in
the
clinic,
suffer
from
poor
drug-‐
like
properties,
including
low
water
solubility,
rapid
plasma
clearance,
and
non-‐specific
bio-‐
distribution
causing
toxic
side
effects.
Protein
polymers
of
appropriate
size
can
address
these

issues
by
solubilizing
hydrophobic
drugs,
extending
their
mean
plasma
residence
time
by

preventing
renal
filtration
and
promoting
tumor
accumulation
through
enhanced
permeation

and
retention(EPR)
effect,
thereby
reducing
drug
toxicity.
Moreover,
the
delivery
vehicle,
being

proteinaceous
in
nature,
is
bio-‐degradable,
non-‐toxic
and
largely
non-‐immunogenic.

The
clinical
utility
of
rapamycin,
a
highly
potent
anti-‐cancer
drug
is
limited
by
low
solubility,
low

bioavailablity
and
rapid
systemic
clearance.
To
address
this,
we
synthesized
FKBP-‐elastin
like

polypeptide(ELP)
fusion
proteins
to
utilize
the
tight
binding
of
rapamycin
to
FKBP
for
effective

nanoparticle-‐based
delivery.
Derived
from
human
tropoelastin,
ELPs
are
a
class
of
thermo-‐
responsive
protein
polymers
with
sequence
of
(VPGXG)
n
,
where
X
is
the
guest
residue
and
n
is

number
of
repetitive
units.
Below
a
characteristic
transition
temperature
(dictated
by
X
and
n),

ELPs
and
ELP
fusion
proteins
stay
in
solution
and
above
the
transition
temperature,
they
self

assemble
nanostructures,
typically
micelles.
To
understand
how
polymer
architecture
affects

carrier
pharmacokinetics(PK)
and
bio-‐distribution,
we
recombinantly
expressed
a
FKBP-‐ELP

fusion
library
consisting
of
FKBP-‐(VPGAG)
192

(FA),
FKBP-‐(VPGAG)
192

-‐FKBP
(FAF),
FKBP-‐
(VPGSG)
48
(VPGIG)
48

(FSI).
At
physiological
temperature,
FA
and
FAF
are
soluble
monomers,
while

FSI
phase
separates
into
25
nm
spherical
micelles
decorated
with
FKBP
on
the
corona.
For
non-‐
invasive
PK
and
bio-‐distribution
evaluation,
we
labeled
these
drug
carriers
with
IR800
without

any
effect
on
their
physico-‐chemical
properties
and
physical
stability.
In
a
breast
cancer
xenograft

mouse
model,
we
seek
to
understand
how
nanoparticle
architecture
and
route
of
administration

(intravenous
vs.
subcutaneous)
affect
their
circulation,
clearance
and
bio-‐distribution
profile.

5

Chapter
1:
Protein
polymers
for
small
molecule
drug
delivery

1.1:
Small
molecule
anti-‐cancer
drugs:
Challenges
for
delivery

Cancer,
a
leading
cause
of
death
worldwide
is
a
group
of
diseases
characterized
by
abnormal
cell

growth
with
the
potential
to
spread
to
other
parts
of
the
body.
In
2012,
there
were
14
million

new
cancer
cases
worldwide
and
8.2
million
cancer-‐related
deaths
[1]
.
In
2016,
an
estimated
1.6

million
new
cancer
cases
will
be
diagnosed
and
nearly
600,000
people
will
die
from
the
disease.

These
numbers
clearly
emphasize
the
need
for
innovations
in
cancer
treatment.
Over
the
past

few
years,
overall
cancer
death
rate
has
declined,
thereby
increasing
the
number
of
cancer

survivors.
These
treads
suggest
progress
towards
cancer
treatment
but
many
questions
yet
need

to
be
answered.

Cancer
chemotherapy
is
currently
the
first
line
of
treatment
against
cancer.
Over
100
anti-‐cancer

drugs
have
been
approved
and
these
drugs
primarily
function
by
inhibiting
mitosis
and
inducing

apoptosis.
Unfortunately,
chemotherapy
is
highly
toxic
as
it
kills
healthy
cells
too,
especially
the

ones
that
naturally
divide
rapidly.
Small
molecule
anticancer
agents
exhibit
poor
drug-‐like

properties,
thereby
posing
delivery
challenges:

a)
Poor
solubility,
bioavailability,
pharmacokinetic
profile:

The
in-‐vivo
effect
of
a
drug
is
a
function
of
pharmacokinetics
(ADME
profile)
and

pharmacodynamics
(potency).
The
critical
physicochemical
properties
of
a
drug
determining
its

ease
of
delivery
are
solubility
and
permeability.
Many
chemotherapeutics
suffer
from
low

bioavailability,
which
could
be
solubility
limited,
or
permeability
limited,
or
both

solubility/permeability
limited.
Classic
examples
of
solubility
limited
poor
bioavailability
include

tamoxifen
[2]
,
rubitecan,
sorafenib
[3]
,
gefitinib
[4]

etc.
While
permeability
limited
poor

bioavailability
is
not
often
observed,
>
60%
drugs
are
substrate
for
one
or
other
types
of
efflux,

suggesting
dominating
role
of
drug
efflux
in
the
oral
bioavailability.
Permeability
limited
poor

availability
is
exhibited
by
cyclophosphamide,
anastrozole
[5]
,
letrozole
[6]
,
doxorubicin
[7]
,

methotrexate
etc.
These
drugs
need
solubility
enhancement/absorption
enhancement
to
make

6

them
clinically
useful.
Rapid
metabolism
and
clearance
by
kidneys
demand
frequent
drug

administration,
further
limiting
the
clinical
use
of
many
chemotherapeutics.

Although
oral
administration
can
be
considered,
concentration
dependent
toxicity
of

chemotherapeutics
precludes
the
use
of
administration
routes
where
the
drugs
can
reach
high

local
concentrations
like
oral,
subcutaneous,
transdermal
etc.
Hence,
intravenous
route
is
widely

used
to
deliver
chemotherapeutics,
but
only
a
small
fraction
of
this
administered
dose
actually

reaches
the
tumor.

b)
Uncontrolled
distribution/Toxicity:

When
chemotherapeutics
reach
systemic
circulation,
there
distribution
is
not
controlled,
and

this
leads
to
dose
dependent
toxicity.
Rapidly
dividing
healthy
cells
of
the
bone
marrow,
gut,

lymphoid
tissue,
and
hair
follicles
are
depleted,
causing
side-‐effects
ranging
from
nausea
and

fatigue
to
blood
disorders
and
neurological
effects
(Figure
1).

Figure
1:
Common
side
effects
of
conventional
chemotherapy.
Source:
NCI

7

c)
Emergence
of
resistance:

Cancer
cells
can
gain
drug
resistance
through
cellular
and
non-‐cellular
mechanisms.
Non-‐cellular

mechanisms
include
restricting
drug
access
through
poor
vascularization,
high
interstitial

pressure
and
low
microvascular
pressure
to
retard
drug
extravasation,
and
acidic
tumor

microenvironment
to
protonate
basic
drugs,
and
impede
their
cellular
diffusion.
Cellular

mechanisms
include
loss
of
a
cell
surface
receptor
or
transporter
for
a
drug
[8]
,
up-‐regulation
of

efflux
transporters,
especially
p-‐glycoprotein
[9]

which
actively
pump
drugs
from
inside
of
cancer

cell
to
outside,
thereby
reducing
drug
sensitivity
and
intracellular
drug
accumulation.
For

instance,
P-‐gp
transports
a
wide
variety
of
hydrophobic
anti-‐cancer
drugs
such
as
vinblastine,

doxorubicin,
vincristine,
and
taxol,
and
therefore
its
increased
expression
has
been
correlated

with
resistance
to
these
[10]
.

Figure
2:
Mechanism
of
drug
resistance
in
cancer

Source:
Gottesman
MM,
Annual
Review
of
Medicine,
53:615-‐627

These
issues
make
current
chemotherapy
a
toxic,
sub-‐optimal,
non
patient
compliant
treatment

strategy
for
cancer.

8

1.2:
Nanoparticle
technology
can
address
problems
with
small
molecule
drug

delivery

National
Cancer
Institute
defines
nanoparticles
as
colloidal
particles
in
the
size
range
of
1-‐100

nm.
Although
there
is
no
size
restriction
on
therapeutic
nanoparticles,
size
range
of
10-‐100
nm

are
most
effective.
More
recently,
scientists
place
less
stringent
limitations
on
the
exact

dimensions,
and
defines
the
'right'
size
in
bio-‐nanotechnology
in
an
operational
fashion,
with

respect
to
addressable
unmet
needs
in
biology.
Nanoparticles
can
be
fabricated
into
different

sizes,
shapes,
surface
properties,
architectures
using
a
variety
of
materials
ranging
from
synthetic

polymers,
metals,
inorganic
materials
to
proteins,
peptides,
lipids,
viruses
etc.
The
advantages
of

nanoparticle
technology
include
increased
solubilization
potential,
protection
from
metabolic

degradation,
flexibility
in
surface
functionalization,
active
targeting
potential,
ability
to

encapsulate
wide
variety
of
drugs
and
ability
to
engineer
drug
release/particle
degradation

profiles.
With
respect
to
anti-‐cancer
drug
delivery,
nanoparticle
technology
seems
promising

because
of
its
following
advantages.

a)
Improve
solubility
of
water
insoluble
drugs:

Large
number
of
chemotherapeutics
and
newly
discovered
molecules
with
cytotoxic
properties

cannot
be
administered
intravenously
because
of
their
low
water
solubility.
Estimates
state
that

40%
of
the
drugs
in
the
pipelines
have
solubility
problems
[11]
.
Literature
states
that
about
60%

of
all
drugs
coming
directly
from
synthesis
are
nowadays
poorly
soluble
[12]
.
Micelles,
the
most

widely
studied
nanoparticles
are
amphiphilic
block
copolymers
that
assemble
to
form
core/shell

structures
in
aqueous
solutions.
The
core
is
composed
of
hydrophobic
block
and
can
solubilize

and
retain
water
insoluble
drugs,
while
the
shell
is
hydrophilic
and
stabilizes
the
core.
The
first

polymeric
micelle
therapeutic
nanoparticle
was
Genexol-‐PM
(PEG-‐poly(D,L-‐lactide)–
paclitaxel)
[13]
.
The
micelle
size,
drug
loading
capacity,
drug
release
kinetics
can
be
controlled
by

tuning
the
block
structures
for
desired
application.

9

Figure
3:
Micellar
structure
of
Genexol-‐PM.
Paclitaxel
is
encapsulated
in
the
hydrophobic
PLA

core
of
micelle

b)
Increased
drug
accumulation
in
tumor/reduced
toxicity:

Nanoparticles
of
the
right
size
can
inherently
get
accumulated
in
the
tumor
by
enhanced

permeation
and
retention
(EPR)
effect
(Figure
4).
The
tumor
neovasculature
is
poorly
developed

and
is
characterized
by
loops,
dead
ends
and
openings
that
lead
directly
into
perivascular

space
[14]
.
Highly
permeable
tumor
vasculature
is
known
to
have
openings
as
big
as
400-‐600
nm
[15]

in
diameter,
facilitating
nanoparticle
extravasation
into
tumor
tissue.
These
pores
are
much

larger
than
the
junctions
in
normal
tissue
where
the
gaps
are
usually
less
than
6
nm.
Although

tumors
manage
to
signal
growth
of
blood
vessels
through
angiogenesis
for
access
to
nutrients,

they
fail
to
develop
a
functional
lymphatic
system,
resulting
in
poor
clearance
of
nanoparticles

once
they
accumulate
in
the
tumor.
These
phenomena
together
comprise
EPR
effect
[16]
.

10

The
selective
accumulation
of
nanoparticles
in
the
tumor
tissue
should
reduce
drug
distribution

to
healthy
tissues,
and
thereby
reduce
toxicity
associated
with
free
drug
administration.
Bae
and

colleagues
systematically
examined
the
biodistribution
of
a
micelle
based
formulation
carrying

adriamycin
[PEG–p(Asp-‐Hyd-‐ADR)]
at
several
time
points
in
experimental
mice
[17]
.
The
circulation

of
drug,
as
measured
by
the
area
under
the
curve
(AUC),
was
15-‐fold
greater
than
that
of
free

ADR.
There
was
also
a
greater
concentration
of
micelles
than
of
free
ADR
in
the
tumor
and
a

lower
concentration
in
the
heart
and
kidney,
explaining
the
enhanced
efficacy
of
the
micelle-‐
delivered
ADR
and
the
reduction
in
side
effects,
such
as
cardiotoxicity
and
nephrotoxicity.
Tumor-‐
specific
accumulation
lasted
for
up
to
50
hours
without
significant
decline.
At
the
same
time,
a

constant
level
of
micelle
accumulation
in
the
liver
and
spleen
was
also
observed.
In
most
relevant

studies,
the
accumulation
of
nanoparticles
in
the
liver,
spleen
or
kidney
is
commonly
observed

depending
on
the
size
and
surface
characteristics
of
the
particle,
and
this
accumulation

constitutes
the
major
concern
regarding
the
toxicity
of
therapeutic
nanoparticles.
Although
acute

toxicities
are
not
usually
observed,
long-‐term
observations
are
still
needed
to
understand
any

potential
harmful
effects
of
therapeutic
nanoparticles
on
major
organ
tissues.

Although
particles
can
accumulate
in
tumor,
their
effectiveness
as
a
therapeutic
is
not

guaranteed
unless
their
drug
release
profile
is
optimized.
Ideally,
they
are
expected
to

encapsulate
the
drug
stably
while
in
circulation
and
release
drug
only
in
the
site
of
action.
The

Figure
4:
Vascular
pathophysiology
and
EPR

effect
in
nanoparticle
delivery.
Scheme

representing
the
microvasculature
of
normal
(A)

and
tumor
(B)
tissue.
Poorly
developed
leaky

vasculature
allows
10-‐100
nm
sized

nanoparticles
to
extravasate
and
gets

accumulated
with
in
solid
tumor.
Within
tumor

depending
on
their
sustained
drug
release

properties,
nanoparticles
keep
releasing
active

drug
for
significantly
longer
time
point.

Nanoparticles
cannot
leak
through
the
intact

blood
vessels,
so
it
considerably
decreases
the

systemic
toxicity.
Source:
Transl
Cancer
Res.
2013

Aug
1;
2(4):
309–319

11

rate
and
mechanism
of
release
need
to
be
optimum
for
anti-‐tumor
efficacy.
A
low
drug
leakage

from
the
particles
in
circulation
is
acceptable
but
rapid
drug
release
in
bloodstream
is
not

characteristic
of
an
efficient
carrier,
as
they
cannot
address
any
disadvantages
of
free
drug.
At

the
same
time,
a
very
slow
rate
of
drug
release
at
the
tumor
site
is
not
desirable
either
because

drug
clearance
may
be
faster
than
release,
and
the
local
drug
concentration
may
not
reach

therapeutic
levels
and
not
confer
any
anti-‐tumor
activity.
An
optimum
rate
of
release
that
can

maintain
local
therapeutic
dose
in
the
tumor
is
desirable,
but
is
difficult
to
achieve.
To
achieve

this,
various
factors
including
polymer
architecture,
hydrophobicity/hydrophilicity,
mode
of
drug

association
with
the
polymer,
such
as
surface
adsorption,
dispersion
homogeneity
in
the
polymer

matrix
and
covalent
linkage
with
the
polymer
backbone
can
be
systematically
altered.

Tumor
targeted
nanoparticles
can
release
drugs
extracelluarly
in
the
tumor
site
and
the
released

drug
can
diffuse
into
cancer
cells
or
can
release
drug
upon
particle
internalization
into
cancer

cells.
Mechanisms
for
extracellular
drug
release
utilize
the
unique
characteristics
of
tumor

microenvironment
or
use
locally
applied
external
forces
like
heat,
sound,
pressure
etc.
The

slightly
acidic
tumor
microenvironment
can
be
used
to
engineer
nanoparticles
that
remain
stable

and
inactive
at
normal
physiological
pH
but
get
activated
and
release
drugs
in
response
to
acidic

tumor
microenvironment.
Local
hyperthermia/hypothermia
can
be
used
to
trigger
thermo-‐
responsive
nanoparticles.
For
example,
Bae
and
colleagues
reported
a
pluronic
nanocapsule
that

maintains
rigid
walls
at
37

o
C
but
the
capsule
permeability
rapidly
increases
when
cooled
to
22

o
C
[18]
.
Extracellular
release
from
particles,
though
possible,
would
suffer
from
possibility
of
drug

resistance
similar
to
free
drug
administration.
Cancer
cells
can
acquire
resistance
by

overexpressing
efflux
transporters,
or
by
activating/deactivating
various
genes
and
proteins.

Apart
from
passive
targeting
via
EPR
strategies,
active
strategies
utilize
molecular
markers
of

cancer,
including
overexpressed
cellular
receptors,
enzymes
and
other
secreted
functional

molecules
like
growth
factors.
These
overexpressed
proteins
typically
aid
in
cancer
survival,

growth
and
metastasis.
Grafting
ligands/antibodies
of
these
overexpressed
receptors
onto

surface
of
nanoparticles
can
actively
accumulate
the
particles
in
tumor
sites
via.
defined
binding

events.
Moreover,
binding
of
few
ligands/antibodies
to
their
cognate
receptors
can
trigger

12

receptor
mediated
endocytosis
[19]

and
facilitate
the
intracellular
entry
of
these
particles.
In
such

cases,
the
particles,
after
internalization,
typically
move
through
maturing
endosomes
[20]
,
before

fusing
to
lysosomes
for
degradation.
This
information
can
be
used
to
advantage
by
engineering

particles
that
can
release
drugs
in
response
to
low
pH
endosomal
and
lysosomal
compartments.

In
few
cases,
the
drug
is
covalently
bound
to
polymer
backbone
through
pH
sensitive
hydrozone

linkers
that
can
hydrolyze
in
response
to
low
endosomal
pH
[21]
.
The
released
drug
can
then

escape
into
cytoplasm
to
demonstrate
its
anti-‐tumor
activity.
In
vitro
and
in
vivo
comparisons

using
internalizing
or
non-‐internalizing
ligands
have
shown
that
the
intracellular
concentration
of

the
drug
is
much
higher
when
it
is
released
from
nanoparticles
into
the
cytoplasm
after

internalization
[22]
.
Intracellular
drug
delivery
can
partially
solve
the
problem
of
cancer
cells

acquiring
resistance,
as
P-‐gp
mediated
active
drug
efflux
is
usually
not
observed
in
this
case.

c)
Improved
drug
pharmacokinetics:

Nanoparticles,
by
virtue
of
their
size,
can
circulate
longer
than
free
drug
by
escaping
glomerular

filtration,
thereby
improving
drug
pharmacokinetics
and
reducing
frequency
of
dose

administration.
Table
1
shows
some
therapeutic
nanoparticle
formulations
extensively
studied

and
how
they
compare
to
free
drug
in
terms
of
circulation
and
clearance
profile.

Name
Carrier
Drug
Circulation

half-‐
time(hr)

Clearance

(ml/minkg)

Fold
change

compared
to
free

drug
(Circulation,

Clearance)

Ref.

SP1049C
Pluronic
micelle
Doxorubicin
2.4
12.6
3.1,
0.88
[23]

NK911
PEG-‐Asp
micelle
Doxorubicin
2.8
6.7
3.5,
0.47
[23]

Doxil
PEG-‐liposome
Doxorubicin
84
0.02
105,
0.001
[23]

Genexol-‐PM
PEG-‐PLA
micelle
Taxol
11
4.8
0.5,
1.3
[23]

Abraxane
Albumin
Taxol
21.6
6.5
0.99,
1.7
[24]

Xyotax
Polyglutamate
Taxol
70-‐120
0.07-‐0.12
3.2-‐5.5,
0.18-‐0.03
[25]

CT-‐2106
Polyglutamate
Camptothecin
65-‐99
0.44
5.6-‐8.5,
0.076
[26]

Table
1:
Pharmacokinetic
information
on
therapeutic
nanostructures
in
humans

13

1.3:
Pharmacokinetics/Clearance
of
nanoparticle
drug
carriers:

When
nanoparticles
reach
the
bloodstream,
they
are
rapidly
coated
by
various
serum
proteins,

complement
proteins
C3,
C4,
C5
and
immunoglobulins
being
the
major
players
by
a
process
called

opsonization
[27]
.
After
opsonization,
macrophages,
primarily
Kupffer
cells
in
liver
can
recognize

opsonins
and
phagocytosis
can
occur
(Figure
5),
which
is
the
engulfing
and
eventual
destruction

or
removal
of
particle
from
the
bloodstream
[28]
.
Since
therapeutic
nanoparticles
are
designed
to

be
bigger
than
the
renal
filtration
cutoff,
particle
phagocytosis
by
macrophages
is
the
major
route

of
clearance.
If
particles
are
not
biodegradable
and
cannot
be
digested
by
lysozymal
enzymes,

they
are
sequestered
in
the
organs
of
reticulo-‐endothelial
system
(RES),
primarily
liver
and

spleen.
The
accumulation
of
particles
in
these
organs
can
occur
leading
to
toxicity
and
other

negative
side
effects.

Figure
5:
In
vivo
clearance
pathway
of

therapeutic
nanoparticles
(nanotubes
in
this

case)
after
binding
to
opsonins
and
after

subsequent
recognition
by
macrophages
in
the

blood
vessels.
The
macrophages
engulf
the

particles
and
sequester
them
to
the
hepato–
biliary
organs,
such
as
the
liver
and
spleen,
for

excretion.
Processed
short
particles
with

favorable
dimension,
orientation,
charge,
and

functionalization
are
eliminated
through
the

renal
excretory
system.

Source:
Int
J
Nanomedicine.
2014
May
6;9

14

Since
clearance
by
RES
can
greatly
reduce
the
half
life
of
particles,
extensive
research
has
been

conducted
to
prevent
opsonization.
It
has
been
found
that
particle
properties
like
size,
shape,

surface
charge,
molecular
weight,
hydrophilicity/hydrophobicity,
surface
architecture
all
play

important
roles
in
extent
and
kinetics
of
opsonization.
Although
there
are
many
contradicting

results
about
how
these
properties
affect
opsonization,
two
findings
could
be
reproduced
by

many
researchers
multiple
times.
Hydrophilic
and
neutral
particle
surfaces
are
less
prone
to

opsonization
and
RES
clearance
compared
to
hydrophobic
and
charged
ones.
One
widely
applied

strategy
to
prevent
RES
clearance
is
to
graft
shielding
groups,
like
polyethylene
glycol(PEG)
which

can
block
the
electrostatic
and
hydrophobic
interactions
that
help
opsonins
bind
to
particle

surfaces.
The
PEG
coating
can
be
achieved
by
covalent
grafting,
entrapping,
or
adsorbing
of
PEG

chains.
It
has
also
been
found
that
longer
PEG
chains
can
improve
particle
circulation
half
life

longer
than
shorter
PEG
chains.

1.4:
Materials
for
nanoparticle
fabrication:

Although
a
wide
variety
of
materials
have
been
used
to
fabricate
therapeutic
nanoparticles,
two

general
classes
stand
out:
synthetic
polymers
and
genetically
engineered
protein
polymers.

Advances
in
polymer
chemistry
[29]

(living
polymerization
techniques)
facilitate
synthesis
of

polymers
with
very
low
polydispersity.
Despite
this
progress,
synthetic
polymers
require
defining

various
terms
to
signify
their
heterogeneous
composition
as
chemical
synthesis
results
in
a

mixture
with
variable
polymer
length,
side
chain
and
drug
substituent
distribution,
and
atomic

composition
in
general.
Though
research
in
drug
delivery
systems
using
synthetic
polymers
laid

the
principles
of
structure-‐function
relationships
[30]
,
present
day
delivery
problems
demand

materials
with
well
defined
molecular
structure
to
facilitate
precise
characterization.
To
achieve

this,
genetic
engineering
[31]

can
be
used
to
provide
unmatched
control
over
macromolecule

components
of
nano-‐sized
systems.
The
following
advantages
of
genetically
engineered,

recombinantly
produced
drug
carriers
make
them
superior
over
synthetic
polymers:

a)
Monodisperse,
well
defined
composition
containing
stereo-‐regular
amino
acids

b)
Relatively
easy
to
include
diverse
amino
acid
residues
and
motif
combinations

15

c)
Ease
of
fusing
peptide
polymer
chains
to
functional/targeting
peptides
and
protein
domains

d)
Programmable
degradation
profile
into
natural
amino
acids

e)
Biocompatibility,
low
or
no
issues
of
immunogenicity

f)
Relatively
low
cost
of
production
on
a
large
scale

g)
Green
and
environment
friendly
approach
compared
to
chemical
synthesis

Recombinant
protein
polymer
synthesis
allows
synthesis
of
protein
polymer
libraries
with
varying

amino
acid
composition,
hydrophobicity,
predictable
secondary
and
higher
order
structures,
and

facilitates
incorporation
of
biologically
functional
peptide/protein
motifs.
Precise
sequence-‐
structure
relationships
can
be
established.
Genetically
engineered
protein
based
drug
carriers

can
be
polymeric
or
non-‐polymeric
in
nature.
Well
studied
polymeric
class
comprise
of
elastin

like
polypeptides
(ELP),
silk
like
polypeptides
(SLP),
extended
recombinant
polypeptide
(XTEN),

and
silk
elastin
like
polypeptides
(SELP).
On
the
other
hand,
non-‐polymeric
protein
based
drug

carriers
are
usually
composed
of
viral
proteins
and
viral
vault
proteins.

1.5:
Elastin
like
polypeptides:

Elastin
is
a
highly
elastic
protein
in
connective
tissue
and
enables
many
tissues
to
restore
their

normal
shape
and
size
after
contracting
or
expanding.
Elastin
is
a
major
component
of
the

extracellular
matrix
of
lungs
[32]
,
blood
vessels,
cartilage,
ligaments,
skin
etc.
The
elastomeric

domains
of
elastin
are
rich
in
hydrophobic
amino
acids
like
valine
and
alanine,
and
regular

presence
of
proline
imparts
a
coiled
structure
to
elastin
fibers.

Elastin
like
polypeptides
(ELP)
are
a
class
of
protein
polymers
derived
from
repeat
sequences
of

elastomeric
domain
of
mammalian
tropoelastin.
ELPs
consist
of
pentameric
repeats
of
[Val-‐Pro-‐
Gly-‐X-‐Gly],
where
X
is
any
guest
amino
acid,
other
than
proline.
This
pentameric
repeat

contributes
greatly
to
the
viscoelastic
properties
of
elastin.
ELPs
are
recombinant
protein

polymers
that
exhibit
many
interesting
properties,
making
them
very
useful
biomaterials.
Firstly,

ELPs
display
near-‐ideal
elasticity
i.e
they
can
return
most
of
the
energy
spent
on
elongation
upon

relaxation.
This
near-‐ideal
elasticity
allows
them
to
convert
energy
within
a
system
through

16

changes
in
intramolecular
interactions
[33]
.
ELPs
also
display
reversible
thermo-‐responsive

properties
(Figure
6).
Upon
heating
a
solution
of
ELP
polymers,
they
transition
from
a
disordered

soluble
phase
to
an
ordered,
compact
two
phase
coacervate
through
hydrophobic
assembly.
The

temperature
above
which
ELPs
coacervate
is
called
transition
temperature.
This
behavior
is

reversible
i.e
upon
cooling,
the
polymers
go
back
into
solution.
The
transition
temperature
can

be
fine
tuned
by
changing
characteristics
of
the
ELP
polymer
itself
or
by
modifying
solution

properties.
ELPs
with
higher
chain
length
(higher
molecular
weight),
and
hydrophobic
guest

residues
display
lower
transition
temperatures.
ELP
concentration,
salt
concentration
and
pH
of

solution
also
impact
their
thermo-‐responsive
behavior.
In
solution,
the
ELP
backbone
is
hydrated

through
highly
ordered
water
structures,
especially
around
the
aliphatic
residues

[34]
.
Upon

heating,
the
hydrophobic
hydration
becomes
disorder
and
drives
the
hydrophobic
assembly
of

ELP
backbones
into
ordered,
condensed
beta
sheet
dominated
spherical
secondary

structures
[35][36]
.

Figure
6:
(A)
ELP
diblock
polymers
undrgo
hydrophobic
collapse
to
form
spherical
micelles
above

transition
temperature
(B)
Above
transition
temperature,
ELPs
form
soluble
state
to
an
insoluble

coacervate

Since
ELPs
display
all
the
desirable
qualities
of
protein
polymer
drug
vehicles
discussed
above,

they
have
a
widespread
application
in
the
field
of
drug
delivery,
employing
various
strategies.

Above
transition
temperature,
ELPs
assemble
spherical
micelles
with
hydrophobic
cores
that
can

17

potentially
solubilize
hydrophobic
drugs.
Certain
ELP-‐drug
conjugates
collapse
into
nanoparticles,

thereby
improving
pharmacokinetics
and
bio-‐distribution
of
the
drug
[37]
.
ELP-‐antibody
fusion

constructs
that
form
multivalent
nanostructures
at
physiological
temperature
improve
PK
and

efficacy
of
antibodies,
while
retaining
the
high
affinity,
high
specificity
binding
of
antibody
to
its

target.
Thermo-‐responsiveness
of
ELPs
can
be
used
to
design
hyperthermia
based
therapeutics.

ELPs
with
transition
temperature
above
body
temperature
but
below
hyperthermia
induced

temperature
can
phase
separate
locally
in
hyperthermic
tissue
resulting
in
high
local

concentration
of
therapeutic
payload,
resulting
in
improved
efficacy
and
reduced
toxic
effects
[38]
.

Chapter
2:
FKBP-‐ELP
fusion
proteins
for
delivery
of
anti-‐cancer
drug

rapamycin

2.1:
Introduction

Rapamycin
(Figure
7),
also
known
as
sirolimus
is
a
macrolide
drug
produced
by
the
bacterium

Streptomyces
hygroscopicus.
It
was
isolated
for
the
first
time
in
1972
on
Easter
Island
and
was

named
rapamycin
after
the
native
name
of
the
island,
Rapa
Nui.
Though
rapamycin
was
originally

developed
as
an
anti-‐fungal,
this
use
was
later
abandoned
due
to
its
potent
immunosuppressive

and
anti-‐proliferative
properties.
Rapamycin
can
inhibit
lymphocyte
proliferation,
and
has
been

repurposed
an
an
immunosuppressant
in
organ
transplantations,
especially
kidney

transplantations.
The
US
FDA
approved
an
oral
formulation
of
rapamycin,
called
Rapamune
in

September
1999.
It
is
available
as
both
oral
solution
and
tablets,
to
be
used
in
organ

transplantations.
Recently,
anti-‐cancer
properties
of
rapamycin
have
been
observed
in
various

pre-‐clinical
models
of
malignancies
of
the
breast
[39]
,
colon
[40]

and
kidney
[41]
.
Though
not
approved

by
FDA
for
treatment
in
cancer,
these
findings
suggest
rapamycin’s
potent
anti-‐proliferative

properties.

18

Figure
7:
Chemical
structure
of
rapamycin

The
mechanism
of
rapamycin’s
cytostatic
activity
has
been
studied
well
in
literature
(Figure
8).

Rapamycin
binds
to
FK-‐506
binding
protein
(FKBP)
with
an
affinity
of
0.2
nM
[42]
.
The
binary

complex
of
rapamycin
and
FKBP
then
binds
to
mTOR
(mammalian
target
of
rapamycin).
The
Rapa-‐
FKBP-‐mTOR
ternary
complex
inhibits
biochemical
pathways
that
are
required
for
cell
progression

through
the
late
G1
phase
or
entry
into
S
phase
of
cell
cycle

[43]
.
G1
cell
cycle
arrest
is
responsible

for
anti-‐proliferative
property
of
rapamycin.

Figure
8:
Mechanism
of

rapamycin’s
cytostatic
activity:

Rapamycin-‐FKBP
binary
complex

binds
to
the
mammalian
target
of

rapamycin
(mTOR).
The
SRL–
FKBP–mTOR
complex
inhibits

biochemical
pathways
that
are

required
for
cell
cycle
progression
1

19

Although
extremely
potent,
the
clinical
use
of
rapamycin
is
limited
by
its
poor
drug
like

properties.
Currently
used
oral
formulations
of
rapamycin
have
low
bio-‐availability
[44]

(14
–
16%)

because
of
its
extremely
low
water
solubility
[45]
(<0.01
mg/ml).
To
improve
aqueous
solubility
and

make
rapamycin
amenable
for
delivery,
various
solubility
enhancing
agents
like
ethanol,
PEG,

poloxamer
188,
polysorbate
80,
propylene
glycol,
carnauba
wax
and
others
are
frequently
used.

These
formulations
are
associated
with
carrier
mediated
toxicity,
especially
toxic
to
liver
and

kidneys,
can
cause
hemolysis
and
hypersensitivity
reactions.
Recent
studies
also
show
that

rapamycin
and
other
mTOR
inhibitors
are
toxic
to
lungs
by
causing
interstitial
pneumonitis
[46]
.

Moreover,
rapamycin
is
known
to
readily
partition
into
erythrocytes
leading
to
reduced
drug

concentrations
at
site
of
action
[47]
.
Current
oral
formulation
of
rapamycin
suffers
from
another

disadvantage,
which
is
variable
PK
across
patients.
The
absorption
of
rapamycin
into
the
blood

stream
from
the
intestine
varies
widely
between
patients,
with
some
patients
having
up
to
eight

times
more
exposure
than
others
for
the
same
dose.
Drug
levels
are,
therefore,
taken
to
make

sure
patients
get
the
right
dose
for
their
condition.
Plasma
drug
levels
need
to
be
continuously

monitored,
especially
before
giving
the
next
dose.

It
is
because
of
these
limitations
that
different
approaches
have
been
taken
to
improve
the

formulation
and
delivery
of
rapamycin.
One
of
these
approaches
has
been
the
development
of

Temsirolimus
(CCI-‐779),
a
water-‐soluble
rapamycin
ester
that
has
demonstrated
promise
in
early

Phase-‐I
trials
[48]
.
However,
CCI-‐779
still
has
limited
solubility
in
water,
~120
μg/ml,
requiring
the

use
of
ethanol
as
a
co-‐solvent
[49]

for
IV
formulations.
Even
though,
phases
I
and
II
clinical
studies

have
reported
significant
changes
in
the
pharmacokinetic
profile
(5-‐fold
increase
in
C
max
,
5-‐fold

decrease
in
t
max
,
3-‐fold
decrease
in
t
1/2
,
and
1-‐fold
decrease
in
AUC)
[50][51]
,
high
inter-‐patient

variability
and
mild
to
moderate
side
effects
such
as
neurotropenia,
thrombocytopenia,
manic-‐
depressive
syndrome,
and
diarrhea
were
observed.

These
disadvantages
can
be
addressed
by
using
protein
nanoparticle
technology
to
improve

solubility
and
PK
of
rapamycin,
attain
a
predictable
plasma
profile
and
prevent
its
distribution

into
erythrocytes
and
other
non-‐target
tissues,
thereby
reducing
toxicity.
Biodegradable
protein

20

based
particles
prevent
vehicle
mediated
toxicity
associated
with
current
rapamycin

formulations.
To
achieve
this,
we
designed
a
library
of
FKBP-‐ELP
fusion
proteins
that
utilize
the

tight
binding
of
rapamycin
to
its
cognate
receptor
(FKBP)
to
solubilize
and
deliver
rapamycin.

2.2:
Materials
and
Methods:

FKBP-‐ELP
expression
and
purification

The
pET25b(+)
vectors
with
FKBP-‐ELP
fusion
gene
were
transfected
into
BLR
(DE3)
E.coli

competent
cells
(Novagen)
and
plated
onto
Agar
plates
with
100
µg/mL
carbenicillin
and

incubated
overnight
in
a
37°C
incubator.
5-‐6
colonies
were
picked
for
each
construct
and

evaluated
for
highest
protein
expression
by
transforming
each
colony
into
50mL
Terrific
Broth

(TB)
media
grown
overnight
supplemented
with
100
µg/mL
carbenicillin
at
37°C.
The
bacterial

culture
grown
from
each
colony
was
amplified
to
1L
TB
media
supplemented
with
100
µg/mL

carbenicillin
and
allowed
to
grow
for
24
h
at
37°C.
The
culture
was
centrifuged
and
bacterial

pellet
was
resuspended
in
Dulbecco’s
sterile
phosphate
buffered
saline
(PBS)
buffer
(Corning).

The
resuspension
was
subjected
to
tip
probe
sonication
for
cell
lysis.
The
supernatant
containing

fusion
protein
was
purified
using
Inverse
Transition
Cycling
(ITC).
The
colony
having
the
highest

protein
expression
was
purified
in
bulk
in
8-‐9
L
TB
media
with
yield
of
50-‐60
mgs/L.
The
purified

protein
was
filtered
through
200
nm
sterile
acrodisc
25
mm
filters
(Pall
Corporation)
and
protein

concentration
was
estimated
using
Beer
Lambert’s
law:

Protein
concentration
(M)
=

("
#$%
&
"
()%

)
×
,-./0-12
345016
789
×.

where
A
280

and
A
350

are
absorbance
at
280
and
350
nm
respectively,
l
is
the
path
length
(cm)
and

MEC
is
the
estimated
molar
extinction
coefficient
at
280
nm,
11585
M
-‐1
cm
-‐1

for
FKBP-‐ELP
and

20190
M
-‐1
cm
-‐1

for
FKBP-‐ELP-‐FKBP.

21

FKBP-‐ELP
physicochemical
characterization

The
purified
fusion
proteins
were
characterized
for
their
physicochemical
properties:
Transition

temperature
using
UV-‐Vis
spectroscopy,
particle
size
and
stability
using
Dynamic
Light
Scattering,

affinity
for
rapamycin
using
Isothermal
Titration
Calorimetry.
Purity
of
ELPs
was
determined
by

running
denatured
samples
on
4-‐20%
gradient
Tris-‐Glycine-‐SDS
PAGE
gel.
6-‐12
µg
protein
in

water
was
mixed
with
SDS
loading
buffer
containing
10%
β-‐mercapto
ethanol
and
heated
at
90°C

for
5
mins
before
loading
onto
the
gel.
Gels
were
stained
using
10%
w/v
copper
chloride
solution

and
imaged
using
BioRad
Gel
Imager
(Figure
2A).

ELP
transition
temperatures
were
obtained
by
measuring
optical
density
at
350nm
as
a
function

of
temperature
on
a
UV-‐Vis
DU
800
spectrophotometer.
A
temperature
ramp
was
performed
by

heating
different
concentrations
of
ELPs
(with
and
without
FKBP)
and
FKBP-‐ELPs
(with
and

without
rapamycin)
in
Beckman
Coulter
Tm
microcells
(Brea,
CA).
The
temperature
was
increased

at
rate
of
1°C/min
with
readings
taken
every
0.3°C
increment.
Temperature
corresponding
to

maximum
first
derivative
of
optical
density
was
defined
as
bulk
transition
temperature.

The
particle
size
of
all
the
FKBP-‐ELPs
was
evaluated
by
measuring
Hydrodynamic
radius
(R
h
)
using

Dynamic
Light
Scattering.
All
samples
and
microwell
plates
were
pre-‐chilled
at
4
°C
before

analysis.
25
µM
concentration
samples
were
filtered
through
200
nm
sterile
acrodisc
13
mm

filters
(Pall
Corporation)
and
60
µL
of
each
sample
was
loaded
in
triplicates
in
384
well
plate

(Greiner
Bio
One)
and
covered
with
15
µL
mineral
oil.
The
plate
was
centrifuged
at
3000
rpm
for

3
minutes
to
clear
any
surface
air
bubbles.
Samples
were
analyzed
using
Wyatt
Dynapro
plate

reader
(Santa
Barbara,
CA)
from
15-‐37°C
at
interval
of
1°C.
The
stability
of
FKBP-‐ELPs
was

evaluated
at
37°C
for
period
of
48h
after
rapamycin
encapsulation
to
study
the
effect
of

rapamycin
binding
on
the
size
and
structural
properties
of
FKBP-‐ELPs.
The
reported
values
are

presented
as
mean
±
SD.

The
affinity
of
FKBP-‐ELPs
for
rapamycin
was
studied
using
Isothermal
Titration
Calorimetry
on
a

MicroCal
PEAQ
ITC
(Malvern
Instruments,
United
Kingdom).
The
reference
cell
of
the
calorimeter

was
filled
with
water
and
all
binding
studies
were
performed
at
37
°C.
Briefly,
300
µl
of
8
µM

22

rapamycin
(PBS,
2.36%
DMSO)
was
carefully
loaded
into
the
calorimeter
cell
using
a
Hamilton

syringe,
making
sure
not
to
introduce
any
air
bubbles.
The
titration
syringe
was
filled
with
100

µM
FKBP-‐ELP
(PBS,
2.36%
DMSO).
While
the
titration
syringe
was
spinning
at
250
rpm,
FKBP-‐ELP

was
injected
into
rapamycin
12
times,
each
injection
being
3
µl,
allowing
3
minutes
between

injections
to
facilitate
equilibration.

At
the
end
of
titration,
the
calorimeter
cell
and
syringe
were

emptied,
washed
with
detergent,
water
and
dried
using
methanol
before
starting
the
next

experiment.
The
resulting
isotherm
was
fitted
to
a
binding
model
in
“Offset
mode”
using
MicroCal

PEAQ
ITC
analysis
software
(Malvern
Instruments,
United
Kingdom)
to
generate
affinity
(k
d
),

stoichiometry
of
binding,
enthalpy
of
binding
(ΔH),
entropy
of
binding
(ΔS)
and
the
Gibbs
free

energy
(ΔG).

Rapamycin
encapsulation
and
formulation
for
in
vivo
injections

Purified
FKBP-‐ELPs
were
used
for
rapamycin
encapsulation
using
two-‐phase
solvent
evaporation

method.
200-‐400
µM
(2mL)
FKBP-‐ELP
in
PBS
was
equilibrated
in
a
glass
vial
to
37°C,
followed
by

addition
of
1.1
mol
equivalent
rapamycin
in
hexane/EtOH
mixture
(7:3
v/v).
The
organic
phase

was
evaporated
under
mild
flow
of
N
2

gas
with
continuous
stirring
for
20
mins.
After
complete

evaporation
of
organic
solvent,
the
remaining
aqueous
solution
was
centrifuged
at
13000
rpm
at

37
°C
to
precipitate
and
pellet
free
unbound
rapamycin.
The
supernatant
was
added
to
a
20
kDa

MWCO
dialysis
cassette
(Thermo
Scientific)
and
was
dialyzed
against
PBS
(1:750

sample:dialyzate)
for
12
hours
to
completely
remove
unbound
rapamycin.

The
sample
after

dialysis
was
filtered
through
200
nm
sterile
acrodisc
25
mm
filters
(Pall
Corporation)
and
an

aliquot
of
filtered,
encapsulated
material
was
injected
onto
a
C-‐18
RP-‐HPLC
column
(Waters,
Inc.)

to
quantify
rapamycin
concentration.
Post
quantification,
all
drug
loaded
FKBP-‐ELP
formulations

were
diluted
to
required
dose
and
aliquots
for
single
injection
were
stored
frozen
at
-‐80°C.

Removal
and
quantification
of
bacterial
endotoxin

Pierce
High
capacity
endotoxin
removal
column
(Thermo
Scientific,
catalog
no:88276)
was

activated
according
to
the
manufacturer’s
protocol.
Rapamycin
encapsulated
FKBP-‐ELP
was

incubated
with
activated
resin
overnight
at
4
°C.
Next
day,
the
sample
was
retrieved,
an
aliquot

23

was
diluted
to
15
µM
rapamycin
and
residual
endotoxin
burden
was
estimated
using
LAL
gel
clot

assay
(Pyrotell,
Associates
of
Cape
Cod
Inc.)
following
the
manufacturer’s
protocol.
The
LAL
lysate

was
reconstituted
using
LAL
Reconstitution
buffer
(Pyrosol,
Associates
of
Cape
Cod
Inc.)
and
the

gel
clot
assays
were
performed
in
sterile
glass
tubes
(Pyrotubes,
Associates
of
Cape
Cod
Inc.)

Rapamycin
release
kinetics
from
FKBP-‐ELPs
by
dynamic
dialysis

Rapamycin
loaded
FKBP-‐ELP
was
added
to
a
20
kDa
dialysis
cassette
(Thermo
Scientific)
and
was

dialyzed
against
PBS
at
37°C
(1:750
sample:dialysate).
Penicillin-‐Streptomycin
1X
solution
was

added
to
PBS
to
prevent
bacterial
contamination.
100
µL
aliquots
were
collected
from
the

cassette
at
fixed
time
intervals
and
rapamycin
concentration
was
quantified
using
RP-‐HPLC.
%

drug
retained
versus
time
was
plotted
and
release
half-‐life
was
calculated
by
performing
non-‐
linear
regression.

Synthesis
of
FL-‐SLF

The
aniline
analog
of
SLF
(Cayman
Chemicals,
5
mg,
1
equiv.)
was
dissolved
in
500
µl

dimethylformamide
(DMF)
and
added
to
the
succinimidyl
ester
of
5-‐carboxyfluorescein(9.5
mg,

2
equiv.,
Molecular
Probes)
in
a
brown-‐glass
vial.
Triethylamine
(10
equiv.)
and
1-‐hydroxy-‐7-‐
benzotriazole
monohydrate
(HObt.H
2
O)
(20
equiv.)
were
added,
and
the
reaction
was
stirred
for

24
h
at
room
temperature.
DMF
was
blown
off
with
a
gentle
stream
of
nitrogen
over
the
reaction

mixture
overnight.
The
mixture
was
purified
using
preparative
TLC.
Presence
of
FL-‐SLF
in
reaction

mixture
was
confirmed
using
ESI-‐MS.

Cyanine5.5
labeling
of
FKBP-‐ELPs:

To
a
solution
of
FKBP-‐ELPs
in
phosphate
buffered
saline
(Corning),
three
times
stoichiometric

excess
of
Cyanine5.5
NHS
ester
(Lumiprobe,
15
mg/mL
in
DMSO)
was
added.
Following
overnight

incubation
at
4
°C,
the
reaction
mixture
was
loaded
onto
a
PD-‐10
desalting
column
(GE

Healthcare)
to
remove
unreacted
free
dye.
Fractions
containing
Cyanine5.5
labeled
FKBP-‐ELPs

were
identified
using
SDS-‐PAGE,
were
pooled
together
and
concentrated
10
times
(Amicon
Ultra,

30
kDa
MWCO).
The
concentrate
was
subjected
to
another
round
of
PD-‐10
purification
and

24

ultrafiltration
for
maximum
removal
of
free
dye.
Concentrations
of
Cyanine5.5
and
FKBP-‐ELP
in

the
purified
material
were
estimated
by
UV-‐Vis
spectroscopy
(Beckman
Coulter
DU-‐800)
using

the
following
equations.

C
Cy5.5

=
Absorbance
at
679
nm
*
dilution
factor

209000

C
FKBP-‐ELP

=
Ab
280

-‐
0.09*Ab
679

*
dilution
factor

M.E.C
FKBP-‐ELP

Labeling
efficiency
=
C
Cy5.5
/C
FKBP-‐ELP

*
100

To
assess
purity
and
efficiency
of
unreacted
dye
removal,
we
used
SDS-‐PAGE,
followed
by

fluorescent
imaging
of
the
gel.
Briefly,
10
µg
of
labeled
protein
after
purification
was
added
to

SDS
sample
buffer
(2%
SDS,
25%
glycerol,
62.5
mM
Tris–HCl,
pH
6.8).
The
denatured
samples

were
run
on
4-‐20%
gradient
Tris-‐Glycine-‐SDS
PAGE
gel
at
constant
voltage
(150
V)
for
30
minutes.

Fluorescent
images
were
acquired
on
a
Typhoon
8610
imager
(Excitation:
633nm
red
laser,

Emission
filter:
670nm
bandpass
30nm).

IR-‐800
labeling
of
FKBP-‐ELPs:

To
a
solution
of
FKBP-‐ELPs
in
phosphate
buffered
saline
(Corning),
0.25
mole
equivalents
of
IR-‐
800
NHS
ester
(LiCor
Biosciences,
20
mg/mL
in
DMSO)
was
added.
Following
overnight
incubation

at
4
°C,
the
reaction
mixture
was
transferred
to
10
kDa
MWCO
dialysis
cassette
(Thermo

Scientific)
and
was
dialyzed
extensively
against
PBS
(1:1000
sample
:
dialysate,
6
buffer
changes,

8
hours
between
buffer
change).
Purity
and
efficiency
of
free
dye
removal
were
assessed
as

discussed
above.
Concentrations
of
IR-‐800
and
FKBP-‐ELP
in
the
purified
material
were
estimated

by
UV-‐Vis
spectroscopy
(Beckman
Coulter
DU-‐800)
using
the
following
equations.

C
IR800

=
Absorbance
at
774
nm
*
dilution
factor

240000

C
FKBP-‐ELP

=
Ab
280

-‐
0.03*Ab
774

*
dilution
factor

M.E.C
FKBP-‐ELP

Labeling
efficiency
=
C
IR800
/C
FKBP-‐ELP

*
100

25

Near
infrared
imaging
of
fluorescently
labelled
FKBP-‐ELPs

All
animal
experiments
were
conducted
as
per
the
guidelines
of
the
American
Association
of

Laboratory
Animal
Care
under
an
USC
approved
protocol.
MDA-‐MB-‐468
cells
(American
Type

Tissue
Culture
Collection)
were
cultured
at
37°C
with
5%
CO
2

in
Dulbecco’s
modified
Eagle’s

medium
(DMEM)/Ham’s
F-‐12
media
(Caisson)
supplemented
with
10%
fetal
bovine
serum.
The

cells
were
sent
for
screening
major
mouse
pathogens
and
human
blood
borne
pathogens

(Charles
River)
prior
to
implantation.
A
single
injection
of
MDA-‐MB-‐468
cells
(1-‐2
x
10
6

cells
in

100
µL
fetal
bovine
serum
free
DMEM
media)
was
implanted
into
the
left
mammary
fat
pad
of
7-‐
8
weeks
old
female
nude
(nu/nu)
athymic
mice
(Harlan,
Inc.)
Tumors
were
allowed
to
grow
to
a

size
of
~
50-‐100
mm
3

and
mice
were
randomized
blindly
into
respective
groups.
Near-‐infrared

dye
labeled
FKBP-‐ELPs
were
administered
subcutaneously
in
the
right
flank
above
the
hind
leg,

or
intravenously
via
tail
vein.
Mice
were
anaesthetized
with
2%
v/v
isoflurane/oxygen
gas
prior

to
injection
of
labelled
FKBP-‐ELPs.
Whole
body
dorsal
and
ventral
scans
were
acquired
using
the

IVIS
optical
spectrum
(Perkin
Elmer)
at
0,1,2,4,8,24
and
48
h
post
injection
using
1
sec
exposure

time
and
small
binning.
Excitation
and
emission
filter
for
Cyanine5.5
were
chosen
to
be
640nm

and
700
nm
respectively.
Images
were
analyzed
using
Living
Image
®

(Perkin
Elmer)
software.

Region
of
Interest
(ROI)
was
drawn
on
tumor,
liver
and
spleen
from
ventral
scans
and
that
of
left

kidney
and
injection
site
from
dorsal
scans.
Fluorescence
from
respective
ROIs
was
quantified
in

average
radiance
efficiency
with
units
(photons/sec/cm
2
/sr)/
(µW/cm
2
)
and
plotted
after

subtracting
respective
ROI
background
intensity
at
0hr.
Organ
distribution
of
Cyanine5.5
labelled

FKBP-‐ELPs
with
respect
to
time
is
presented
as
mean
±
SD
with
n
=
4
per
group
(Figure
XYZ).
After

48h,
mice
were
euthanized
and
small
volume
of
blood
was
withdrawn
via
cardiac
puncture.
Mice

were
then
skinned
and
imaged
dorsally
and
ventrally
for
any
evidence
of
lymph
node

accumulation.
The
carcasses
were
then
dissected
and
individual
organs
along
with
blood

withdrawn
earlier
were
scanned
using
1
sec
exposure
time
with
small
binning.
The
fluorescence

from
the
dissected
organs
was
quantified
as
described
earlier
(Figure
XYZ).

26

2.3:
Results
and
discussion:

To
solubilize
and
deliver
rapamycin,
we
designed
a
FKBP-‐ELP
library
consisting
of
four
members:

FKBP-‐(VPGAG)
192

(FA),
FKBP-‐(VPGAG)
192

-‐FKBP
(FAF),
FKBP-‐(VPGSG)
48
(VPGIG)
48

(FSI),
and
FKBP-‐
(VPGVG)
48

(V48).
Based
on
previous
experience
in
the
lab,
we
expected
these
fusion
proteins
to

have
different
thermo-‐responsive
properties,
particle
sizes
and
drug
carrying
capacities.
We

characterized
and
studied
these
four
different
fusions
to
understand
how
nanoparticle

architecture
affects
PK
and
bio-‐distribution
of
particles,
and
improve
our
understanding
of
ELP

based
drug
delivery.
The
carriers
with
optimum
PK
and
bio-‐distribution
profile
can
efficiently

deliver
rapamycin.

Recombinant
expression
of
FKBP-‐ELPs:

Recombinant
expression
of
FKBP-‐ELP
library
resulted
in
yields
of
50-‐60
mg/L.
Inverse
transition

cycling
was
effective
in
purifying
ELP
fusions
from
bacterial
proteins.
Typically,
3
rounds
of
ITC

resulted
in
>99%
pure
proteins.
SDS-‐PAGE
(Figure
9A)
was
used
to
assess
purity
and
confirm

molecular
weights
of
FKBP-‐ELPs.

FKBP-‐ELP
fusion
proteins
retain
thermo-‐responsive
properties
observed
with
ELPs:

FKBP-‐ELPs
and
control
ELPs
(without
FKBP)
exhibited
concentration
dependent
phase
transition

temperature
(Figure
9B).
As
expected,
the
transition
temperature
dropped
with
increasing

solution
concentration.
We
obtained
linear
graphs
when
transition
temperature
was
plotted

against
log(concentration)
and
data
points
were
fit
to
equation
T
t

=
b
–
m
[Log
10
(concentration)].

These
phase
diagrams
help
us
understand
the
physical
state
of
an
ELP
at
a
particular
temperature

and
concentration.
At
physiological
temperature
of
37

o
C,
FA
and
FAF
remain
monomeric
in

solution
over
a
wide
range
of
concentrations
(5-‐100
µM)
while
FSI
assembles
nanoparticles.
At

37

o
C,
FV
remains
soluble
at
concentrations
lower
than
100
µM,
and
forms
a
coacervate
at
higher

concentrations.
FKBP-‐ELPs
retain
their
concentration
dependent
thermo-‐responsive
properties

even
after
encapsulating
rapamycin
(Figure
9C).

27

After
evaluating
thermal
properties,
we
measured
the
hydrodynamic
radii
(R
h
)
of
these
ELPs
using

dynamic
light
scattering
(Figure
9D).
FV,
FA
and
FAF
(at
25
µM)
exhibited
temperature

independent
(between
15
to
37

o
C)
hydrodynamic
radii
of
3.6
nm,
8
nm
and
10
nm
respectively.

On
the
other
hand,
FSI
remained
soluble
at
temperatures
below
24

o
C
(hydrodynamic
radius:
5.3

nm),
and
assembled
into
25
nm
spherical
micelles
above
the
transition
temperature.
The
renal

filtration
cutoff
for
nanoparticles
is
extensively
studied
and
is
generally
accepted
to
be
5.5
nm
[52]
.

Nanoparticles
smaller
than
the
glomerular
filtration
cutoff
are
rapidly
cleared
by
kidneys
and

would
not
make
efficient
drug
carriers.
With
a
hydrodynamic
radius
well
above
5.5
nm
at
37

o
C,

FA,
FAF,
and
FSI
can
be
expected
to
have
desirable
PK
properties
of
a
drug
carrier.
On
the
other

hand,
FV
being
smaller
than
renal
filtration
cutoff
might
get
rapidly
cleared
when
administered

in-‐vivo
and
would
not
make
a
great
carrier.
For
this
reason,
FV
was
excluded
from
the
library
and

was
not
evaluated
in-‐vivo.
Since
physical
stability
is
an
important
characteristic
of
a
stable
drug

carrier,
we
evaluated
the
stability
of
FKBP-‐ELPs
at
37

o
C.
All
FKBP-‐ELP
nanoparticles
showed
no

signs
of
aggregation
and
remained
stable
for
at
least
24
hours
(Figure
9E).

28

Figure
9:
A)
SDS-‐PAGE
gel
shoeing
purity
of
FKBP-‐ELPs
and
control
ELPs.
B)
Concentration

dependent
phase
transition
behavior
of
FKBP-‐ELPs.
C)
Concentration
dependent
phase

transition
behavior
of
rapamycin
loaded
FKBP-‐ELPs.

D)
Temperature
dependence
of
particle

size
of
FKBP-‐ELPs.
E)
Physical
stability
of
rapamycin
loaded
FKBP-‐ELPs
at
37

o
C

29

FKBP-‐ELPs
bind
to
rapamycin
with
a
low
nanomolar
dissocation
constant:

To
study
the
binding
of
rapamycin
to
FKBP-‐ELPs,
we
used
Isothermal
Titration
Calorimetry
(ITC)

to
evaluate
binding
affinity
and
thermodynamics.
ITC
is
a
powerful
technique
to
evaluate
a
wide

range
of
bio-‐molecular
interactions
with
affinities
in
low
nanomolar
to
millimolar
range.
It
works

by
directly
measuring
the
heat
absorbed/released
during
a
binding
event.
Being
a
label
free

technique,
ITC
does
not
require
any
fluorescent
tags
or
immobilization
and
measures
binding

parameters
in
native
states
of
binding
partners.
A
single
ITC
experiment
can
simultaneously

determine
binding
stoichiometry,
binding
affinity,
enthalpy
of
binding,
entropy
of
binding
and

the
binding
mechanism
(Figure
10).
Although
very
useful,
one
disadvantage
of
ITC
is
its
low

sensitivity
when
compared
to
techniques
like
fluorescence
polarization,
surface
plasmon

resonance
and
nuclear
magnetic
resonance.
During
the
experiment,
one
component
of
the

ligand-‐receptor
complex
is
titrated
into
the
other
component
and
the
incremental
heat
changes

for
each
step
of
the
titration
are
measured.
This
raw
data
is
converted
to
a
binding
isotherm
that

is
fitted
to
a
suitable
binding
model
by
non-‐linear
least
squares
fit
to
retrieve
the
desired
binding

parameters.
The
shape
of
this
binding
isotherm
depends
on
the
ratio
of
the
receptor

concentration
divided
by
the
dissociation
constant,
also
called
the
c
value.
It
is
generally
accepted

that
c
values
in
a
certain
range
(10-‐100)
provide
the
best
sigmoidal
shape
for
obtaining
reliable

Table
2:
Physicochemical
properties
of
ELP
protein
polymers
with
and
without
FKBP

Label

a
Amino
acid

sequence

Expected

MW

(kDa)

R
h

at

20
C°

(nm)

R
h

at

37
C°

(nm)

c
Slope,
m

[⁰C

Log
(µM)]

c
Intercept,
b

(⁰C)

V48
MG(VPGVG)
48
Y
19.7

7.01
48.8

FKBP-‐V48
(FV)
FKBP-‐G(VPGVG)
48
Y
31.5
3.6
3.6
10.26
56.7

S48I48
(SI)

MG(VPGSG)
48
(VPGIG)
48
Y
39.6

FKBP-‐S48I48
(FSI)
FKBP-‐G(VPGSG)
48
(VPGIG)
48
Y
51.4
5
25

A192
MG(VPGAG)
192
Y
73.5
7
7

FKBP-‐A192
(FA)
FKBP-‐G(VPGAG)
192
Y
85
8
8

FKBP-‐A192-‐FKBP

(FAF)

FKBP-‐G(VPGAG)
192
-‐FKBP
97
10
10

a
FKBP
amino
acid
sequence:

MGVQVETISPGDGRTFPKRGQTCVVHYTGMLEDGKKFDSSRDRNKPFKFMLGKQEVIRGWEEGVAQMSVGQRAKLTISPDYAYGATGH
PGIIPPHATLVFDVELLKLE

c
Phase
diagrams
were
fit
with
the
following
linear
relationship:
T
t

=
b
–
m[Log
10
(concentration)].
Mean
±
95%
CI.
R
2

=
0.99.

30

k
d

values.
However,
ligand
and
protein
solubility
can
strictly
limit
the
achievable
c
value,

particularly
for
weak
binders.

In
a
typical
ITC
experiment,
a
low
concentration
macromolecule
in
calorimeter
cell
is
titrated

against
high
concentration
ligand
in
titration
syringe.
In
the
case
of
rapamycin,
its
low
aqueous

solubility
limited
reaching
high
enough
concentrations
required
to
be
used
in
titration
syringe.

Hence
we
titrated
rapamycin
in
the
calorimeter
cell
against
FKBP-‐ELPs
in
the
titration
syringe.

The
resulting
binding
isotherms
are
shown
in
Figure
11.
As
shown
in
the
figure,
when
rapamycin

was
titrated
against
FKBP-‐ELPs,
the
heat
released
reduced
with
every
injection
as
the
titration

moved
towards
saturation.
This
is
expected
because
with
every
injection,
the
number
of
binding

events
decrease
and
hence
the
heat
released
decreases.
The
raw
data
represents
a
typical

binding
profile
seen
with
ITC.
On
the
other
hand,
when
rapamycin
was
titrated
against
A192
and

Figure
10:
Raw
data
(upper
panel)
generated

by
an
ITC
experiment
representing
the
heat

released/absorbed
during
the
duration
of

the
titration.
This
raw
data
is
converted
into

the
binding
isotherm
(below)
by
integration

of
each
injection
peak
giving
the
thermal

energy
(∆H)
of
each
titration
step.
Upon

saturation
of
the
protein
in
the
cell
with

added
ligand,
the
signal
is
reduced
until
only

the
background
heat
of
dilution
remains.

From
the
binding
isotherm
(heat
plotted

against
the
molar
ratio
of
ligand/protein),

the
change
in
enthalpy
∆H,
the

stoichiometry
n,
and
the
binding
affinity
k
d

can
be
calculated.

31

S48I48
control
ELPs,
we
could
only
detect
heat
of
dilution
that
remained
constant
with
every

injection.
The
data
obtained
from
controls
ELPs
did
not
show
any
signs
of
binding
to
rapamycin.

Figure
11:
ITC
titration
curve
and
binding
isotherm
for
FA-‐rapa
(left)
and
FAF-‐rapa
(right)

By
transforming
raw
data
to
binding
isotherms,
we
could
fit
the
data
to
a
binding
model
and

extract
binding
parameters
(Table
3).
The
binding
stoichiometry
was
close
to
1
in
case
of
FSI
and

FA.
Since
each
FAF
has
two
copies
of
FKBP,
it
can
theoretically
bind
two
rapamycin
molecules.

ITC
proved
this
to
be
true
since
FAF
Rapamycin
binding
resulted
in
a
binding
stoichiometry
close

to
2.
The
fact
that
binding
stoichiometry
was
not
an
exact
whole
number
reflects
small
errors
in

concentration
measurements
of
rapamycin
or
FKBP-‐ELP.
Another
less
likely
explanation
is
an

existence
of
partially
misfolded
FKBP-‐ELP
population
that
is
not
able
to
participate
in
binding.

0 10 20 30
11.0
11.5
12.0
12.5
13.0
TIme(min)
Power(µJ/sec)
FA-Rapa
0 1 2 3
-80
-60
-40
-20
0
Molar ratio
ΔH(kJ/mol)
0 10 20 30
10.5
11.0
11.5
12.0
12.5
13.0
TIme(min)
Power(µJ/sec)
FAF-Rapa
0.0 0.5 1.0 1.5
-150
-100
-50
0
Molar ratio
ΔH(kJ/mol)

32

Figure
11
(cont.):
ITC
titration
curve
and
binding
isotherm
for
FSI
(left)
and
titration
curves
for

A192
and
SI(right).
A192
and
SI
do
not
fit
a
binding
profile.

Binding

stoichiometry

(n)

Dissociation

constant

(nM)

Enthalpy
of

binding

(kJ/mol)

Free
energy

of
binding

(kJ/mol)

-‐TΔS

FA
0.92
±
0.05
7.42
±
2.01
-‐61.76
±
1.67
-‐48.36
±
0.70
13.4
±
1.90

FAF
1.82
±
0.10
7.04
±
1.67
-‐59.90
±
3.43
-‐48.60
±
0.69

11.3
±
3.97

FSI
0.80
±
0.07
5.59
±
2.34
-‐49.56
±
2.13
-‐49.70
±
0.10
2.11
±
0.17

FV
1.09
±
0.02
5.97
±
1.24
-‐58.53
±
2.70
-‐48.67
±
0.90
9.72
±
1.98

Table
3:
Binding
parameters
for
FKBP-‐ELP
rapamycin
binding.
All
experiments
have
been

performed
at
37

o
C
and
100
uM
ELP
was
titrated
into
8
uM
Rapamycin
(2.36%
DMSO/PBS).
All

figures
are
represented
as
Mean
±
SD
of
three
independent
experiments.

The
k
d

for
FKBP
Rapamycin
binding
was
previously
estimated
to
be
0.3
nM.
For
FKBP-‐ELP
binding

to
rapamycin,
we
derived
a
dissociation
constant
of
6-‐7
nM
(Table
2),
independent
of
ELP

composition.
Although
conjugation
of
ELP
to
FKBP
resulted
in
approximately
20-‐fold
reduction
in

0 10 20
9.5
10.0
10.5
11.0
11.5
TIme(min)
Power(µJ/sec)
FSI-Rapa
0 1 2 3
-50
-40
-30
-20
-10
0
Molar ratio
ΔH(kJ/mol)
0 10 20 30
9
10
11
TIme(min)
Power(µJ/sec)
A192-Rapa
0 5 10 15 20 25
11.0
11.1
11.2
11.3
TIme(min)
Power(µJ/sec)
SI-Rapa

33

affinity
of
FKBP
for
rapamycin,
FKBP-‐ELP/rapamycin
binding
is
still
strong
enough
for
the
purpose

of
drug
delivery.
A
high
binding
enthalpy
of
approximately
-‐60
kJ/mol
suggests
an
enthalpy
driven

binding
mechanism.
A
positive
-‐TΔS
revealed
an
entropic
cost
associated
with
FKBP-‐
ELP/rapamycin
binding.
This
is
expected
because
upon
binding,
rapamycin
is
transitioning
from

a
free,
random
state
in
solution
to
a
bound,
ordered
state,
thereby
reducing
the
entropy
of
the

system.
Nonetheless,
a
negative
gibbs
free
energy
for
all
FKBP-‐ELPs
binding
to
rapamycin

indicates
the
binding
is
energetically
favorable.

Bivalent
FAF
extends
rapamycin
release
by
30
fold
over
FA

We
studied
the
kinetics
of
rapamycin
release
from
FKBP-‐ELPs
by
dynamic
dialysis
method.

Dynamic
dialysis
is
used
to
study
protein
binding
of
small
molecules
based
on
the
determination

of
rate
of
dialysis
of
a
small
molecule
from
a
protein-‐containing
compartment.
The
method
is

based
on
the
fact
that
non-‐diffusible
protein-‐small
molecule
complexes
are
reversibly
formed
in

the
protein
compartment
and
that
the
rate
of
loss
of
small
molecule
from
that
compartment
is

directly
proportional
to
the
concentration
of
unbound
small
molecule,
provided
that
sink

conditions
are
maintained
for
the
diffusing
species,
i.e.
that
back
diffusion
into
the
protein

compartment
is
insignificant.

Rapamycin
encapsulation
had
an
efficiency
of
70-‐90%
with
FA
and
FSI,
and
~160%
with
bivalent

FAF.
Drug
loaded
FKBP-‐ELPs
were
added
to
a
dialysis
cassette
and
dialyzed
under
sink
conditions

(>5
times
the
volume
of
buffer
needed
to
saturate
the
rapamycin
in
cassette).
By
determining

total
rapamycin
concentration
in
the
cassette
at
various
time
points
of
dialysis,
%drug
retained

(vs)
time
was
plotted
and
was
fit
to
a
one-‐phase
exponential
decay
model
by
non-‐linear

regression
analysis.
With
a
half
life
of
1500
hours,
FAF
reproducibly
demonstrated
a
30-‐fold

slower
release
(Figure
12)
compared
to
FA
(release
half
life
=
46
h).

As
shown
in
figure
12,
the
dissociation
rate
constant
(k
off
)
for
FAF
is
30
fold
lower
than
FA.
This

suggests
that
FAF
might
behave
as
a
sustained
release
drug
carrier
in-‐vivo
and
extend
circulation

half
life
of
rapamycin.
On
the
other
hand,
dissociation
constant
(K
d
),
which
is
a
measure
of
affinity

34

and
is
defined
as
ratio
of
on-‐rate
(k
on
)
and
off-‐rate
(k
off
),
is
comparable
for
both
FA-‐Rapa
and
FAF-‐
Rapa
(Table
3).
Based
on
these
results,
we
hypothesized
that
FA
and
FAF
have
equal
affinity
for

rapamycin,
but
different
binding
kinetics.
To
prove
this
hypothesis,
we
are
currently
using
Surface

Plasmon
Resonance
(SPR)
to
calculate
both
on
and
off
rates
of
binding.
We
expect
to
see

compensation
in
on
and
off
rates
such
that
K
d

still
remains
the
same
for
FA-‐rapa
and
FAF-‐rapa

binding.

Figure
12:
Differential
release
kinetics

Non-‐invasive
bio-‐distribution
of
FKBP-‐ELPs
using
near
infrared
imaging:

To
evaluate
circulation,
clearance
and
bio-‐distribution
profile
of
FKBP-‐ELPs,
we
labeled
them
with

near
infrared
dye
and
tracked
in-‐vivo
non-‐invasively
using
near
infrared
imaging.
Near
infrared

(700nm–
900nm)
probes
provide
the
following
advantages
for
optical
imaging
when
compared

to
probes
that
emit
in
visible
region:

1)
NIR
light
can
penetrate
deeper
into
tissue
than
light
at
visible
wavelengths
(Figure
13),
thus

enabling
the
assessment
of
information
from
deeper
structures.

35

2)
Less
autofluorescence
is
present
at
the
NIR
compared
to
visible
wavelengths,
enabling
higher

signal-‐to-‐background
ratios.
Whenever
tissue
absorbs
light,
there
is
a
chance
that
fluorescent

light
will
be
emitted.
Figure
14
demonstrates
the
relationship
of
excitation
and
emission

wavelengths
to
tissue
autofluorescence.
Green
autofluorescence
of
the
skin
and
viscera,
and

especially
the
gallbladder,
small
intestine
and
bladder,
is
very
high
when
excited
with
blue
light.

Autofluorescence
of
the
gallbladder
and
bladder
are
greatly
reduced
using
a
‘red’
filter
set
(green

light
excitation),
but
intestinal
autofluorescence
remains
significant.
Use
of
a
NIR
filter
set

essentially
eliminates
autofluorescence.
Hence,
high
tissue
autofluorescence
precludes
the
use

of
visible
light
for
most
in
vivo
imaging
applications,
and
NIR
light
solves
this
problem
by
reducing

fluorescence
background.

FKBP-‐ELPs
were
labeled
with
Cyanine-‐5.5
using
NHS
chemistry.
With
excitation
maximum
and

emission
maximum
at
673nm
and
707nm
respectively,
and
a
high
extinction
coefficient
of

209000
M
-‐1
cm
-‐1
,
cyanine-‐5.5
is
a
widely
used
NIR
dye.
Using
NHS
chemistry,
we
non-‐specifically

labeled
side
chain
amino
and
carboxy
groups
of
FKBP-‐ELPs
with
cy-‐5.5.
Following
unreacted
dye

removal,
we
assessed
the
purity
of
labeled
conjugates
using
SDS-‐PAGE
(Figure
15)
and
verified

that
labeled
protein
contributed
to
>85%
fluorescence.
Labeling
proteins/nanoparticles
with

Figure
13:
Spectral
attenuation
of
light
transmitted

through
the
thoracic
tissue
of
a
mouse
along
the

dorsal–ventral
axis.
[ln(1/I
t
)]
is
relative
spectral

attenuation,
where
I
and
I
t

are
the
intensity
of
the

incident
and
transmitted
light
for
0.8
cm
thickness

in
a
living
mouse
(in
vivo,
solid
curve)
and
post-‐
mortem
(ex
vivo,
dashed
curve).
Attenuation
drops

substantially
at
wavelengths
longer
than
600
nm,
in

the
red
part
of
the
spectrum,
making
this
region

ideal
for
non-‐invasive
imaging.

Source:
J.
Biomed.
Opt.
13,
0440081–0440089.

36

hydrophobic
dyes
can
sometimes
drastically
alter
their
properties.
Using
dynamic
light
scattering,

we
assessed
if
dye
conjugation
was
impacting
the
size
or
stability
of
FKBP-‐ELP
particles.
It
was

observed
that
stability
of
dye
labeled
particles
was
a
function
of
labeling
efficiency.
At
37

o
C,
cy-‐
5.5
labeled
FA
and
FAF
retained
their
hydrodynamic
radii,
but
FSI
formed
stable
25
nm
particles

only
if
labeling
efficiency
was
<20%
(Figure
16),
in
other
words,
only
when
less
than
20%
of
FSI

molecules
were
labeled.
When
labeling
efficiency
was
>20%,
FSI
aggregated
to
form

microparticles
as
soon
as
warmed
to
37

o
C
(data
not
shown).

Figure
14:
(a)
Physical
location
of
organs
in
mouse
(b)
Green
autofluorescence
observed
with
a

blue
excitation/green
emission
filter
set
(c)
Reduced
autofluorescence
of
gall
bladder
and

bladder
using
red
emission
filter,
but
GI
tract
signal
is
still
strong
(d)
NIR
filter
set
has
no

autofluorescence
Source:
Current
Opinion
in
Chemical
Biology
Volume
7,
Issue
5,
October

2003,
Pages
626–634

Figure
15:
SDS-‐PAGE,
followed
by
fluorescent
imaging
of
FKBP-‐ELPs
(represented
by
FSI)

37

Mice
bearing
tumors
of
size
50-‐100
mm
3

were
injected
with
450
µM
Cy5.5
labeled
FKBP-‐ELP
(80

µM
cy5.5)
subcutaneously.
Whole
body
dorsal
and
ventral
scans
were
acquired
at
0,1,2,4,8,24

and
48
h
post
injection
(Figure
17)
using
1
sec
exposure
time
and
small
binning.
Excitation
and

emission
filter
for
Cyanine5.5
were
chosen
to
be
640
nm
and
700
nm
respectively.
As
seen
from

Fig
17,
though
our
optical
range
was
in
near-‐infrared
region,
we
observed
a
very
high
background

signal
and
it
was
not
easy
to
resolve
true
signal
of
Cy5.5
FKBP-‐ELPs
from
background
noise.
The

reason
for
high
background
was
unrefined
chlorophyll
containing
ingredients,
particularly
alfalfa

that
is
found
in
regular
laboratory
animal
diets.
Before
optical
imaging,
animals
need
to
be
shifted

to
a
low
fluorescence
alfalfa
free
diet,
but
we
failed
to
do
this.
Since
animals
were
on
regular
diet

and
chlorophyll
fluoresces
at
680
nm,
we
had
strong
background
signals
from
GI
tracts
of
animals

masking
fluorescence
from
underlying
organs,
making
data
analysis
difficult.

Nonetheless,
we
obtained
reliable
and
very
useful
information
from
optical
imaging
of
excised

organs
after
48
hrs
of
injecting
Cy5.5-‐FKBP-‐ELPs
(Figure
18).
As
expected,
excised
large
and
small

intestines
in
PBS
group
displayed
high
fluorescence,
confirming
the
source
of
background
noise

in
full
body
scans.
FKBP-‐ELPs
show
promising
tumor
localization,
with
FA
and
FAF
having
a
higher

tumor
localization
compared
to
FSI.
This
data
is
coherent
with
our
in-‐vivo
tumor
regression

studies
where
we
observed
better
efficacy
with
rapamycin
loaded
FAF
when
compared
to
FSI.

Significant
signal
from
liver
and
kidneys
suggested
these
organs
as
primary
routes
of
clearance,

a
common
observation
with
nanoparticles.
This
also
raises
the
question
of
possible
toxicity
to

liver
and
kidneys
when
rapamycin
is
delivered
using
FKBP-‐ELPs.
Additional
experiments
need
to

be
carried
out
to
answer
this.

Figure
16:
Physical
stability
of

Cyanine5.5
labeled
FKBP-‐ELPs

as
measured
by
DLS.
Data
is
plotted

as
mean
±
SD

38

Figure
18:
(A)
Representative
near
infrared
images
of
excised
organs
after
48hrs
(B)
Average

radiant
efficiency
from
excised
organs
in
FA,
FAF,
FSI
groups
(n=4)

Figure
17:
Representative
dorsal
and
ventral
scans

of
breast
cancer
xenograft
bearing
mice
injected

with
Cy5.5
FKBP-‐ELPs
at
4
hrs
and
24
hrs
post

injection.
Site
of
injection
(subcutaneous)
is

evident
in
dorsal
scans
as
the
brightest
spot
above

hind
leg.

39

References

[1]
American
Cancer
Institute:
http://www.cancer.gov/about-‐cancer/understanding/statistics

[2]
A.F.
El-‐Kattan,
C.S.
Asbill,
N.
Kim,
B.B.
Michniak,
Int.
J.Pharm.
215
(2001)
229–240

[3]
L.
Jain,
E.R.
Gardner,
W.D.
Figg,
M.S.
Chernick,
H.H.
Kong,
Pharmacotherapy
30
(2010)
52–56

[4]
S.
Hofer,
K.
Frei,
H.P.
Rutz,
Cancer
Biol.
Ther.
5
(2006)
483–484

[5]
H.S.
Friedman,
O.M.
Colvin,
S.M.
Ludeman,
S.C.
Schold
Jr.,
V.L.
Boyd,
L.H.
Mulhbaier,
D.D.

Bigner,
Cancer
Res.
46
(1986)
2827–2833

[6]
N.
Suvannang,
C.
Nantasenamat,
C.
Isarankura-‐Na-‐Ayudhya,
V.
Prachayasittikul,
Molecules

16
(2011)
3597–3617

[7]
P.
Paixão,
L.F.
Gouveia,
J.A.G.
Morais,
Int.
J.
Pharm.
429
(2012)
84–98

[8]
Longo-‐Sorbello
GS,
Bertino
JR,
Haematologica
86:121-‐27

[9]
Borst
P,
Evers
R,
Kool
M,
J.
Natl.
Cancer
Inst.
92(2012)1295-‐301

[10]
Gottesman
MM,
Fojo
T,
Bates
SE,
Nat
Rev
Cancer.
Jan
2(1)
2002:48-‐58

[11]
Speiser
PP.
Poorly
soluble
drugs:
a
challenge
in
drug
delivery.
In:
Müller
RH,
Benita
S,
Böhm

B,
editors.
Emulsions
and
nanosuspensions
for
the
formulation
of
poorly
soluble
drugs.

Medpharm
Stuttgart:
Scientific
Publishers;
1998.
pp.
15–28

[12]
Merisko-‐Liversidge
E.
Particles;
April
20–23,
2002;
Orlando,
Florida,
USA.
2002

[13]
Kim
TY,
Kim
DW,
Chung
JY,
Shin
SG,
Kim
SC,
Heo
DS,
Kim
NK,
Bang
YJ,
Clin
Cancer
Res.
2004

Jun
1;
10(11):3708-‐16.

[14]
Carmeliet
P,
Jain
RK
Nature.
2000
Sep
14;
407(6801):249-‐5

[15]
Yuan
F,
Dellian
M,
Fukumura
D,
Leunig
M,
Berk
DA,
Torchilin
VP,
Jain
RK,
Cancer
Res.
1995

Sep
1;
55(17):3752-‐6.

[16]
Yuan
F,
Leunig
M,
Huang
SK,
Berk
DA,
Papahadjopoulos
D,
Jain
RK,
Cancer
Res.
1994
Jul
1;

54(13):3352-‐6

[17]
Bae
Y,
Jang
WD,
Nishiyama
N,
Fukushima
S,
Kataoka
K.
Mol
Biosyst.
2005
Sep;1(3):242-‐50

[18]
Zhang
W,
et
al.,
Nanotechnology.
2009;20:275101

[19]
Heidel
JD,
Yu
Z,
Liu
JY,
Rele
SM,
Liang
Y,
Zeidan
RK,
Kornbrust
DJ,
Davis
ME,
Proc
Natl
Acad

Sci
U
S
A.
2007
Apr
3;
104(14):5715-‐21

40

[20]
Wang
X,
Li
J,
Wang
Y,
Cho
KJ,
Kim
G,
Gjyrezi
A,
Koenig
L,
Giannakakou
P,
Shin
HJ,
Tighiouart

M,
Nie
S,
Chen
ZG,
Shin
DM,
ACS
Nano.
2009
Oct
27;
3(10):3165-‐74

[21]
JA
MacKay,
M
Chen,
JR
McDaniel,
W
Liu,
AJ
Simnick,
A
Chilkoti,
Nature
materials
8
(12),

993-‐999

[22]
Sugano
M,
Egilmez
NK,
Yokota
SJ,
Chen
FA,
Harding
J,
Huang
SK,
Bankert
RB,
Cancer
Res.

2000
Dec
15;
60(24):6942-‐9

[23]
Sutton
D,
et
al,
Pharm.
Res.
2007;24:1029–1046

[24]
Gradishar
WJ,
et
al.,
J.
Clin.
Oncol.2005;23:7794–7803

[25]
Boddy
AV,
et
al.,
Clin.
Cancer
Res.2005;11:7834–7840

[26]
Bhatt
R,
et
al.,
J.
Med.
Chem.
2003;46:190–193

[27]
Gref
R.;
et
al,
Colloids
Surf.2000,
18,
301–313

[28]
Nagayama
S.;
Ogawara
K.;
Fukuoka
Y.;
Higaki
K.;
Kimura
T.,
Int.
J.
Pharm.
2007,
342,
215–
221

[29]
K
Matyjaszewski,
J
Spanswick,
Materials
Today,
2005,
8(3),
26–33

[30]
S.
Kim,
J.
H.
Kim,
O.
Jeon,
I.
C.
Kwon
and
K.
Park,
Eur.
J.
Pharm.
Biopharm.,
2009,
71,
420

[31]
M.
Haider,
Z.
Megeed
and
H.
Ghandehari,
J.
Controlled
Release,
2004,
95,
1

[32]
D.
W.
Urry,
K.
D.
Urry,
W.
Szaflarski
and
M.
Nowicki,
Adv.
Drug
Delivery
Rev.,
2010,
62,

1404

[33]
D.
W.
Urry,
What
Sustains
Life?
Consilient
Mechanisms
for
Protein-‐
based
Machines
and

Materials,
Springer,
Birkhauser,
LLC,
2006.

[34]
M.
R.
Banki,
L.
Feng
and
D.
W.
Wood,
Nat.
Methods,
2005,
2,
659

[35]
D.
W.
Urry,
R.
D.
Harris
and
K.
U.
Prasad,
J.
Am.
Chem.
Soc.,
1988,
110,
3303

[36]
D.
K.
Chang,
C.
M.
Venkatachalam,
K.
U.
Prasad
and
D.
W.
Urry,
J.
Biomol.
Struct.

Dyn.,1989,
6,
851

[37]
J.
A.
MacKay,
M.
Chen,
J.
R.
McDaniel,
W.
Liu,
A.
J.
Simnick
and
A.
Chilkoti,
Nat.
Mater.,

2009,
8,
993

[38]
W.
Liu,
M.
R.
Dreher,
D.
Y.
Furgeson,
K.
V.
Peixoto,
H.
Yuan,
M.
R.
Zalutsky
and
A.
Chilkoti,
J.

Controlled
Release,
2006,
116,
170

41

[39]
Yu
K1,
Toral-‐Barza
L,
Discafani
C,
Zhang
WG,
Skotnicki
J,
Frost
P,
Gibbons
JJ.
Endocr
Relat

Cancer.
2001
Sep;8(3):249-‐58

[40]
Seeliger
H1,
Guba
M,
Koehl
GE,
Doenecke
A,
Steinbauer
M,
Bruns
CJ,
Wagner
C,
Frank
E,

Jauch
KW,
Geissler
EK,
Clin
Cancer
Res.
2004
Mar
1;10(5):1843-‐52

[41]
Luan
FL1,
Ding
R,
Sharma
VK,
Chon
WJ,
Lagman
M,
Suthanthiran
M.
Kidney
Int.
2003

Mar;63(3):917-‐26

[42]
Bierer,
B.
E.,
Mattila,
P.
S.,
Standaert,
R.
F.,
Herzenberg,
L.
A.,
Burakoff,
S.
J.,
Crabtree,
G.,

Schreiber,
S.
L,
1990,
Proc
Natl
Acad
Sci
U
S
A,
87,
9231-‐5.

[43]
Bjornsti,
M.
A.
&
Houghton,
P.
J.
2004,
Nat
Rev
Cancer,
4,
335-‐48.

[44]
Buck,
Marcia
L.,
Pediatric
Pharmacotherapy
12
(2),
2006

[45]
Simamora,
P;
Alvarez,
JM;
Yalkowsky,
SH,
International
journal
of
pharmaceutics
213
(1-‐2):

25–9

[46]
Chhajed,
P.
N.,
Dickenmann,
M.,
Bubendorf,
L.,
Mayr,
M.,
Steiger,
J.,
Tamm,
M.,Respiration,

73,
367-‐74,
2006

[47]
Yatscoff,
R.
W.,
Wang,
P.,
Chan,
K.,
Hicks,
D.
&
Zimmermann,
J.
1995,
Ther
Drug
Monit,
17,

666-‐71

[48]
Temsirolimus:
CCI
779,
CCI-‐779,
cell
cycle
inhibitor-‐779.
Drugs
R
D.
2004;5:363–367.

[49]
Raymond
E,
Alexandre
J,
Faivre
S,
Vera
K,
Materman
E,
Boni
J,
Leister
C,
Korth-‐Bradley
J,

Hanauske
A,
Armand
JP,
J
Clin
Oncol.
2004
Jun
15;
22(12):2336-‐47

[50]
Hidalgo
M,
Buckner
JC,
Erlichman
C,
Pollack
MS,
Boni
JP,
Dukart
G,
Marshall
B,
Speicher
L,

Moore
L,
Rowinsky
EK,
Clin
Cancer
Res.
2006
Oct
1;
12(19):5755-‐63

[51]
Raymond
E,
Alexandre
J,
Faivre
S,
Vera
K,
Materman
E,
Boni
J,
Leister
C,
Korth-‐Bradley
J,

Hanauske
A,
Armand
JP,
J
Clin
Oncol.
2004
Jun
15;
22(12):2336-‐47

[52]
Hak
Soo
Choi,
Wenhao
Liu,
Preeti
Misra,
Eiichi
Tanaka,
John
P.
Zimmer,
Binil
Itty

Ipe,
Moungi
G.
Bawendi,
John
V.
Frangioni,
Nat
Biotechnol.
2007
Oct;
25(10):
1165–1170.

Asset Metadata

Creator Santosh, Peddi (author)

Core Title Effects of particle architecture on in-vivo pharmacokinetics and bio-distribution of therapeutic nanostructures

Contributor Electronically uploaded by the author (provenance)

School School of Pharmacy

Degree Master of Science

Degree Program Pharmaceutical Sciences

Publication Date 07/28/2016

Defense Date 06/18/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag drug delivery,elastin-like polypeptides,EPR effect,FKBP,fusion protein,nanoparticle,OAI-PMH Harvest,rapamycin

Format application/pdf (imt)

Language English

Advisor Andrew, Mackay (committee chair)

Creator Email santosh.peddi1992@gmail.com,speddi@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-284676

Unique identifier UC11279401

Identifier etd-SantoshPed-4659.pdf (filename),usctheses-c40-284676 (legacy record id)

Legacy Identifier etd-SantoshPed-4659-0.pdf

Dmrecord 284676

Document Type Thesis

Format application/pdf (imt)

Rights Santosh, Peddi

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract Small molecule chemotherapeutics, although routinely used in the clinic, suffer from poor drug-like properties, including low water solubility, rapid plasma clearance, and non-specific bio-distribution causing toxic side effects. Protein polymers of appropriate size can address these issues by solubilizing hydrophobic drugs, extending their mean plasma residence time by preventing renal filtration and promoting tumor accumulation through enhanced permeation and retention(EPR) effect, thereby reducing drug toxicity. Moreover, the delivery vehicle, being proteinaceous in nature, is bio-degradable, non-toxic and largely non-immunogenic. ❧ The clinical utility of rapamycin, a highly potent anti-cancer drug is limited by low solubility, low bioavailablity and rapid systemic clearance. To address this, we synthesized FKBP-elastin like polypeptide(ELP) fusion proteins to utilize the tight binding of rapamycin to FKBP for effective nanoparticle-based delivery. Derived from human tropoelastin, ELPs are a class of thermo-responsive protein polymers with sequence of (VPGXG)n, where X is the guest residue and n is number of repetitive units. Below a characteristic transition temperature (dictated by X and n), ELPs and ELP fusion proteins stay in solution and above the transition temperature, they self assemble nanostructures, typically micelles. To understand how polymer architecture affects carrier pharmacokinetics(PK) and bio-distribution, we recombinantly expressed a FKBP-ELP fusion library consisting of FKBP-(VPGAG)₁₉₂ (FA), FKBP-(VPGAG)₁₉₂ -FKBP (FAF), FKBP-(VPGSG)₄₈(VPGIG)₄₈ (FSI). At physiological temperature, FA and FAF are soluble monomers, while FSI phase separates into 25 nm spherical micelles decorated with FKBP on the corona. For non-invasive PK and bio-distribution evaluation, we labeled these drug carriers with IR800 without any effect on their physico-chemical properties and physical stability. In a breast cancer xenograft mouse model, we seek to understand how nanoparticle architecture and route of administration (intravenous vs. subcutaneous) affect their circulation, clearance and bio-distribution profile.