10
2/7/2016 1 Tuula Salo 1 , Susanna Teppo 1 , Sini Nurmenniemi 1 , Marilena Vered 2 , Dan Dayan 2 , Carolina Bitu 1 and Pia Nyberg 1 . 1 Department of Diagnostics and Oral Medicine, Institute of Dentistry, University of Oulu, Oulu, Finland, and 2 Institute of Pathology, Chaim Sheba Medical Center Center, Tel Hashomer, Ramat Gan, Israel. Tuula Salo Professor of Oral Pathology Universities of Oulu and Helsinki Human uterus leiomyoma tissue for invasion studies My dream: To create a model for analyzing the invasion process: the interaction between human carcinoma cells and tumor microenvironment matrix using human tumor tissue in vitro Human cancer cells Rat tail type I collagen + Mouse EHS-tumor Matrigel + Human Fibroblasts Nylon membrane Medium (+ e.g. inhibitors) Steel grid The Classic Collagen 3D-Organotypic Model ”rat-mouse-man” (since 1980´s; Fusenig NE et al.) Problems with classic 3D ”rat-mouse-man” model ”Comparing murine with human tissue structures suggest, that even the structural and physical parameters diverge (Wolf et al. Seminars in Cell and Dev Biol, 2009) Human tumor microenvironment model = Myoma model: Uterine Bening Smooth Muscle Neoplasm Myoma is common: incidence 20 – 50 % Most likely to be diagnosed with leiomyomas at the age of 40 – 50 yr Easy to obtain as a left- over material after the tumor operation Uterus Myoma Cutting myoma slices

PowerPoint Presentation seminar 2015 Salo.pdf · cytoskeleton organization biogenesis Totally 1.4% (751) genes were differentially expressed (FC 1.5 or more) For microarray analysis,

  • Upload
    lamdat

  • View
    215

  • Download
    0

Embed Size (px)

Citation preview

2/7/2016

1

Tuula Salo1, Susanna Teppo1, Sini Nurmenniemi1, Marilena Vered2, Dan Dayan2,

Carolina Bitu1 and Pia Nyberg1.

1Department of Diagnostics and Oral Medicine, Institute of Dentistry, University of Oulu, Oulu, Finland, and2 Institute of Pathology, Chaim Sheba Medical Center Center, Tel Hashomer, Ramat Gan, Israel.

Tuula Salo Professor of Oral Pathology

Universities of Oulu and Helsinki

Human uterus leiomyoma tissue for invasion studies

My dream: To create a model for analyzing the invasion process:the interaction between human carcinoma cells and

tumor microenvironment matrix using human tumor tissue in vitro

Human cancer cells

Rat tail type I collagen+

Mouse EHS-tumorMatrigel

+Human Fibroblasts

Nylon membraneMedium

(+ e.g. inhibitors)

Steel grid

The Classic Collagen 3D-Organotypic Model

”rat-mouse-man” (since 1980´s; Fusenig NE et al.)

Problems with classic 3D”rat-mouse-man” model

”Comparing murine with human tissue structures suggest, that even the structural and physical

parameters diverge “

(Wolf et al. Seminars in Cell and Dev Biol, 2009)

Human tumor microenvironment model = Myoma model:

Uterine Bening Smooth Muscle Neoplasm

• Myoma is common: incidence 20 – 50 %

• Most likely to be diagnosed with leiomyomas at the age of 40 – 50 yr

• Easy to obtain as a left-over material after thetumor operation

Uterus

Myoma

Cutting myoma slices

2/7/2016

2

8 x 4 mm punch biopsy slices of an average size myoma will give ~125 discs

6 cm

Myoma disks have been cut out from the original tissue with 8 m biopsy punch.

4 mm

8 mm

Figure 1.

HSC-3 in fbl + collagen gelNurmenniemi et al. Am J Pathol 2009

Tongue SCC, patient sample

HSC-3 in myoma disc

Pancytokeratin AE1/AE3

Highly invasive tongue carcinoma cells (HSC-3) invade in myoma similar to OTSCC in vivo=

(mesenchymal and collective or chain migration)

Growth pattern of OTSCC cells in myoma vs. 3D organotypic collagen model?

Over 100 myomas are currently tested:HSC-3 invasion varies in different myoma discs

7A 28 16 34A

32C54

5026B49B47

32A 41C

39B 54 26A 53

35A235148

37

27

53

Cells:VIM= fibroblastsSMA= smooth muscle cells CD 45 and CD 68

= inflammatory cellsFVIII =endothelial cells

Nurmenniemi et al. Am J Pathol 2009

Composition of Myoma tissue (IHC)

Matrix molecules:Hyaluronic acidCollagen types I, III, IVLaminins

Hyaluronic acid

Most of the cells in myoma discs are nonvital (apoptotic) after storage in liquid nitrogen

TUNEL (green) &

DAPI (blue), Roche

ApopTag,

Chemicon

Alahuhta et al. Exp Cell Res 2015

2/7/2016

3

Growth pattern:Tongue carcinoma cells invade mostly in budding pattern

(less than 5 cells in a group); whereas mucoepidermoid carcinoma cell line invades in

collective clusters into myoma disc

HSC-3 Highly invasive tongue SCC

MUC-1 Mucoepidermoid carcinoma cells

HSC-3 cells invade up to 7 x deeper in myoma than in the 3D collagen + fibroblasts

Nurmenniemi et al. Am J Pathol, 2009

… but they proliferate less (Ki-67 index) in myoma than in collagen

Nurmenniemi et al. Am J Pathol, 2009

Some cancer cells are positive for both mesenchymal (VIM) and epithelial (AE1/AE3) markers in myoma.

Sign for epithelial-mesenchymal-transition (EMT)

Nurmenniemi et al. 2009 Nurmenniemi et al. Am J Pathol 2009

VIM AE1/AE3 AE1/AE3 + VIM

Nurmenniemi et al. 2009

Inhibition of invasion and type I collagen degradation

with MMP inhibitor GM6001

Analyses of collagen degradation products (RIA) from media samples reflect carcinoma cell invasion depth

In myoma discs, not in collagen, ICTP-RIA measures the invasion depth of HSC-3 cells

Type I collagen fragments12 % SDS-PAGE: media samples collected from days 2, 5, 7

- bMe + bMe

28

36

55

72

95

130

250

d2 d5 d7 d2 d5 d7

”Rinsing media” from intact myoma tissue

Are soluble factors relieced during rinsing

period involved in invasion?

2/7/2016

4

Yes - Rinsing of myoma tissue affects SCC cell invasion!

Alahuhta et al. Exp Cell Res 2015

”Anti-invasive” arresten transfected HSC-3 cells

Aikio et al. Plos One 2012

3D Collagen gel+fbl

Intact myoma

Myoma discs were rinsed for 14 days

in DMEM

w/o serum before the experiment

with the SCC cells Invaded similar to Ctrl

No invasion

Rinsed myomaNo invasion

Rinsed Myoma

Invasion depth

Teppo et al., Exp Cell Res 2013

Hypoxic conditions induced HSC-3 invasion in rinsed myoma, but

…intact myoma provides the best hypoxic TME for HSCC-3 invasion. Why?

Intact Myoma

LOX is secreted by hypoxic tumours; facilitates invasion and metastases formation. (Erler et al. Nature 2007)

Invasion inducing and hypoxia factorsare present in intact myoma tissue!

Myoma tissue extract w/o HSC-3 cells: Western blot

Intact Rinsed

Teppo et al., Exp Cell Res 2013

MMP-11 has pro-invasive & anti-apoptotic properties.(Fromigué O, et al. Int J Cancer. 2003)

MMP-11 LOX-1

Intact Rinsed

Myoma tissue extract w/o HSC-3 cells: western blot

Some examples of the published resultsusing

myoma invasion assayss

The myoma model has been used in more than 20 publications.

So far it has not been criticized by reviewers

Lymph node metastatic, more aggressive cell line (SCC-9 ZsG LN-1) of the primary tongue carcinoma (SCC-9ZsG) cells invaded significantly

deeper into myoma

Agostini M et al. Mol Cancer Ther. 2014

Toll-like receptor 9 activation with CpG-oligonucleotide enhancesthe invasion of hMSCs compared to activation with non-CpG

Non-CpG CpG

Nurmenniemi et al. Exp Cell Res. 2010

Human bone marrow derived mesenchymal stem cells

(hBMMSC) do not invade in myoma

CpG induced invasion!

2/7/2016

5

An increase in MMP-13 synthesis is detected in the CpG-activated MSCs; addition of MMP-13 antibody significantly

diminished the CpG induced invasion

RT-PCR

Western blot

MMP-13 (arrow) in invading MSCs into myoma

Nu

rmen

nie

mi e

t a

l. E

xpC

ellR

es20

10

• MMP-13 participates in MSCs invasion

Trypsin-2 transfected HSC-3 cells invaded deeper in myoma than controls -

MM-14 specific inhibitor (peptide G) reduced the invasion

Peptide G is a selective MMP-14 peptide-inhibitor (GACFSIAHECGA)• It do not affect the activities

of MMP-1, -2, -3, -7, -8, -9, -10, -11, -12, -13, -15, -17 or -20

• The peptide effectively inhibited the migration and (Transwell) invasion of cancer cell lines and reduced the growth of tongue carcinoma xenografts in mice.

[Suojanen et al. 2009]

MMP-14 induces HSC-3 invasion

+ peptide G

Vilen et al. 2012

HSC-3 HSC-3 + GF HSC-3 + MSC

DOK DOK + GF DOK + MSC

Oral dysplastic cells (DOK) did not invade in myoma model, not even after co-culturing with gingival fibroblasts (GF) or

with mesechymal stem cells (MSC)

¤ Conforming the expression of molecules –

Myoma model in a laser capture method

A

B

HSC-3 cells grown in myoma disks, stained with cathepsin K ab

Laser capture of HSC-3 cells

β-Actin

cell lines microdissected myoma

Cat-K

The expression of cat K was confirmed by RT-PCR of the RNA isolated from lazer captured cells from either frozen sections or from paraffin embedded samples

Bitu CC et al. Plos ONE 2013

”In SCC, carcinoma associatedfibroblasts (CAFs) are always the leading cell of the invading cohortwith the SCC cells closely followingbehind. ”

”Hitching a ride” mechanism(Gaggioli et al. Nat Cell Biol 9: 1392-400, 2007)

In the myoma model: bone marrow derived mesenchymal stem cells (MSC) and SCCs are mixed:

Some of the MSCs are leading the invading SCCscolonies

Stem cells + HSC-3 in myoma

HSC-3 + M1-M -> reducedHSC-3 + M2-M -> induced HSC-3 invasion compared to HSC-3 cells alone

Pir

ilä e

t a

l. P

los

On

e 20

14

p<0,02

p=0,002

p=0,034

THP-1 leukemic cells were induced to macrophages (M) with:PMA+ LPS + IFN-γ -> M1-M (pro-inflammatory)PMA+ IL-4 + IL-13 -> M2-M (pro-tumorigenic)

Co-cultures of macrophage subtypes and HSC-3 in myoma

Depth of HSC-3 Invasionwas measuredfrom myoma

2/7/2016

6

What to do with discarded discsand

the left-over myoma tissues

?

Nothing must go to waste!

What if we prepare Myogel similar to Matrigel*?

Crushing the myomatissue (pooled fromseveral patients) in liquid nitrogen and mortar

Homogenizing the myoma pulp tissue

homogenizer and centrifugation

Purifying the myoma gel

Four dialysissteps

Kibbey MC. Maintenance of the EHS sarcoma and Matrigel preparation.J Tissue Cult Methods. 1994;16:227–30.

*

Hynda Kleinman, NUH, USA –”the mother” behind Matrigel

Interview of Hynda Kleinman yr 2013:

“Are you surprised that Matrigel was so successful?”Hynda Kleinman:“I’m shocked that it’s this useful. I’m also shocked that no one has invented anything better. Because it’s a no brainer that you could make it from lots of other tissues, but it’s still made the exact same way we made it 25 years ago. It was just a wild tumor. There must be so many thousands of tumors—you can go into a bank and pull out all kinds of tumors and make it. Nobody’s done that. No company’s done it.”

Matrigel:15-20 mil.€/yr Engelbreth-Holm-Swarm(EHS) sarcomagrowssubcutaneouslyin mice

Millions of mice havebeen killedto produceMatrigel!

Basement membrane extracts (BME)prepared from EHS mouse sarcoma tissue

(Patent for Matrigel ended yr. 2006)

Composition:

Laminin

Collagen

Heparin sulphate proteoglycan,

Entactin/nidogen

Growth factors (TGF-b, EGF)

Composition varies from lot to lot

Matrigel®, ECMatrix™, Cultrex ®, BME ®, Geltrex ®

• Adhesion experiments

• Invasion (3D)

• Drug screening

• Angiogenesis

(tube formation)

• Xenografts

• Directed differentation

Other human ECM extracts prepared similarlyare all from normal tissues or from cell cultures

• Skeletal muscle myogel

adipogenic and support the ex vivo amplification of

corneal epithelial cells

• Amnion tissue extract (Amgel, HumBiogel)

Goodly LJ et al. Tumour Biol. 1994;15:326–36.

• MaxGel™ Human ECM (Sigma), AlphaMAX3D (Neuromics)

co-culture of fibroblasts and keratinocytes

-> matrix for BME extract

Abberton KM, et al. Cells Tissues Organs. 2008;188:347–58Francis D. Exp Eye Res. 2009;88:339–46.

Crushingthe myomadiscs

• Using liquid nitrogen and mortar

Homogenizingthe myoma pulp

• Using tissue homogenizer and centrifugation

Purifying the myoma gel

• Four dialysissteps

MyogelPreparation of the gel:

Human myoma -> Myogelvs

Mouse EHS-tumor -> Matrigel®

A novel human leiomyoma tissue derived matrix for cell culture studiesSalo T, Sutinen M, Apu EH, Cervigne N, deOliveira C, Akram SU, Ohlmeier S, Suomi F, Ekholm L, Juusela P, Astrom P, Bitu CC, Santala M, Savolainen K, Korvala J, Leme AFP,

Coletta RD; BMC Cancer, 2015

2/7/2016

7

Proteomic of Myogel vs. Matrigel

MyoGel: out of 765 identified proteins 34% were the same as in Matrigel® and 66% were different

MyoGel was lacking some growth factors and MMPs present in Matrigel®

Myogel had a number of ECM and connective tissue forming proteins lacking in Matrigel®

Myogel

with SCC cells

Matrigel®

with SCC cells

Myogel

without cells

0 h 7.0 - 7.5 8.0 - 8.5 7.0 - 7.5

17 h 7.0 - 7.5 8.0 7.0 - 7.5

48 h 6.5 - 7.0 6.0 - 6.5 7.0 - 7.5

The pH:

Myogel is neutral & more stable than Matrigel®

Thin coating: adhesionCarcinoma cells adhere most to Matrigel

Coating the bottlesnoneBSAMatrigelMyogel

+ HSC-3 cells

Thin coating: gene expression

Carcinoma cells expressed genes differentially on top of

Myogel vs plastic

Mostly pathways were related to: intracellular organelle cytoskeleton organization biogenesis

Totally 1.4% (751) genes were differentially expressed (FC 1.5 or more)

For microarray analysis, 90,000 HSCseeded into uncoated or Myogel coated 6each). The next day the cells were harvested for RNA extraction using a Qiagen RNA kit. Three samples of each group (on top of plastic or Myogel coating) were pooled; the pools contained an equal amount of RNA from each sample. AffymetrixPlus 2.0 Arrays were used for microarray analysis and experimental procedures were performed according to the Expression Analysis Technical Manual. Briefly, 1used as a template to synthesize biotinylated GeneChip 3’IVT Express kit (manufacturer’s instructions. The prior to hybridization to Affymetrixarrays containing approximately 55,000 human transcripts. The array was washed and stained with streptavidinProbes). Finally, biotinylated antiwas used to amplify the staining signal and a second staining was performed with streptavidinon a GeneChip Scanner 3000 (Kingdom). The expression data was analyzed to find genes with fold changes (FC) of 1.5 or more using

FC 1.2 or more were divided into Gene Ontology (GO) categories using

a dChip enrichment analysis tool

Uncoated Myogel coated after 24 h HSC-3 cells collected

RNA harvested Affymetrix GeneChip Human Genome U133 Plus 2.0 Arrays

Significant changes related to actin polymerization and reorganization

PBS

Thin coating: horizontal migrationThe HSC-3 migration is faster on Myogel than on Matrigel®

Myogel

Matrigel

PBS

Salo et al, BMC Cancer 2016

Matrigel

Myogel

Incubation for 12 to 48 hours

Nylon filter membrane

Human cancer cells(50 000/well)

Medium +chemoattractant

Transwell insert

Thick coating: Transwell invasion

Invading cells

OR

2/7/2016

8

Cel

l co

un

t/

fram

e

12h 24h 48h

Concentration of Myogel and Matrigel was 2.4 mg/ml

Matrigel vs. Myogel vs. No gel HSC-3 cells invade more efficiently through Myogel than Matrigel

Matrigel

Myogel

No gel

Salo et al. BMC cancer 2015

Thick coating with Myogel-LMA:

Myogel-LMA:

• Low Melting Agarose (LMA) is added to Myogel for keeping it in gelatinous form

• Cells did notinvadethrough LMA w/o myogel

SCC-25 SAS Bowes Pa02c Pa03c Pa04c

p 0.002 p 0.002 p 0.015 p 0.002 p 0.002 p 0.026

** ********

Thick coating:

Transwell migration of melanoma & pancreatic cancer celllines was increased in Myogel-LMA vs Matrigel

Myogel and Matrigel 2.4 mg/ml (unpublished)

Thin and thick coating:Scratch wound healing assay with IncuCyte

MDA

HSC-3

HSC-3 GFP

Pa01c

Pa02c

Pa03c

Pa04c

empty

Myo Migration Myo Invasion Mat invasion Mat Migration

More cell lines are invading through Myogel than Matrigel(unpublished)

Hanging drop spheroids:

Carcinoma (HSC-3) cells form larger colonies within Myogel-LMA than in plain conventional low melting soft agarose (LMA)

1

2

4

3

Inner border

Plate lead

HangingDrop

Flipping the covered plate upside downIn a humadifying chamber

Plate containing drops in 4compartments from eachnumbered cell embeddedsample mixtures

Keep in incubator for 5 minutes

Keep in incubator for ≥ 3 hours, after that add 1ml 2% FBS containing media/compartment inthe microscopy unit.

Hanging-drop assay:Myogel-collagen, Matrigel-collagen or collagen

2/7/2016

9

HSC-3 cells move faster in Myogel-collagen matrix than inMatrigel®-collagen matrix in the

3D hanging-drop cultures

Salo et al, BMC Cancer

Angiogenesis:Myogel efficiently induces endothelial cell tube formation:

more, smaller & longer lasting (up to 72h) tubes compared to Matrigel-GFR or ECMatrix

Salo

et

al,

BM

C C

an

cer,

201

6

100-1000candidates

Myogelthin/thick

coatedviability

proliferation, invasion

A) Analyzing most efficient drugs using 2D /3D cell cultures

Myoma Discs3D Invasion

B) Testing best candidate drugs in “natural 3D tumor

matrix”

Mice/animal experiments

Clinical trials

Possible applications for ”Myoma Kits” - A part of cancer drug testing ?

Viability of the drug-test cells:Pancreas carcinoma cells (Pa02c) on top of plastic, Matrigel or Myogel

CellTiter-Glo® ATP amount

Matri0.15

mg/ml

Matri0.31

mg/ml

Matri0.62

mg/ml

Myo0.15

mg/ml

Myo0.31

mg/ml

Myo0.62

mg/ml

FIMM-Wennerberg lab

No coating

(unpublished)

Drug sensitivity and resistance testing (DSRT) of Pa02c (FIMM-Wennerberg lab)

https://www.fimm.fi/en/services/technology-centre/htb/instructions/workflows

132 compounds”drugs for leukemia”5 concentrations each

1. No coating2. Matrigel (0,62 mg/ml) 3. Myogel (0,62 mg/ml)

Three days incubation

Viability measuring byThe CellTiter-Glo®

Luminescencedetection using thePHERAstar®

Data analysisby Dotmatics(EC50, IC50)

Name Mechanism/Targets IC50 GRAPH IC502 GRAPH3 IC503 GRAPH5

Panobinostat HDAC inhibitor 70,42 54,3 48,4

Dinaciclib CDK inhibitor 18,32 18 17,1

Selumetinib MEK inhibitor 97,31 342,5 91

GDC-0994 ERK inhibitor 679,58 175 1879,2

Mercaptopurine Antimetabolite 1031,3 295,2 717,8

Palbociclib Cdk inhibitor (Cdk4/6) 8409,1 3839 680,1

Clofarabine Anti-metabolite:Purine analog 3873,3 516,5 988,5

No gel Myogel Matrigel

IC50 of 132 tested compounds against Pa02cAbout one third of the drugs (48) had differend IC50 values when

tested on top of Myogel vs plasticIC50: no coating ≈ Myogel = 64 %

Same effect

IC50: no coating > Myogel = 20% Drug more effective in Myogel

IC50: no coating < Myogel = 16 % Drug more effective on plastic

More than one third of the drugs (56) had differend IC50 values when tested on top of Matrigel vs plastic

IC50: no coating ≈ Matrigel = 57 % Same effect

IC50: no coating > Matrigel = 27% Drug more effective in Matrigel

IC50: no coating < Matrigel = 16 % Drug more effective on plastic

(unpublished)

2/7/2016

10

MyogelHuman tumor

extracellular matrixmg/ml

Adhesion experiments

Invasion (3D)

Drug screening

Angiogenesis

• Xenografts?

• Directed differentation?

• What else?

My dream: Myogel - a commercial product replacing at least partially the use of Matrigel

• Patent application is filed• TEKES TUTLI project (8/2015-1/2017)

to further evaluate the properties of myogel and possibilities for producing Myogel as a commercial product

But: Collaborative work is still needed before we can evaluateif Myogel would be a useful multipurpose product

Tuula Salo1, Susanna Teppo1, Sini Nurmenniemi1, Marilena Vered2, Dan Dayan2,

Carolina Bitu1 and Pia Nyberg1.

1Department of Diagnostics and Oral Medicine, Institute of Dentistry, University of Oulu, Oulu, Finland, and2 Institute of Pathology, Chaim Sheba Medical Center Center, Tel Hashomer, Ramat Gan, Israel.

Researchers and collaborators in myoma projects

Oulu groupPia Nyberg, PhD, docentJohanna Korvala, PhDMaija Risteli, PhDElias Sundquist, DDS, PhD-studentIlkka Alahuhta, DDS, PhD-studentEhsanul Hogue Apu, DDS, PhD-studentPast members:Sirpa Salo, PhDSini Nurmenniemi, PhDSusanna Teppo, MScMeeri Sutinen, PhDOulu collaboratorsEklund groupMäkinen groupTasanen-Määttä groupHelsinki groupAhmed Al-Samadi, DDS, PhDJenni Vasara, MScKatja Puskala, MScFIMM collaboratorWennenberg group

Brazil:Coletta, Graner and Leme groups

Current main financial support

Brazil collaboratorsColetta, Graner and Leme groups

Current financial support