References - Masarykova univerzita

shootceaselessUrban and Civil

Nov 16, 2013 (3 years and 8 months ago)

117 views

MASARYKOVA

UNIVERZITA

P
ŘÍRODOVĚDECKÁ FAKULT
A

ÚSTAV BIOCHEMIE








Diplomová práce
















Brno 2013

Andrej Bešše


MASARYKOVA

UNIVERZITA

P
ŘÍRODOVĚDECKÁ FAKULT
A

ÚSTAV BIOCHEMIE







Studium molekulární podstaty chemo
-

a
radiorezistence na základě komplexní
molekulárně
-
genetické charakterizace
tkáně glioblastomu a jeho recidivy


Diplomová práce




Andrej Bešše







Vedoucí práce:
D
oc. RNDr. Ondřej Slabý, Ph.D.

Brno
,

2013

Bibliografický záznam

Autor:

Bc. Andrej Bešše

Přírodovědecká fakulta, Masarykova univerzita

Ústav Biochemie

Název práce:

Studium molekulární podstaty chemo
-

a
radiorezistence na základě komplexní molekulárně
-
genetické charakterizace tkáně glioblastomu a jeho
recidivy

Studijní program:

Biochemie

Studijní obor:

Biochemie

Vedoucí práce:

d
oc.
RNDr. Ondřej Slabý, Ph.D.

Akademický rok:

2013

Počet stran:

77

Klíčová slova:

glioblastom, rezistence, miRNA


Bibliographic Entry


Author

Bc. Andrej Bešše

Faculty of Science, Masaryk University

Department
of

Biochemistry

Title of Thesis:

Study of chemo
-

and radio
-
resistance molecular
basis through complex molecular
-
genetic
characterization of paired primary and relapsed
glioblastoma tissues

Degree programme:

Biochemistry

Field of Study:

Biochemistry

Supervisor:

d
oc.
RNDr. Ondřej Slabý, Ph.D.

Academic Year:

2013

Number of Pages:

77

Keywords:

glioblastoma, resistance, miRNA




Abstrakt

Glioblastoma multiforme (GBM) je nejčastější a nejmalignější nádor mozku s
astrogliální

diferenciací. GBM tvoří 15 až 20 % všech primárních intrakraniálních
nádorů při incidenci 3
-

4 nové případy na 100 000 obyvatel. I přes včasné
stanovení diagnózy, radikální resekci primárního GBM, konkomitantní
radiochemoterapii a adjuvantní chemoterapii

s temozolomidem (TMZ) je
prognóza pacientů s GBM pouze 12
-

15 měsíců od stanovení diagnózy. Cílem
této diplomové práce je prokázat souvislost mezi mikroRNA (miRNA) a
rezistencí GBM k

léčbě
,

a to na základě

miRNA profilování

párových vzorků
primárních a r
ekurentních GBM pomocí expresního miRNA profilování. Dalším
cílem je funkční charakterizace deregulovaných miRNA pomocí
in vitro

funkčních testů na stabilních buněčných liniích primárního GBM.




Abstract

Glioblastoma multiforme (GBM) is one of the most co
mmon and most
malignant brain tumors with astroglial origin. GBM accounts for 15
-
20% of all
primary intracranial tumors with incidence of 3
-
4 new cases per 100

000
individuals. Despite the early diagnosis, radical resection, concomitant
radiochemotherapy a
nd adjuvant chemotherapy with temozolomide (TMZ) of
primary GBM, the prognosis of patients remains only 12
-
15 months after
diagnosis. The aim of this diploma thesis is to determine the relationship
between microRNA (miRNA) and GBM resistance to the treatme
nt on the basis
of
miRNA profiling

of paired samples of primary and recurrent GBM
s

using
miRNA expression profiling. The next aim is functional characterization of
deregulated miRNA
s

by
in vitro

functional analyses o
n stable cell lines derived
from primary

GBM tumors
.







Poděkování

Na tomto místě bych chtěl

poděkovat
svému školiteli
doc. RNDr. Ondřeji
Slabému,
Ph.D. za odborné vedení a pomoc při realizaci této práce, Mgr. Jiřímu
Šánovi za odborné konzultace, cenné rady a pomoc při plánování

experimentální části,

Mgr.
Lence Kubiczkové za pomoc při sepisování práce a
Zdeňce Kosařové za skvělou spolupráci
.











Prohlášení

Prohlašuji, že jsem svoji diplomovou práci vypracoval samostatně s využitím
informačních zdrojů, které jsou v práci cito
vány.



Brno 9. května
=


=
………………………………
=
Andrej Bešše
=
8


Contents

List of abbreviations

................................
................................
.......................

11

Introduction

................................
................................
................................
.....

13

1

Theoretical part and main issues

................................
............................

14

1.1

Glioblastoma multifome

................................
................................
.......

14

1.1.1

Epidemiology

................................
................................
...............................

14

1.1.2

Etiology

................................
................................
................................
........

15

1.1.3

Diagnosi
s

................................
................................
................................
......

16

1.1.4

Therapy

................................
................................
................................
........

17

1.1.5

Prognosis

................................
................................
................................
......

18

1.2

Molecular biology of GBM

................................
................................
....

18

1.2.1

EGFR/PTEN/AKT/mTOR Pathway
................................
................................
.

19

1.2.2

TP53/MDM2/p14ARF Pathway
................................
................................
....

20

1.2.3

p16INK4a/RB1 Pathway

................................
................................
...............

20

1.3

MiRNA

................................
................................
................................
....

20

1.3.1

MiRNAs involved in radioresistance

................................
............................

23

1.3.2

MiRNAs involved in chemoresistance

................................
.........................

28

2

Objectives

................................
................................
................................
.

32

3

Material and M
ethods

................................
................................
...............

33

3.1

Chemicals

................................
................................
..............................

33

3.2

Instruments

................................
................................
...........................

34

3.3

Patients

................................
................................
................................
..

34

3.4

Isolation from PFFE

................................
................................
..............

35

3.5

Isolation of RNA enriched of miRNAs

................................
.................

36

3.6

TaqMan Low Density Arrays (TLDA)

................................
...................

37

3.6.1

Reverse transcription for TLDA

................................
................................
....

37

3.6.2

Quantitative PCR (qPCR)

................................
................................
..............

38

9


3.7

Analysis of miRNA arrays

................................
................................
....

39

3.7.1

miRNA specific qRT
-
PCR

................................
................................
..............

39

3.8

In vitro

functional studies

................................
................................
....

42

3.8.1

In vitro

cultivation

................................
................................
........................

42

3.8.2

Cell lines

................................
................................
................................
.......

43

3.8.3

Treatment

................................
................................
................................
....

43

3.8.4

Transfection

................................
................................
................................
.

43

3.8.5

MTT viability assay

................................
................................
.......................

44

3.8.6

Cell cycle

................................
................................
................................
......

45

3.8.7

Apoptosis

................................
................................
................................
.....

45

3.8.8

Statistical evaluation

................................
................................
....................

46

4

Results

................................
................................
................................
......

47

4.1

MiRNA qRT
-
PCR/ TLDA TaqMan Low Density Arrays

.......................

47

4.2

MiRNA validation

................................
................................
..................

47

4.3

Transfection of miRNAs

................................
................................
.......

48

4.4

Treatment

................................
................................
..............................

49

4.5

Cell viabi
lity

................................
................................
...........................

49

4.5.1

Radiation treatment

................................
................................
....................

49

4.5.2

TMZ treatment

................................
................................
.............................

51

4.6

Cell

cycle

................................
................................
...............................

52

4.6.1

Radiation treatment

................................
................................
....................

52

4.6.2

TMZ treatment

................................
................................
.............................

53

4.7

Apoptosis

................................
................................
..............................

55

4.7.1

Radiation

................................
................................
................................
......

55

4.7.2

TMZ treatment

................................
................................
.............................

57

5

Discussion

................................
................................
................................

60

10


Souhrn

................................
................................
................................
.............

65

Summary

................................
................................
................................
..........

66

References

................................
................................
................................
.......

67




















11


List of
abbreviations

5
-
FU

5
-
Fluorouracil

ABCG2

ATP
-
binding cassette, sub
-
family G member 2

ACF1

A
TP
-
Utilizing Chromatin Assembly And Remodeling Factor 1

ATM

Ataxia telangiectasia
mutated

BAD

Bcl
-
2 associated death promoter

BCL
-
2

B
-
cell leukemia/lymphoma 2

BRCA1

Breast cancer 1

BRD7

B
romodomain containing 7

CCND3

Cyclin

D3

CDK4

Cyclin
-
dependent kinase 4

CDK6

Cyclin
-
dependent kinase 6

cDNA

Complementary DNA

C
T

C
ycl
e threshold

CUL5


Cullin 5

DNA

Deoxyribonucleic acid

DNA
-
PKs

DNA
-
dependent protein kinases

E2F1

E2F transcription factor 1

E2F3

E2F transcription factor 3

EDTA

Ethylenediaminetetraacetic acid

EGFR

Epidermal growth factor receptor

EPHX1

Epoxide hydrolase 1

FBS

Fetal bovine serum

FFPE

Formalin
-
Fixed Paraffin
-
Embedded

FGFR2

Fibroblast growth factor receptor 2

FHIT

Target fragile histidine triad

GABRA1

Gamma
-
aminobutyric acid A receptor

GalNT7

N
-
acetylgalactosaminyltransferase 7

GBAS

Glioblastoma Amplified Sequence

GSK3B

Glycogen synthase kinase 3

GTP

Guanosine triphosphate

IDH
-
1

Isocitrate dehydrogenase
-
1

IGFR1

Insulin
-
like growth factor t
ype 1 receptor

12


LOH

Loss of heterozygosity

LRRFIP1

Lucine rich repeat (in FLII) interacting protein 1

MAPK

Mitogen
-
activated protein kinase

MGMT

Methylguanine methyltransferase

miRISC

miRNA
-
induced silencing complex

miRNA

microRNA

MMR

Mismatch repair

mRNA

Messenger ribonucleic acid

MTAP

Methylthioadenosine phosphorylase

mTOR

Mammalian target of rapamycin

MXI1

MAX
-
interacting protein 1

NF1

Neurofibromatosis 1

NFκB

Nuclear factor kappa
-
B

NHEJ

Non
-
homologous end joining

PI3K

Phosphoinositide 3
-
kinase

PIP2

Phosphatidylinositol 4,5
-
bisphosphate

PIP3

Phosphatidylinositol 3,4,5
-
trisphosphate

pre
-
miRNA

Precursor miRNA

pri
-
miRNA

Primary miRNA

PTEN

Phosphatase and tensin homolog

RB

Retinoblastoma 1

RNA

Ribonucleic acid

RT

Reverse transcription

SIRT1

Silent mating type information regulation 2
homolog 1

SMARCA5

SWI/SNF related, matrix associated, actin dependent regulator of
chromatin, subfamily a, member 5

SWI/SNF

SWItch/Sucrose Non Fermentable

TLDA

TaqMan Low Density Arrays

TNF
-
α

Tumor necrosis factor
-
alpha

TRBP

Trans
-
activator RNA binding protein

VEGFR

Vascular Endothelial Growth Factor Receptor

WDR11

WD repeat
-
containing protein 11

XRCC

X
-
ray cross
-
complementary genes


13


Introduction

Tumors of central nervous system (CNS) are a heterogeneous group that
contains both curable tumors
,

such as meduloblastoma
,

and tumors with very
poor prognosis
,

such as
high
-
grade
gliomas

(HGG)
.
HGG

constitute
of more than
50% of brain tumor

diseases. The most frequent form of gliomas originating from
astrocytomas is glioblastoma multiforme (GBM). Although the incidence of GBM
is not high, tumo
rs are extremely aggressive

and

patients

without any ther
apy die
within 3
-
6 months. During the past years, only modest improvements were done
in the treatment of GBM
, c
urrent standard therapy,
which
include

maximal
surgical resection followed

by

concomitant chemoradiotherapy with temozolomide
(TMZ)

and adjuvant TMZ
,

can extend median of overall survival only
approximately to 12
-
15 months from diagnosis.
One of the reasons of short
survival might be presence of resistant tumor cells
to the indicated therapy.

With the discovery of RNA interferen
ce an
d microRNAs (miRNAs),
increased evidence suggested contribution of

mi
RNAs

in molecular biology of
tumors. MiRNAs are small non
-
coding RNAs with a length of approximately 22
nucleotides, which are able to post
-
transcriptionally regulate gene expr
ession.
The
y were shown to
re
gulate multiple processes in

cells, such as apoptosis,
proliferation, angiogenesis, differentiation, etc. So far, a lot of papers have been
published about miRNA profiling in diffe
rent solid tumors and their deregulated
levels

in tumors w
hen compared to healthy tissues. Depending on the mRNA
target which
miRNAs
bind and regulate
, they

can act
either as

oncogenes or
as
tumor suppressors.

The aim of this diploma thesis

is to define the role of miRNAs in the
resistance of GBM and the respons
e of the tumor cells to the treatment. This was
done using microRNA arrays profiling where initial tumors were compared with
their relapses, followed by validation of differentially expressed miRNA
s

on bigger
cohort of patients and miRNA
s

functional analys
es.






14


1

Theoretical part and main issues

1.1

Glioblastoma multifome


Glioblastoma multiforme (GBM) belongs to
the largest group of
primary
intracranial neoplasms known as gliomas
, which account
approximately
for
50%
of all primary brain tumors. World Health
Organization (WHO) classifies gliomas
into 4 grades (I
-
IV) (Table I
) from which
GBM represent
s

the most aggressive
form
(
1
)
.

GBM tumors

are classified into primary and secondary subtypes, 95% of
GBMs belong to the primary tumors developing
de novo
. Second
ary GBMs arise
and progress from lower grade gliomas
, most frequent
ly from astrocytomas
(
2
)
.



Table I
:
Classification of primary brain tumors and their incidence in population
(adapted from CBRUS 2012)


1.1.1

Epidemiology

GBM represent up to 60% of infiltrative gliomas
(
3
)
. Primary GBM

develop in
older patients (50
-
55 years), while pati
ents with secondary GBM

are mostly
affected in younger age, about 40 years old
(
4
)
. The ratio of mortality (6:100 000)
and incidence (7.8:100 000) in Europe show similar patterns for both genders
(EUCAN 2
012). Occurrence of GBM differs also across worldwide distribution.
Most affected are Caucasian followed by Asian and African born indiv
iduals (6.78,
4.94 and 3.84/100
000

respectively
) (Figure 1)
(
5
)
.

Grade

Histology

% Brain tumors

Incidence

(per
100

000)

I

Pilocytic astrocytoma

1
.
7

0.
33

II

Oligodendroglioma,
diffuse astrocytomas

1.
9

0.
36

III

Anaplastic astrocytoma

2.
1

0
.
4

IV

Glioblastoma
multiforme

17
.
1

3
.
17

15





Figure 1:
Proportional distribution of worldwide incidence of GBM.


1.1.2

Etiology

As in other types of cancer, there are risk factors that might contribute to the
occurrence of GBM. The risk factors can be divided in to two groups: factors
closely conducted to the envir
onment (sporadic factors) and factors associated
with family case history (hereditary factors).


Sporadic

Gliomas with sporadic origin emerge more often as hereditary; however,
their cause still remain unexplained in most of the cases
(
6
)
. Although the
environmental influence has been published across the whole spectrum of
diseases, the possible contribution of environment to GBM development is
hindered by many factors, such as relative rarity of the disease and rapid death of
patients. So
me studies have shown a slightly elevated risk of gliomas in ‘white
-
collar’ workers, including social science, financial workers, managers and people
with higher socioeconomic status
(
7, 8
)
.

Likewise in many others cancers, especially cancers of digestive

tracts,
gliomas could be linked with a diet as well. Higher intake of cored meat,
processed pork and fried bacon which contains nitrate/nitrite and secondary
amines compounds can elevate the risk of glioma development in adults and in
16


prenatal development
. On contrary, vitamins C and E are able to inhibit the
formation of nitroso compounds from their precursors
(
9
-
11
)
. Several studies
have shown increased risk of glioma development in patients who received polio
vaccines contaminated with SV40 virus
(
12
)
.
Moreover, cytomegalovirus has
been found in
high grade gliomas (
HGG
)

and also validated it contribution to GBM
development
(
13, 14
)
.

Although the known risk factors are multifactorial, one factor was proved to
be most significant


radiation.
Radiation is
the only accepted and validated risk
factor in glioma, which was confirmed to increase
the
risk about 22 times
(
12
)
.
This was supported by a study with children who received radiotherapy. In this
group, the occurrence of brain tumors was much higher, 12.8%
, than in patients
of the same study who did not receive radiotherapy, 0%
(
15
)
. In the last few
years, there were plenty of medial discussions about the relationship between
exposure to electromagnetic fields, cell
-
phones, wireless radio devices and
possible risk of glioma development. However, the validation of such data was
unsucc
essful
(
16
-
18
)
.


Hereditary

Studies focused on families with GBM incidence did not validate a
heritability of GBM; it was shown that the pattern of tumor inheritance was not
clear because of variable incidence of tumors through generations and time.
Moreo
ver, the diagnosis in
-
between parent
-
child generation was unsuccessful,
because the children were often diagnosed with the tumor before the adults in the
time of observation
(
19
)
.


1.1.3

Diagnosis

The symptoms of brain tumors are either produced by tumor mass i
n the
brain (edema, swelling) or by infiltration and damage of normal tissue.
The most
characteristic symptoms are
increased intracranial pressure
, weakness, loss of
somatosensory and
visual

perception
, headache
s
, vomiting and focal or
generalized epilepti
c seizure
s. However, they

are present only
in 7% of patients
with GBM and

arise in
-
between a year from the time of diagnosis
(
20, 21
)
.
17


Headache occurs approximately in 50% of patients with brain tumors
(
21
)
. Up to
this day, neither any option is available
for early detection, prevention, or
screening of gliomas, nor any specific GBM markers. The investigation of gliomas
generally consists of topographic studies, which involve magnetic resonance
imaging (MRI) and computerized tomography (CT). Their advantage
s are higher
sensitivity, soft
-
tissue contrast and better depiction of the extents of tumors
(
22
)

.

Development of new diagnostic methods, such as magnetic resonance
(
MR
)

proton spectroscopy, enables better distinction of neoplastic from non
-
neoplastic
bra
in lesions
(
23
)
.
As the main problems

in selection of the appropriate treatment
are the heterogeneity and infiltrat
ive origin of brain tumors,
imaging is an integral
component for clinical evaluation, diagnosis, treatment and therapy response.
After
intravenous application o
f contrast agent,

increased cellularity, foca
l
necrosis and edema can be observed

(
24
)
. These advance techniques are useful
for preoperative localization at the time of diagnosis and during follow up of
patients.


1.1.4

Therapy

Surgical
intervention with maximally safe tumor tissue resection is a
standard procedure in GBM treatment. This is the first line of therapy, which is
generally followed by
concomitant
radiotherapy and chemotherapy
(
25
)
.
Radiotherapy is employed to dispose the rest

of tumor cells and thereby increase
the treatment efficacy
.
Unfortunately, chemotherapy in gliomas is difficult because
of the presence of blood
-
brain barrier, which doesn’t allow for most of
medicaments to get into the brain from the blood vessels
(
26, 2
7
)
. The most
commonly used chemotherapeutic agent is
temozolomide (
TMZ
)
, an alkylating
agent that damages the DNA and triggers tumor cells apoptosis
(
28
)
. TMZ

can
prolong the median progression
-
free survival time from 13 weeks to 21 weeks
(
29
)
.

However, s
urvival rate varies from patient to patient, this

means that the
subsequent treatment after diagnosis has to be highly personalized
(
30
)
.


18


1.1.5

Prognosis

As was mentioned above, t
he current standard of care

in GBM

is based on
maximal
surgical resection followed by a

concomitant chemoradiotherapy with
temozolo
mide (TMZ) and adjuvant TMZ.

Despite such aggressive treatment,

and
advances in medicine,

recurrence is
still
near
-
universal phenomenon and
prognosis remains poor
(
31, 32
)
.

Median survival of GBM patients without any therapy is 3
-
6 months.
Therapy containing combination of treatment with TMZ and radiotherapy can
extend the median survival up to 12
-
15 months.
Nevertheless, several GBM
cases (3
-
5%)

with the survival rate

more t
han 3 years

have been described

(
30
)
.


1.2

Molecular biology of GBM

Using c
onventio
nal molecular methods,

several genetic alterations

have
been found
, which are involved in initiation and progression of GB
M development.
To the most frequent genetic alterations belong amplification of
EGFR
, mutation
of
p53
, loss of heterozygosity (LOH) or deletion on chromosome 10q
,

where
tumor
-
suppressor genes such as
PTEN

are

located
,

and

mutations of isocitrate
dehydrogenase
-
1
(IDH
-
1
).
From
epigenetic alteration
s, frequent is
methylation of
the methylguanine methyltransferase (
MGMT
) gene promoter
(
33, 34
)
.

During last few years many papers were published about several
constitutive abnormalities that influence glioma genesis. These abnormalities can
be manifested as mutations in genes responsible for regulation of apoptosis,
proliferation or DNA repair
(
35
)
.
Among genes ensuring apoptosis regulation
belong i.e. neurofibromatosis 1 (
NF1
) and
p53
, genes responsible for DNA repair


mismatch repair (
MMR
) genes, x
-
ray cross
-
complementary genes (
XRCC
) and
MGMT

genes. These genes and their products belong to ATM and

PI3K signaling
pathways that are often deregulated in gliomas
(
35
-
37
)
.

Genetic alterations, which contribute
to
glioma development, differ in
primary and secondary gliomas.
Primary GBMs are often characterized by
Epidermal Growth Factor Receptor (
EGFR
)

and
MDM2

amplification/overexpression,
PTEN

mutations, and
p16

deletions
,

whereas
secondary GBMs frequently contain
p53

mutations. Loss of heterozygosity (LOH)
19


is the most frequent genetic a
lteration in GBM. It is estimated that there is

75% of
LOH on 10q
, 47% on 9p, 29% on 19q and 19% on 1p in primary GBM
(
38
)
.
Particular genomic alterations are fundamental to both formation and malignant
progression of GBM.
Broad genetic screens have shown that
genetic aberrations
are spread across the entire genome and
affect nearly all chromosomes. The
most altered regions have been found to be on chromosome 1p, 7, 8q, 9p, 10,
12q, 13q, 19q, 20 and 22q. These regions contain genes with putative synergistic,
tumor promoting function, such
as EGFR, MYC, MDM2, GBAS, PTEN,
CDK6,
MTAP, IGFR1, MXI1, WDR11
and

FGFR
2
(
39
)
. These genes belong to frequently
deregulated signaling pathways which play a role in GBM pathology and thus are
described below.


1.2.1

EGFR/PTEN/AKT
/mTOR p
athway

The characterist
ic chromosomal aberration is

loss of

chromosome 10 or

its
long arm, where lies
a gene for tumor
-
suppressive protein
PTEN
. LOH 10q
occurs at similar frequencies in primary and secondary GBMs
(
40
)
.

Gains of gene material due to genetic alterations have also been
demonstrated, but they occur less frequently than genetic losses. The most
common event is
gain of the entire chromosome 7 with 7p12
region amplification
that
occurs

in 40% of primary GBMs but rarely in secondary GBM
(
41
)
. This
region contains several genes related to gliomagenesis, such as
EGFR

or
GBAS

(
Glioblastoma Amplified Sequence).


Growth factor pathways and their components
, such as epidermal and
vascula
r epithelial growth factor receptors (EGFR, VEGFR) play a significant role
in glioma cell proliferation, migration, and neovascularization
(
31, 42
)
.

Amplification and overexpression of the

EGFR

gene occur

only rarely in
secondary GBM, while in primary is the incidence much higher
,

about 34
-
50%
(
43
)
.
Previously mentioned genetic alterations can contribute to the induction of
glial transformation and progression. It causes activation of downstream signaling
lipid kinase PI3K/Akt and MAPK pathways
(
44, 45
)
.

Activation can also occur in a
ligand
-
independent fashion in case of
EGFR

mutations or high
-
level wild
-
type
receptor overexpression. This activation supports signal
-
transduction pathways,
which lead to tra
nscriptional activation of genes involved in cell division, tumor
20


invasion, promotion of angiogenesis and resistance to apoptosis
(
46
)
. The
majority of GBMs with
EGFR

amplification also contain the mutant
EGFR

gene,
EGFRvIII
, which carries deletion from ex
on 2 to 7
(
47, 48
)
. This mutation has
been shown to be responsible for radiation resistance
(
49
)
. Receptors activate
downstream effectors molecules such as AKT (protein kinase B) and mTOR, the
mammalian target of rapamycin. This leads to cell proliferation

and increased cell
surviv
al by blocked

apoptosis. Moreover, loss of
PTEN

does not inhibit the PIP3
signal and in that way cell can continuously proliferate. Response to EGFR kinase
inhibitors requires presence of both EGFRvIII and PTEN
(
50
)
.

1.2.2

TP53/MDM2/p14ARF p
athway

The p53 pathway is involved in controlling the cell cycle. Secondary GBMs
have higher incidence of
p53

mutations (>65%) than primary GBMs. MDM2
amplification has been observed in up to 10% of GBMs, mostly in primary GBMs
that lack
p53
mutations
(
51
)
. Loss of
p14

ARF

expression due to homozygous
deletion or promoter methylation of
p14
ARF
gene has appeared in about 76% of
GBMs. In addition, in secondary GBMs promoter methylations are more frequent
(
52
)
.

1.2.3

p16INK4a/RB1
p
athway

The p16IN
K4a/RB1 seems to be important in both types of GBM. Promoter
methylation of the
RB1
gene was significantly more frequent in secondary (43%)
than in primary GBMs (14%)
(
53
)
.


1.3

MiRNA


MicroRNAs (miRNAs) are a novel class of endogenously produced short
(about
22 nt) non
-
coding RNAs. The process of miRNA biogenesis starts in the
nucleus. Some genes encoding miRNAs have their own promoters and are
transcribed by RNA polymerase II to pri
-
miRNAs. Pri
-
miRNAs are then modified
and spliced to short
,

cca 70 nucleotides

long
,

hairpin structures called pre
-
miRNAs. The splicing is processed by microprocessor complex consisting of
ribonuclease Drosha and protein Pasha that binds to pre
-
miRNA. The export from
nucleus to the cytoplasm is mediated by GTP dependent transport co
mplex,
21


Exportin 5/ RanGTP. Exportin 5 recognizes characteristic structure of pre
-
miRNA,
which is 2
-
nucleotides overlap on 3´ end of pre
-
miRNAs
(
54
)
. Then, pre
-
miRNAs
is spliced by Dicer and TRBP (trans
-
activator RNA binding protein) to duplex of
leading st
rand (miRNA) and

passenger strand (miRNA*). After that
, leading or
guide strand represented by miRNA is incorporated to the miRISC (miRNA
-
induced silencing complex) and passenger strand, miRNA*, is usually degraded
(
55
)
.

The main domains of RISC complex constitute of proteins from the
argonaute family Ago2 with catalytic enzymes of RNA interference Gemin2 and
Gemin3 (Figure 2)
(
56
)
. These enzymes are able to degrade RNA with bounded
miRNA and the interaction between miRNA
and mRNA is processed on
polyribosomes. The main advantage of miRNA is their negative regulation of gene
expression at the posttranscription
al level by interaction with 3`UTR

of miRNAs
target mRNA
(
57
)
.

There are two possibilities how to decrease the prot
ein level after miRNA
binding to mRNA on the basis of complementar
it
y. If there is

only partial
complementarity between miRNA and

mRNA, it leads to inhibition of translation,
whereby the level of mRNA remains unchanged
(
58
)
. H
owever, if there is a
complete

homology between miRNA and mRNA target, mRNA is
degraded

(
59
)
.

As the binding of miRNA to

its target site does not require a perfect
complem
entarity of nucleotides,

one miRNA can regulate more target genes as
well as one gene can be regulated by several m
iRNAs.
Recently, number of
studies focused on a relationship between miRNAs and signaling pathways
involved
in GBM pathogenesis
(
60
)
.

22



Figure 2:
Canonical pathway of miRNA maturation and posttranscriptional
regulation of mRNA

(adapted from Slabý et al.)
(
61
)
.

23


1.3.1

MiRNAs involved in radioresistance

The goal of radiotherapy is to eliminate as many of the remaining cells as
possible or

arrest the remaining tumor cells in a non
-
invading, sleeping state for
as long as possible. Ionizing radiation (IR) causes water

ionization within the cells
and so gives rise to production of reactive radicals, which subsequently interact
with DNA and disrupt the phosphate DNA backbone. DNA strand breaks caused
by this interaction can be either repaired or can lead to cell cycle ar
rest.
Depending on the response to the therapy, we can observe a long
-
term effect of
IR which is manifested as senescence of the tumor cel
ls, or a short term effect
observed as

cell death via apoptosis
(
62
)
. Relapse of the tumor after radiotherapy
is commo
n and then, tumor often progresses into mor
e aggressive form

associated with poor prognosis and resistance to further treatments
(
32
)
. It was
previously described in several studies that IR triggers DNA repair mechanisms
and activa
tes several signaling pat
hways,
such as PI3K/AKT or ATM/Chk2/
p53

that subsequently lead to higher proliferation, invasivity and survival of glioma
cells (Figure 3)
(
36, 37
)
.
In the past several years, there has been published an
increasing number of studies describing miRNAs as i
mportant regulators of
different pathways involved in development of gliomas as well as their role in the
resistance to the treatment.


24



Figure 3:

MiRNA involved in regulation of pathways connected with
radioresistance

(adapted from
Besse at al
.)
(
63
)
.


PI3K/AKT/mTOR pathway

Deregulation of PI3K/AKT/mTOR pathway occurs frequently in GBM
(
64
)
.
Physiologically the
PI3K signaling pathway
is

critical for normal brain
development; however, it has also been found to be hyperactivated in brain
tumors
(
65, 66
)
.
EGFR

plays an important role as an activator of this pathway
,

since mutations in
EGFR
lead to tumor cell proliferation, increased survival,
angiogenesis and metastasis. Activation of EGFR allows for phosphorylation of
PIP2 via PI3K to PIP3. Subsequently, P
IP3 activates AKT, an important regulator
of various processes in the cell. Downstream signaling components of AKT
belong to regulators of the cell cycle i.e. glycogen synthase kinase 3 (GSK3B),
proliferation
-

nuclear factor kappa
-
B (NFκB) or mammalian ta
rget of rapamycin
(mTOR) and apoptosis (BAD, MDM2, Caspase9)
(
66
)
. As indicated, a
ctivation
of

AKT

can

result in cell growth and cell survival which are closely linked to
treatment resistance. These capabilities make AKT an attractive therapeutic
target.

25


O
ne of the ways of targeting the
PI3K/AKT pathway and its downstream
components for intervention is offered by

miRNAs. Study of miRNA expression
profiles after IR exposure in the U87MG glioblastoma cell line showed
downregulation of miR
-
181a. Based on the c
omputational analysis and predictive
tools
,

a downstream protein of AKT signaling, BCL
-
2, was chosen as a direct
target for further validation. Transient overexpression of miR
-
181a sensitized the
U87MG cell line to the IR and led to downregulation of mRNA
and protein level of
BCL
-
2 which is not only associated with radioresistance but also plays a
protective role against apoptotic cell death and is frequently overexpressed in
human tumor cells
(
67, 68
)
. Another miRNA involved in the AKT signaling is miR
-
21, which is generally classified as an oncomiR. MiR
-
21 was one of the first
identified miRNAs to play an important role in glioma pathogenesis with an
antiapoptotic effect on HGGs. Computational analysi
s revealed phosphatase and
tensin homolog (
PTEN
), a direct negative regulator of AKT, as a target gene of
miR
-
21. However, a
fter downregulation of miR
-
21 the expression of EGFR and
BCL
-
2 was decreased, AKT and cyclin D were activated, but the status of PTE
N
appeared to be independent of miR
-
21
(
69
)
. In another study
,

PTEN was found to
be regulated directly by miR
-
26a
(
70
)
. MiR
-
26a alone is able to transform cells by
promoting GBM cells growth
in vitro

and
in vivo.

Cell growth is enhanced either by
decreased

PTEN, RB1, or

MAP3K2/MEKK2 protein levels
, which subsequently
leads to increased AKT activation and promotes proliferation, or by decrease of c
-
JUN N
-
terminal kinase
-
dependent apoptosis
(
71
)
.

Additional miRNAs linked to AKT regulation
are miR
-
7

and miR
-
451
(
72,
73
)
. The

involvement of miR
-
7 in this pathway was evaluated on U251 and U87
glioma cell lines. Ectopic overexpression of

miR
-
7

attenuated EGFR and AKT
expression and radio
-
sensitized both glioma cell lines. Further, it prolonged
radiation
-
induced γH2AX foci formation and led to downregulation of DNA
-
dependent protein kinases (DNA
-
PKs). This indicates a relationship between
delayed

DNA damage repair and increased radiation
-
induced cell killing by
overexpression of miR
-
7, which leads to downregulation of the EGFR
-
AKT
signaling network
(
37
)
. AKT also plays an important role in DNA repair after IR
through controlled DNA
-
PK expression.
After irradiation, recognition of DNA
double
-
strand breaks during non
-
homologous end joining (
NHEJ)

is carried out by
the heterodimers Ku70/Ku80, which are activated by the DNA
-
PKs
(
74, 75
)
.
It
26


seems that miR
-
7 could be a useful therapeutic target for over
coming the radio
-
resistance of human cancers with activated EGFR
-
PI3K
-
AKT signaling
(
37
)
.



ATM/Chk2/p53 pathway

Loss of
ATM/Chk2/p53 pathway components accelerates glioma
development and contributes to radiation resistance. In response to IR cells
activate the sensor kinases ATM, ATR and DNA
-
PKs that phosphorylate multiple
downstream mediators, including the checkpoint kina
ses Chk1 and Chk2, which
lead to checkpoint initiation in cell
-
cycle and/or to apoptosis
(
36
)
. A lower level of
ATM was observed in the M059J radiosensitive cell line when compared to the
M059K radioresistant cell line due to deficiency in DNA
-
PK. This eff
ect might be
caused by overexpression of miR
-
100, which was predicted to be a direct
regulator

of ATM
(
76
)
.

After IR of both M059K and M059J cell lines, several
miRNAs were upregulated: miR
-
17
-
3p, miR
-
17
-
5p, miR
-
19a, miR
-
19b, miR
-
142
-
3p, and miR
-
142
-
5p. Mo
reover, miR
-
15a, miR
-
16, miR
-
21, miR
-
143, and miR
-
155
were found to be upregulated only in the M059K cell line with normal DNA
-
PK
activity
(
77
)
. These findings are supported by a study focused on lymphoblastoid
cells, where low
-
dosage IR also caused upregu
lation of miR
-
15a, miR
-
16 and
miR
-
21 in cells with a normal status of p53
(
78
)
. Furthermore, miR
-
143 was found
to directly target fragile histidine triad (FHIT), which is often downregulated in
epithelial tumors. Cells with homozygous deletion of FHIT show

higher resistance
to multi
-
DNA damage inducers, including IR. Overexpression of miR
-
143 protects
cells from DNA damage
-
induced killing by downregulation of FHIT expression and
leads to significant G2
-
phase arrest
(
79
)
. Increased level of miR
-
155 was also
found in lung cancer cells. This miRNA protected the cells against IR and
inhibition of this miRNA led to sensitization of cells to radiation
(
80
)
. Overall, these
observations point to the involvement of miRNAs in the different responses of
GBM cells to tr
eatment by IR, which are often seen in brain tumors, especially in
HGGs
(
78
)
.

MiR
-
101 is another miRNA associated with the protein levels of ATM and
DNA
-
PK in the U87MGD cell line. Upregulation of miR
-
101 by lentiviral

transduction sensitized tumor cells to radiation
in vitro

and

in vivo
(
81
)
. These
27


observations are supported by a study on lung cancer cell lines, which
demonstrated that ectopic expression of miR
-
101 could sensitize human tumor
cells to radiation by targeting ATM and DNA
-
PK
(
82
)
.

For repair of double
-
strand breaks nascent after irradia
tion, chromatin
remodeling protein complexes are required. MiR
-
99 expression was described to
correlate with sensitivity to IR as it targets the SWI
/
SNF chromatin remodeling
factor SNF2H
/
SMARCA5, a component of the ACF1 complex, which plays an
important ro
le in double
-
strand break repair
(
83
)
. Moreover, it has been
elucidated that reduction of BRCA1 level at the DNA damage site was the result
of downregulation

of SNF2H, which was caused by induction of miR
-
99a and
miR
-
100. These observations were further supported by an experiment where
ectopic expression of the

miR
-
99

family

in cells reduced the rate of overall
efficiency of repair by both homologous recombina
tion and non
-
homologous end
joining
(
83, 84
)
.


ATM encodes a protein kinase that acts as a tumor suppressor. This
kinase is activated by IR damage of DNA, it stimulates DNA repair and blocks cell
cycle progression. One of the mechanisms of ATM function is
by ATM dependent
phosphorylation of p53, which either arrests the cell cycle at a restriction point to
allow for the DNA damage repair or leads to the apoptosis of damaged cells. The
p53 is indeed a central regulator of cell response to stress and it has t
o be tightly
regulated
(
85
)
. Bioinformatics suggested that miR
-
125b is a negative regulator of
p53
-
induced apoptosis during cell stress
(
86
)
. Likewise, miR
-
34a acts as a tumor
suppressor in p53
-
mutant glioma cells U251. Overexpression of miR
-
34a, which
is
transcriptionally activated through p53, led to cell growth inhibition, cell cycle
arrest in G0
-
G1, induction of apoptosis and significantly reduced migration and
invasion capabilities. Such events could also be due to regulation of SIRT1, which
is predict
ed to be a direct target of miR
-
34a
(
87
)
. This is also supported by a
study, where high dosage of IR lead to induction of miR
-
34 and reduced the p53
expression level
(
88
)
. Effects of miRNA modulation on radioresistance are
summarized in Table II.




28


Table
II:

Effect of miRNA modulation on radioresistance

(adapted from
Besse at
al
.)
(
63
)
.

miRNA

Up
-
/down
-

regulation

Effect on
radioresistance

Direct and indirect
targets

References

miR
-
181a



+

BCL
-
2

(
68, 89
)

miR
-
7



+

EGFR, AKT, DNA
-
PKs

(
37
)

miR
-
143



+

FHIT

(
90, 91
)

miR155



+

GABRA1

(
92, 93
)

miR
-
101



+

ATM,DNA
-
PK

(
94
)

miR
-
99



-

SNF2H/SMARCA5

(
95
)

miR
-
100



-

SNF2H/SMARCA5

(
84
)

mi
R
-
34a



+

p53,SIRT

(
88, 96
)

miR
-
17
-
3p

IR induction


GalNT7,vimentin,MDM2

(
90, 97, 98
)

miR
-
17
-
5p

IR induction


E2F1,
PTEN

(
90, 97, 98
)

miR
-
19a

IR induction


PTEN,TNF
-
α

(
90, 99
-
101
)

miR
-
19b

IR induction


CUL5


(
90, 99
)

miR
-
142
-
3p

IR induction



(
90
)

miR
-
142
-
5p

IR induction



(
90
)


1.3.2

MiRNAs involved in chemoresistance

Temozolomide (TMZ) is a commonly used alkylating agent in the treatment
of gliomas; however, after prolonged exposure a development of resistance is
frequently observed. Unfortunately, there is little known about the signaling
pathways affected by TMZ
(
102
)
. Cytotoxicity of TMZ is based on methylation of
guanines in DNA at the O
6

position and also on the

b
ase
-
pair mismatch with
thymine. Glioma cells
have frequently
activated repair mechanisms to overcome
this substitution through DNA repair
enzyme MGMT

and
become resistant to TMZ
exposure
(
103
)
. In non
-
resistant glioma cells, TMZ induces cell death partly by
the fact that the MGMT promoter is hypermethylated
(
104
)
. Interestingly, miR
-
181
family members have a capacity to regulate MGMT and so increase the
che
mosensitivity to TMZ. Silencing of the miR
-
181 family led to increased levels
of MGMT, suggesting that the miR
-
181 family could be involved in
post
t
ranscriptional regulation of MGMT and could also be used as a predictive
factor of the response to TMZ
(
105,

106
)
. An experiment with the chemoresistant
glioma cell line U118 indicated a functional relationship between TMZ and
survival pathways PI3K/AKT and ERK1/2 MAPK. These pathways are activated
29


during treatment with TMZ; however, after inhibition of these pa
thways the
chemoresistance was partially eradicated

(Figure 4)
(
102
)
.





Figure 4:
MiRNA
s

involved in regulation of pathways connected with
chemoresistace

(adapted from
Besse at al
.)
(
63
)
.


As mentioned above, many studies
have
described the overexpression of
miR
-
21 in gliomas
(
107
-
112
)
. Downregulation of miR
-
21 in GBM cells led to
repression of cell growth, increased cellular apoptosis and cell
-
cycle arrest, which
might enhance the chemotherapeutic effect in cancer

therapy
. Overexpression of
miR
-
21 in glioma cell line U87 protected the cells against TMZ induced apoptosis
via decreasing the BAX/BCL
-
2 ratio and Caspase
-
3 activity, thereby contributing
to chemoresistance
(
109, 110
)
. An opposite effect was observed for miR
-
221/
222
downregulation in the same cell line, where it caused an increase of sensitivity to
TMZ and apoptosis, independently of p53 status. Simultaneously an increase in
expression of apoptotic factors BAX, cytochrome c, and cleaved
-
caspase
-
3 was
observed
(
113
)
.
In the TMZ resistant GBM cell line U251R, the expression levels
of miR
-
195, miR
-
455
-
3p, and miR
-
10a(*) were increased in comparison to the
30


U251MG TMZ sensitive cell line. Inhibition of miR
-
455
-
3p or miR
-
10a(*) showed a
modest cell killing effect in the
presence of TMZ compared to the suppression of
miR
-
195 which in combination with TMZ strongly enhanced cell death
(
114
)
. An
inverse correlation was found between E2F transcription factor 3 (E2F3) and
cyclin D3 (CCND3) protein levels and expression of miR
-
1
95 in glioma cells,
suggesting that these proteins might be direct targets of miR
-
195. Downregulation
of E2F3 can suppress the transcription of cell cycle related genes and cause
accumulation of cells in G1 phase and inhibition of cell cycle progression. A
s for
CCND3, its suppression mediated by miR
-
195 can upregulate the expression of
cytoplasmic
p27
Kip1
,which

negatively

regulates

G1

phase

and

so

affects

the

expression

of

other

proteins

involved

in

cell

migration,

resulting

in

repression

of

GBM

cells

inva
sion

(
115
)
.


Besides TMZ, there are other therapeutic agents used in the treatment of
gliomas, such as VM
-
26 (Teniposide). MiR
-
21 was found to mediate
chemoresistance

to VM
-
26 in U373MG cells, where suppression of miR
-
21 led to
enhancement of cytotoxicity of VM
-
26.
LRRFIP1

was confirmed and validated as
a direct target of miR
-
21, whose product is an inhibitor of NF
-
κB factor, a
downstream effector of PI3K/AKT signaling

(
112
)
. In another study, an anti
-
sense
-
miR
-
21 oligonucleotide significantly improved the effect of 5
-
FU on U251
glioma cells and increased apoptosis with a simultaneous decrease of migration
ability
(
108
)
.

There

are

likely

other

mechanisms

which

are

involved

in

the

chemoresistance

of

glioma

cells.

One

of

these

mechanisms

is

based

on

the

activity

of

proteins

from

ATP
-
binding cassettes family, which are able
to transport
various molecules across cellular membr
anes, i.e. they are able
to export anti
-
tumor drugs or their metabolites from cells
(
116
)
.
ATP
-
binding cassette, a sub
-
family of G member 2 protein family (ABCG2), is highly expressed in GBM and is
associated with multi
-
drug

resistance. MiR
-
328 contributes to the
chemosensitivity by decrease of ABCG2 expression. Thus, inhibition of ABCG2
by miRNA can lead to better treatment outcomes for patients
(
117
)
. Effects of
miRNA modulation on radioresistance are summarized in Table III
.




31


Table III:

Effect of miRNA modulation on chemoresistance

(adapted from
Besse
at al
.)
(
63
)
.

miRNA

Up
-
/down
-

regulation

Effect on
radioresistance

Direct and indirect
targets

References

miR
-
181



-

MGMT

(
105, 118
)

miR
-
21



-

BAX,BCL
-
2

(
119, 120
)

miR
-
221/222



-

BAX, cyt c, Caspase3

(
121
)

miR
-
195



+

E2F3,CCND3

(
122, 123
)

miR
-
455
-
3p



+


(
122
)

miR
-
10a(*)



+

EPHX1,BRD7

(
122
)

miR
-
328



-

ABCG2

(
124
)























32


2

Objectives


1)

miRNA expression profiling,

comparison of

pair
ed

samples of patient
s

with
GBM (initial vs. recurrent)

2)

on the basis of miRNA profiling, validate
selected miRNAs
on bigger cohort

of patients

3)

in vitro

functional analyse
s of selected miRNAs on GBM cell lines




33


3

Material and Methods

3.1

Chemicals



96% ethanol (La
chema, Czech Republic)



Hanks Balanced Salt Solution, H6648 (Sigma
-
Aldrich Inc., Saint Louis,
MO, USA)



Propidium iodide ≥94% (HPLC),
81845

(Sigma
-
Aldrich Inc., Saint Louis,
MO, USA)



Dulbecco’s phosphate buffered saline (PBS), D8537 (Sigma
-
Aldrich Inc.,

Saint Louis, MO, USA)



Dulbecco’s modified eagle medium (DMEM) 1×, Cat. No. 41966
(Invitrogen, Carlsbad, CA, USA)



0.25% Trypsin
-
EDTA 1×, Cat. 25200 (Invitrogen, Carlsbad, CA, USA)



Fetal bovine serum (BIOCHROM AG, Berlin, Germany)



OPTI
-
MEM® Reduced Serum Med
ium 1×, Cat. 31985 (Invitrogen,
Carlsbad, CA, USA)



Lipofectamine® RNAiMAX
, Cat. 11668 (Invitrogen, Carlsbad, CA, USA)



Trypan Blue Stain 0.4%, Cat. 15250 (Invitrogen, Carlsbad, CA, USA)



DEPC
-
Treated water, Cat. AM9906 (Ambion INC, Austin, Texas)



Acid Phenol
: CHCl
3

, Cat
. No.

AM720 (Ambion INC, Austin, Texas)



MTT Formazan powder
,

M2003

(Sigma
-
Aldrich Inc., Saint Louis, MO,
USA)



Annexin V : PE Apoptosis Detection Kit I

(BD Bioscience, CA, USA)



mirVana
TM

miRNA Isolation Kit (Ambion INC, Austin, Texas)



Proteinas
e K,
P6556

(Sigma
-
Aldrich Inc., Saint Louis
, MO, USA)



TaqMan® Array Human MicroRNA Set Cards v3.0, Cat.

No. 4444913

(Applied Biosystems, CA, USA)



TaqMan® 2× Universal PCR Master Mix, No AmpErase® UNG, 5 ml

(Applied Biosystems, CA, USA)



TaqMan® MicroRNA Re
verse Transcription Kit, 200 Reactions (Applied
Biosystems, CA, USA)



TaqMan® MicroRNA Assays


hsa
-
miR
-
34a, hsa
-
miR
-
338
-
5p, hsa
-
miR
-
1274a and RNU48 (Applied Biosystems, CA, USA)

34




Pre
-
miR
TM

miRNA Precursor Molecules


pre
-
miR
-
34a, pre
-
miR
-
338
-
5p,
and pre
-
miR
-
Negative Control #1 (Ambion INC, Austin, Texas)


3.2

Instruments



Microscope OLYMPUS CKX41 (OLYMPUS Europa GMBH, Hamburg,
Germany)



Centrifuge IEC CL30R (Thermo Fisher Scientific Inc., Waltham, MA, USA)



Minicentrifuge Eppendorf (Hamburg, Germany)



Bio
-
Rad

T100 Thermal Cycler
(BIO
-
RAD, USA)



NanoDrop® ND
-
1000 (Thermo Fisher Scientific Inc., Wilmington, DE, USA)



StepOne™ Real
-
Time PCR
System (Applied Biosystems, CA, USA)



Real
-
Time 7900 HT

Real
-
Time PCR System (Applied Biosystems, CA,
USA)



Synergy HT Multi
-
Mode Microplate Reader (BIO
-
TEK, Winooski, VT, USA)
+ KC4TM Data Analysis Software vs. 3.4



Applied Biosystems Sequence Detection System ABI 7500 (Applied
Biosystems, CA, USA)



BD FACS

Canto II

(BD Bioscience, CA, USA)



OGL
-
1 VF
γ
-
radiant (
Če
rná Hora, Czech Republic)


3.3

Patients

For this study
,

13 pair
ed GBM samples (initial tumor sample
vs

recurrent
tumor sample)
were
compared
.

All patients underwent
surgical
resection followed
by radiotherapy (2 Gy per fraction for 6 weeks, total dose of 60 Gy
) and
concomitant chemotherapy with TMZ (75 mg

m
-
2

daily, for 6 weeks)
.


Three samples were used for miRNA
TLDA
arrays and 10 pairs were used
for validation of chosen miRNAs

by qRT
-
PCR
.
F
ormalin
-
fixed paraffin
-
embedded
(FFPE)

s
amples were obtained from Ma
saryk Memorial Cancer

Institute, Brno,
Czech Republic
and Hradec Kralove Faculty Hospital, Czech Republic,
only after
signing an informed consent form

(Table IV).



35


Table IV:
Characteristics of patients involved in study


Total (n=13)

%

Age



<50

5

39

>50

8

61

Gender



Male

7

54

Female

6

46

Karnofsky p
erformance status


0 and 1

13

100

2

0

0

Extent of resection


Total

2

15

Subtotal

9

70

Partial

2

15

Adjuvant TMZ


Yes

4

30

No

9

70


3.4

Isolation from PFFE

Isolation of RNA enriched for miRNAs
from FF
P
E

paired samples was carried
out according to followed protocol.


1.

Place 4µm paraffin sample to 1.5 ml micro tube and homogenize by a
needle

2.

Add 1 ml of Xylene and
twist

for 5´

3.

Incubate 3´ at 50°C

4.

Spin 2´/ max. rpm, discard supernatant

5.

Add 1 ml of a
bsolute ethanol, vortex

6.

Spin 2´/ max. rpm, discard supernatant

7.

Repeat step 5 and 6

8.

Place microtube to prewarmed thermoblock for removing residual ethanol

36


9.

Add 100 µl of buffer (Table.V) adjusted to pH8, 16 µl of 10% SDS and 40
µl of proteinase K (20 mg

l
-
1
)

(Sigma)

10.

Vortex and shortly spin

11.

Incubate
for
3 hours at 55°C


Table V:

Buffer for proteinase K

Reagents

Quantity

NaCl

0.292 g

Tris

0.06 g

EDTA

0.466 g

Nuclease
-
free water

50 ml


3.5

Isolation of RNA enriched of miRNAs



MiRNAs were isolated us
ing
mirVana
TM

miRNA Isolation Kit b
y
Acid
Phenol:
chloroform

extraction.


1.

Add 600 µl of Lysis Binding
Buffer
to a microtube with sample

2.

Add 100 µl of microRNA homogenate additive, gently vortex

3.

Incubate 10´ on ice

4.

Add
600 µl of A
cid

Ph
enol
:

chloroform, vortex

5.

Spin
5‘/ max. rpm

6.

Transfer upper water phase to new 2ml microtube

7.

Add 1.25 volume of absolute ethanol

8.

Prepare new microtube with inserted column and transfer 700 µl of solution

9.

Spin 15‘‘/ 10000 rpm, discard supernatant

10.

If needed
, repeat step 8 and 9

till all s
olution is filtered

11.

Add 700 µl of Wash Solution 1

12.

Spin 15‘‘/ 10000 rpm, discard supernatant

13.

Add 500 µl od Wash Solution 2/3

14.

Spin 15‘‘/ 10000 rpm, discard the supernatant

15.

Repeat steps 11
-
14

16.

Spin 1‘/ 10000 rpm

17.

Transfer column to a new microtube

37


18.

Add 30 µl

of pre
-
warmed 95°C Elution Solution

19.

Spin 30‘‘/ max. rpm

20.

Concentration and purity of miRNA was m
easure
d

by
NanoDrop® ND
-
1000

and store
d

at
-
80°C.


3.6

TaqMan Low Density Arrays

(
TLDA
)

TaqMan® Array Human MicroRNA Set Cards v3.0

contains

2 platforms A
and B, which allows to study 667 human miRNAs in one run for each sample.


3.6.1

Reverse transcription for TLDA

Megaplex
TM

RT (Human pool A and B) stem loop primers from TaqMan®
Array Human MicroRNA Set v3.0 were used. Each pool contains 368 RT pr
imers
for cDNA synthesis of 667 human miRNAs

and controls

in 2 reactions (Figure 5)
.


1.

Prepare MasterMix for RT

primers

for pool A and B according (Table VI),
vortex and shortly spin

2.

Add 3 µl of (350 ng) total RNA to 8 tube strips

3.

Add 4.5 µl of RT master mi
x, vortex and spin

4.

Incubate 5´ on ice

5.

Perform RT in thermocycler according (Table VII)

6.

Store cDNA at
-
20°C


Figure 5:
Specific TaqMan stem loop primers (adapted from Slabý et al.)
(
61
)
.

38


Table VI.

Composition of RT mixture for TLDA

Reverse transcrtiption

components

Volume per

sample [μl]

jegap汥lTj⁒ ⁰物me牳
N0×F

0⸸0

d乔ms⁷楴á

d呔m
 00jF

0⸲0

MultiScribeTM Reverse Transcriptase (50 U·μl
-
1
)

1.50

10× RT pufr

0.80

MgCl2 (25 mM)

0.90

RNase Inhibitor (20 U·μl
-
1
)

0.10

Nuclease
-
free water

0.20

T
otal

4.50


Table VII.
Th
ermocycler conditions

of RT for TLDA

RT cycles

Temperature

Time

40 cycles

16 °C



42 °C



50 °C

1´´

hold

85 °C



hold

4

°C


=
=
P⸶KO
=
兵an瑩瑡瑩íe⁐䍒
=m䍒F
=
=

=
m牥ra牥=m䍒=ja獴e爠jáx=⡍j⤠楮áN⸵ml=m楣牯瑵be
=
a捣o牤楮g=⡔able=sf䥉⤬F
vo牴rx⁡nd⁳=o牴ró⁳=楮
=

=
Add=U⁸‱
MM=‘氠映jj⁴o⁴he⁤e獫⁁=o爠䈠景爠r
m䍒=
=

=
䍥C瑲í晵ge‱4MM⁲=m⼠PxN´
=

=
pea氠lhe=de獫sand⁣=琠íhe=f楬汩ng⁨o汥l
=

=
創o= qm䍒
=
and= éer景牭= ana汹獥猠 on= TVMM䡔= se牳ron= OKP= pequen捥=
䑥ae捴楯n⁓ó獴sms⁵náve牳慬⁴he牭o=é牯f楬e
qab汥l䥘F
K
=
=
=
=
=
39


Table VIII:

Composition of qPCR mixture for TLDA

Master mix reagents

Volume

per 1 plate [
µl
]

2x
Universal PCR
Master Mix,
No AmpErase® UNG

450

MegaplexTM RT product

6

Nuclease
-
free water

444

Total

900


Table IX.

Thermocycler conditi
ons of qPCR for TLDA

Cycles

Temperature

Time

hold

50 °C



hold

95 °C

10´

40 cycles

95 °C

30´´

60 °C




3.7

Analysis of miRNA arrays

Samples were analyzed
using

SDS RQ manager. For normalization of data,
a ΔCT method was used. All data were normalized to CT 40, because of the lack
of suitable endogenous control. For normalization, following equation
s were

used:

ΔCT
(target gene)
= CT
(target gene)


40

RQ*
(tar
get gene)

= 2
-
ΔCT
(target gene)

*Relative quantity

3.7.1

miRNA specific qRT
-
PCR

Reverse transcription using miRNA specific RT stem loop primers.

1.

Prepare MasterMix according

to

(Table X)

2.

Add 3.
3 µl of (
6.
66 ng
·μl
-
1
) total
miRNA/RNA

to 8 tube strip

3.

Add 4.
67 µl of
RT master mix

4.

Add 2 µl miRNA specific stem loop primer, vortex and spin

5.

Incubate 5´ on ice

6.

Perform r
everse transcription in thermocycler according Table XI

7.

Store cDNA at
-
20°C

40


Table X:
Composition of RT mixture

Master mix for RT

Volume per sample [μl]

1
00mM dNTPs (with dTTP)

0.1
0

MultiScribe Reverse Transcriptase, 50 U

ul
-
1

0.67

10x Reverse Transcription Buffer

1
.00

Rnase Inhibitor, 20 U

uL
-
1

0.13

Nuclease
-
free water

2.77

Total

4.67


Table XI:

Thermocycler conditio
ns of RT

Cycles

Temperature

Time

hold

16 °C

30´

hold

42 °C

30´

hold

85 °C



hold

4
°C


=
=
兵an瑩瑡瑩íe⁐䍒
=

=
é牥ra牥‴Uⁱm䍒é瑩ía氠le汬⁰污le
=

=
Add†OM=‘氠l映m楸f⁲=agen瑳ía捣o牤楮g⁴o
呡b汥lu䥉F
=

=
pé楮á4U⁷e汬⁰污le
=

=
創oⁱm䍒Ⱐ
u獩湧
=
瑨e牭oérof楬
e
=
Eqab汥⁘䥉䤩
=

=
Ana汹ze⁲=獵汴猠l獩湧=
p瑥élne⁳of瑷a牥r
v
O⸳
䅰é汩ed⁂楯獹獴sm猩
=
=
Table XII:

Composition of qPCR mixture

Reagents

Volume
per sample
[μl]

2x
TaqMan universal MM
no UNG

10
.0

Nuclease
-
free water

7.6

Probe

1
.0

cDNA (0.
66 ng

µl
-
1
)

1.4

T
otal

20
.0



41


Table XIII.
Thermocycler

conditions

of qPCR

cycles

Temperature

time

hold

50 °C



hold

95 °C

10´

40 cycles

95 °C

15´´

60 °C




Samples were run in duplicates. For normalization of data

for validation phase,

2
-
ΔCT

method was used
, for in vitro experiments
2
-
ΔΔCT
. Expression

of patients
samples were normalized to C
T

40. Expression of cell lines were normalized to
endogenous control RNU44,
using
following equation
s:

ΔC
T(target gene)

= C
T(target gene)

-

C
T(endogenous control)

ΔΔC
T(target gene/test sample A)

= C
T(target
gene/test sample A)

-

C
T(target gene/calibrator)

RQ*
(target gene)

= 2
-
ΔΔCT
(target gene)

*Relative quantity












42


3.8

In vitro

functional studies

Functional analyses of chosen miRNAs in GBM

cell lines
were performed using
workflow described in (Figure 6).
All experiments were performed in three
independent experiments.


Figure 6:
Workflow of
in vitro

functional analyse
s

3.8.1

In vitro

c
ultivation

For

in vitro

experiments,
GBMs cell lines A172, T98G and U87MG were
used
.
Cell lines were obtained from ATCC and grow
n in

culture medium DMEM
(Sigma) supplemented with 10% heat
-
inactivated fetal bovine serum (FBS), 1%
nonessential amino acids, 2 mM L
-
glutamine, 4.5 g/l D
-
glucose, 110 mg/l sodium
pyruvate, 100 µg/ml streptomycin and 100 U/ml penicillin. Cells were cultiva
ted at
37°C in humidified atmosphere containing 5% CO
2
.


43


3.8.2

Cell lines

A172

-

derived

from a GBM of 53 year old male, non
-
tumorigenic

-

doubling time 37.5 hours

T98G

-

derived

from a GBM of 61 year old Caucasian male, non
-
tumorigenic

-

doubling time 34.5
hours

U87MG

-

derived

from a GBM of 44 year old Caucasian male, tumorigenic

-

doubling time 24 hours


3.8.3

Treatment

Radiation

As a source of radiation,
Cs
-
137 γ
-
radiation (2 Gy

min
-
1
)

was used. The
suitable dosage of radiation was evaluated by counting the cells after 24, 48, 72
and 96 hours post radiation, treated with different dosages of radiation. The used
dosages of radiation were 0, 2, 5, 10, 15 and 20 Gy.

Temozolomide

Temozolomid
e was dissolved in DMSO to 25 mM stock soluti
on. To
examine exact dosage of t
emozolomide, a scale up of different concentrations
was performed 0.1; 1; 10; 100; 500 µM, where DMSO was used as a control. The
most suitable concentration for further experiment
s was chosen.


3.8.4

Transfection

The GBM cell lines were transfected by oligo
nucleotides pre
-
miR
-
34a and
pre
-
miR
-
338
-
5p
using
Lipofectamine® RNAiMAX
. Cells were seeded 24 hours
before transfection

and
transfected with 10 nM pre
-
miRNA or negative control
(Mock)
. E
f
fectivity of transfection was evaluated using qRT
-
PCR 24 hour after
transfection. For transfection of cells seeded on 60 mm dishes
, the

amount of
reagents
was
multiplied by 10.


44


Day before transfection

Seed the cells on
24 well plates in concentration

2.5 x 10
4

cells/ 500 μl of
DMEM
,

10% FBS medium without antibiotics.


Day of transfection

1.

Prepare transfection mix

2.

Add 0.25 µl
of
Lipofectamine RNAi Max to 50ul of OptiMem in microtube

3.

Add 0.12 µl (50µM) of miRNA to the 50 µl of OptiMem in microtube

4.

Incub
ate 5´

5.

Transform 50 µl of OptiMem with miRNA to the 50 µl of OptiMem with
Lipofectamine and mix gently

6.

Incubate 20´/RT

7.

Add 100 µl of solution Lipofectamine+miRNA+Optimem per well

8.

Change the medium after 24 hours


3.8.5

MTT viability assay

The proliferation of
GBM cell lines A17
2, T98G and U87MG was

examined by
MTT

proliferation assay. Cells were seeded at 25
000/well into 24
-
well plates and
incubated overnight, then transfected with miRNAs mimics (10 nM) and negative
control miRNA (10 nM), respectively. After 24

hours of incubation, medium was
changed with 500 µl

of

fresh medium supplemented with 10% FBS

and
antibiotics. Cells were the
n treated either with radiation or with TMZ. Prolifer
ation
of cells was evaluated after following

24, 48, 72 and 96 hours
. P
lates
were read
on Synergy HT Multi
-
Mode Microplate reader at a wavelength of 570 nm.


1.

Add 50 µl of MTT solution (5mg

ml
-
1
) per well

2.

Incubate for 1 hour, 37°C and 5% CO
2

3.

Discard medium

4.

Dissolve MTT in 500 µl of DMSO

5.

Measure absorbance at 570nm on spectrophotomet
er


45


3.8.6

Cell cycle

Cells were cultivated on 60 mm dish
es. The distribution of cells in

specific
cell cycle stages was evaluated by assessment of DNA content by flow cytometry.
Cells were stained with PI (propidi
um iodide) and analyzed on flow cytometer

based on
DNA conte
nt. Results

were analyzed usi
ng the FlowJo software 7.2.2.


1.

Harvest
the
cell
s

by using trypsin a
nd

inactivate by medium DMEM with
10% FBS

2.

Add
cells
suspension to flow cytometry tube and spin 5´/ 1500

rpm

3.

Add 1

ml of 70% cold ethanol

4.

Incubate 30´
on ice

5.

Spin 5´/ 1500

rpm, discard supernatant

6.

Wash with PBS, spin 5´/1500

rpm

7.

Add 250 µl PBS

8.

Add 50 µl RNAse A 0.
1mg

ml
-
1

9.

Incubate 10´ 37°C

10.

Add 10 µl PI (1mg

ml
-
1
), incubate 10´/RT

11.

Incubate 20´ on ice

12.

Measure on flow cytometer FACS Canto II

13.

Analyze by

software FlowJo v 7.2.2


3.8.7

Apoptosis

Cells were cultivated on 60 mm dishes.


1.

Harvest the cells using trypsin and inactivate by medium DMEM 10%FBS

2.

spin 5´/1500

rpm

3.

Wash the cells by 1

ml Binding buffer

4.

Repeat step 2

5.

Resuspend the cells in 100

µl of binding
buffer

6.

Add 5 µl of 7
-
AAD and 5 µl of anexin
-
PE

7.

Incubate 25´ in dark at RT

8.

Add 400

µl of binding buffer

9.

Measure at Canto BD

II, in software BD

Diva
-
Software
TM

46


3.8.8

Statistical evaluation


Statistical analyses were performed on GraphPad

Prism 5 software. For
validation phase of study, Wilcoxon non
-
parametic paired test was used.

In

vitro
experiments

were evaluated by a

two
-
tailed

non
-
parametric t
-
test. P values lower
than 0.05 were considered as statistically significant.


























47


4

Results

4.1

MiRNA qRT
-
PCR/ TLDA TaqMan Low Density Arrays

Three paired samples of GBM were chosen for global miRNA profiling and
miRNAs with the highest difference among initial and recurrent GBM were chosen
for further validation.
For presentation
of miRNA expression profile
,

a heat map
was
used. Green color indicates

dowregulated expression of miRNA in sample.
Red color means
upregulation of miRNAs

(Figure 7).



Figure 7:

Heat map of three most significant
ly deregulated

miRNAs
between
primary GBM
samples and recurrent samples
obtained from large
-
scale miRNA
profiling
.


4.2

MiRNA validation

Three m
iRNAs
,

miR
-
34a, miR
-
338
-
5p and miR
-
1274a
, that were f
ound to be
deregulated using TLDA microarrays
,

were further validate
d

on 10 pair
ed

samples of initial and recurrent GBMs. All three validated miRNAs were
significantly upregulated in recurrent
samples
compared to initial GBM

sample
48


miR
-
34a (P=0.0195)
,
miR
-
338
-
5p (P=0
.0195) and miR
-
1274a (P=0.0098)

(Figure
8).



Figure 8:

Validation of

different miRNA expression between 10 paired GBM
samples using
Wilcoxon

test.

Levels of miRNA: A)
miR
-
34a,
B)
miR
-
338
-
5p and
C)
miR
-
1274a
are expressed as relative expression
2
-
ΔCT
.


4.3

Transfection of miRNAs


Cell lines A172
,

T98G and U87MG were transfecte
d using Lipofectamin

RNAiMAX and pre
-
miRNA
-
34a or

pre
-
miR
-
338
-
5p oligonucleotides. Expression
level
s

of miR
-
34a and miR
-
338
-
5p were

measured 24 hour after transfection

(Figure 9).


49



Figure 9:

Expression level
s

of
A)
miR
-
34a

and
B)
miR
-
338
-
5p

24 hours
after
transfection.

Normalization was performed using
2
-
ΔΔCT

equation to RNU44
endogenous control. Mock is expressed as baseline and expression of miRNAs in