• No results found

Knockout studies of Panc1 cells

N/A
N/A
Protected

Academic year: 2021

Share "Knockout studies of Panc1 cells"

Copied!
44
0
0

Loading.... (view fulltext now)

Full text

(1)

Knockout studies of Panc1

cells

Effect of CCN1 gene knockout on Panc1

cell viability

MARTIN SUNDIN

Master’s Programme, Engineering Physics, 120 credits Date: June 22, 2021

Supervisor: Rainer Heuchel Examiner: Hjalmar Brismar

School of Engineering Sciences Host company: Karolinska institutet

Swedish title: Knockout studier av Panc1-celler TRITA-SCI-GRU 2021:069

(2)

© 2021 Martin Sundin

(3)

Abstract | i

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal form of cancer with very few available treatment options of which none has great effect.

Cancer cells and stromal cells such as stellate cells which exist in abundance in PDAC interact by crosstalk, resulting in a tumorigenic collective response.

With the help of a previously developed 3D co-culture spheroid model the effect of a CRISPR/cas9 knockout of the cellular communication cetwork factor 1 (CCN1) gene together with gemcitabine (GEM)treatment has been investigated in terms of Panc1 cell viability and gene expression. Spheroids consisting of wild-type and knockout cell lines, each identified by western blots were cultured, imaged and treated. Viability assays and RNA extraction followed by PCR showed that the viability of the cancer cells in the spheroids were higher for the cells with CCN1 knockout. Cancer cells were also co- cultured with stellate cells with the goal of investigating the effect of the cellular crosstalk on chemoresistance.

Keywords

PDAC, crosstalk, stroma, spheroid, western blot, viability, gemcitabine.

(4)

Sammanfattning | ii

Sammanfattning

Pankreatisk duktal adenokarcinom (PDAC) är en ytterst dödlig form av cancer med få tillgängliga behandlingsalternativ, varav ingen är särskilt effektiv.

Cancerceller och stromala celler såsom de stellatceller som rikligt förekommer i PDAC interagerar med varandra genom överhörning, vilket leder till en effekt som hjälper tumören att proliferera. Effekten av en CRISPR/cas9 knockout av genen CCN1 tillsammans med behandling med gemcitabin vad gäller cellviabilitet och genuttryck studerades med hjälp av en tidigare utvecklad fleratig sfäroidmodell. Sfäroider, bestående av vildtypceller och knockoutcellerlinjer som identifierades med western blots, odlades, fotades och behandlades.

Viabilitetstester och extraktion av RNA följt av PCR visade att viabiliteten av cancerceller i sfäroiderna var högre för de celler som var knockout.

Cancerceller samodlades även med stellatceller med målet att undersöka effekten av cellernas överhörning på motståndet mot kemoterapi.

Nyckelord

PDAC, överhörning, stroma, sfäroid, western blot, viabilitet, gemcitabine.

(5)

Acknowledgments | iii

Acknowledgments

First I would like to express my gratitude to my supervisor Rainer Heuchel for accepting me into his research team and providing invaluable guidance throughout my stay.

Likewise to my tutors in the team, Xinyuan Liu and Beate Gündel for assisting and teaching me with great patience and kindness.

Thank you also to Maura Krook for providing the opportunity for me to do this thesis by setting up the contact between me and the research team.

Last but not least loving thanks to my parents for their ever-present encouragement and support in every way possible.

Stockholm, June 2021 Martin Sundin

(6)

CONTENTS | iv

Contents

1 Introduction 1

1.1 Background . . . 1

1.2 Problem . . . 1

1.3 Purpose . . . 2

1.4 Goals . . . 2

1.5 Structure of the thesis . . . 2

2 Background 4 2.1 CRISPR/Cas9 . . . 4

2.1.1 CCN1 . . . 4

2.2 Crosstalk . . . 5

2.3 The spheroid model . . . 5

2.4 PDAC treatment options . . . 5

3 Methods 6 3.1 Protocols . . . 6

3.2 Data Collection . . . 9

3.3 Assessing reliability and validity of the data collected . . . 10

3.3.1 Validity of method . . . 10

3.3.2 Reliability of method . . . 10

3.3.3 Data validity . . . 10

3.3.4 Reliability of data . . . 10

3.4 Data Analysis . . . 11

3.4.1 Software Tools . . . 11

4 Results and Analysis 12 4.1 Results. . . 12

4.1.1 Western blot . . . 12

4.1.2 2D culture and viability . . . 16

4.1.3 3D culture and viability . . . 19

(7)

Contents | v

4.1.4 Gene expression . . . 22

4.2 Reliability Analysis . . . 23

4.3 Validity Analysis . . . 23

4.3.1 Western blot . . . 23

4.3.2 Viability assay . . . 23

5 Discussion 24 5.1 Viability observations . . . 24

5.1.1 Spheroid aggregation and physical traits . . . 24

5.1.2 Off-target effects . . . 25

5.2 Assays . . . 25

6 Conclusions and Future work 26 6.1 Conclusions . . . 26

6.2 Future work . . . 26

6.2.1 Unfinished work . . . 26

6.2.2 Possibilities of future work . . . 26

6.3 Reflections . . . 27

References 29

A Methylcellulose stock solution 33

B Poster 34

(8)

LIST OF FIGURES | vi

List of Figures

4.1 Western blot of HSP60 . . . 12

4.2 Western blot of CCN1 . . . 13

4.3 Western blot of CCN1 . . . 14

4.4 Western blot of CCN1 . . . 14

4.5 Bright-field images of untreated 2D cultures.. . . 16

4.6 Bright-field images of 2D cultures treated with 5 µM gemcitabine 17 4.7 2D cell viability . . . 18

4.8 Bright-field images of untreated 3D cultures.. . . 19

4.9 Bright-field images of 3D cultures treated with 5 µM gemcitabine. 20 4.10 3D cell viability . . . 21

4.11 Relative gene expression . . . 22

B.1 Poster . . . 34

(9)

List of acronyms and abbreviations | vii

List of acronyms and abbreviations

ATP adenosine triphosphate BSA bovine serum albumin

CAFs cancer-associated fibroblasts

CCN1 cellular communication cetwork factor 1 CYR61 cysteine-rich angiogenic inducer 61 DCK deoxycytidine Kinase

DMEM Dulbecco’s Modified Eagle’s Medium DMSO dimethyl sulfoxide

ECM extracellular matrix FBS fetal bovine serum GEM gemcitabine

GSEA gene set enrichment analysis PBS phosphate buffered saline

PDAC Pancreatic ductal adenocarcinoma PS Penicillin-Streptomycin

PSCs pancreatic stellate cells

SDS-PAGE Sodium dodecyl sulphate–polyacrylamide gel electrophoresis shRNA small hairpin RNA

SLC29A1 solute carrier family 29 member 1 TBST tris-buffered saline tween

WT wild-type

(10)

Introduction | 1

Chapter 1

Introduction

1.1 Background

PDACis characterized by late diagnosis and almost complete therapy resistance, leading to a poor prognosis and high lethality. This is emphasized by numbers showing that PDAC constitutes 2.5% of cancer diagnoses in the world, but as high as 4.5% of the mortality, [1] and while treatment options for other types of cancers are being developed and improved, the lethality of PDAC remains high.

A specific trait of PDAC is its densely packed stromal component, which predominately consists of pancreatic stellate cells (PSCs). These PSCs are known to communicate bidirectionally with cancer cells and other stromal cells. PSCs, when activated facilitate stromal fibrosis,[2] and the product of the crosstalk of all celltypes combined with the excessive fibrosis (desmoplasia) is generally believed to play a major part in the therapy-/ chemoresistance of PDAC.[3]

Bygene set enrichment analysis (GSEA)of cancer/stromal cell co-culture, a number of genes has been identified as possible targets [4] of which CCN1 in Panc1 cells was selected for knockout by CRISPR/cas9 technique.

1.2 Problem

While PSCs constitute the majority of the stromal component in PDAC, the stroma also consists of extracellular matrix (ECM) proteins secreted by the PSCs when activated by carcinogenesis. The activated PSCs also transform into myofibroblasts andcancer-associated fibroblasts (CAFs). This makes the stroma a complex structure to study, especially when attempting to understand

(11)

Introduction | 2

the cancer-stroma crosstalk. The palpable effect of a dense stroma however is an increase in pressure inside the tumor which acts as a constricting force on blood vessels, creating a nutrient and oxygen poor environment. This also affects the supply of blood-bourne therapies, since circulation is compromised into the tumor[5]. PDAC cells however have adapted and survive in these nutrient poor, hypoxic conditions and have thus acquired an innate advantage to thrive where healthy cells do not.

More knowledge about the crosstalk between tumor cells and PSCs is thus needed in order to better understand and tackle the tumor-supporting properties of the stromal component in PDAC.

1.3 Purpose

To investigate and gain further understanding about the effect the knockout of a previously identified gene in Panc1 cells has on proliferation and chemoresistance.

1.4 Goals

1. To verify the knockout efficiency of CCN1 in several Panc1 cell lines.

2. To grow Panc1 monospheroids and compare the effectiveness of existing chemotherapeutic options between knockout and wild-type (WT) cell lines.

3. To investigate the effect of the CCN1 knockout on the gene expression of selected genes.

1.5 Structure of the thesis

Chapter2covers a more detailed description of background, including theory and past work in the same area.

Chapter3goes through the methodology applied in this work, as well as the materials used. The focus here will be to present the methods and give an overview of the steps taken, as well as any deviations made from standardized protocols.

Chapter4presents all observed or measured results from the experiments described in chapter 3, with analysis explaining the meaning of the results in context to the desired outcome.

(12)

Introduction | 3

Chapter 5 further explores the meaning of the results with emphasis on discussion about character and magnitude of any errors.

Chapter6explains what conclusions can be made from the results in this work, while also stating and suggesting what future work that can and/or will be made in this area.

(13)

Background | 4

Chapter 2

Background

2.1 CRISPR/Cas9

With CRISPR/Cas9 it is possible to essentially edit the nucleotide sequence of a specific gene in the genome, and in this particular case of Panc1 cells. The gene CCN1 mentioned in chapter1was the subject of attempts to be removed by the Karolinska Genome Engineering (KGE) core facility, by introduction of Cas9 protein along with a piece of guide RNA corresponding to the CCN1 gene into Panc1 cells. Out of the pool of all cells treated in this way, single cells were picked randomly and transferred into a 96-well cell culture plate to grow single clones. The actual success rate of the gene editing of CCN1 was not known and had to be tested by investigating if the cells from the different clones indeed have lost the expression of this gene by looking for the CCN1 protein, the product of the CCN1 gene. Indeed the CRISPR technique is not always successful as will later be shown in chapter4as the results of western blots of different Panc1 cell lines differ. There is also the problem of off-target effects, where the removal of a target gene comes with alterations in other unknown sites.[6] This unpredictable event calls for caution when drawing conclusions about what the removal of the target gene actually leads to.

2.1.1 CCN1

CCN1 orcysteine-rich angiogenic inducer 61 (CYR61) is a gene coding for a protein with the same name. Importantly here it plays a role for example in cell proliferation and adhesion, as well as in formation ofECM.[7] It was found that the gene expression of CCN1 in Panc1 was higher when co-cultured with murine PSCs[4], and the CCN1 gene was selected for knockout with

(14)

Background | 5

CRISPR/cas9.

2.2 Crosstalk

The crosstalk mentioned in chapter 1 is believed to play a major role in the high lethality and therapy resistance of PDAC. Crosstalk is a complex and not yet fully described concept of intercellular communication, and is defined by many different signal transduction pathways affecting each other. The alterations of the gene expression on cellular level caused by the tumor-stroma interactions are believed to effectively strengthen the tumor’s resistance to therapies and facilitate growth of the tumor.

2.3 The spheroid model

3D tumor and tumor-/stroma cell spheroid models were previously developed[8].

The main advantage of 3D growth over 2d/monolayer growth is the increased likeness to in vivo tumors. In a spheroid much like in a real tumor there will be oxygen and nutrient gradients, as well as 3D cell-cell interactions which have effects on how cells behave [9]. In order to analyse tumor-stroma crosstalk with this model, a stromal component like PSCs can be mixed with pancreatic cancer cells to form a heterospheroid model.

2.4 PDAC treatment options

Existing treatment options for PDAC include chemo- and radiotherapy as well as surgical resection, but as the 5 year survival rate for PDAC of 7%[10]

suggest none of them are particularly effective. Stroma-targeted therapies have been suggested as the stroma strongly contributes to the lethality ofPDAC, but as of yet there are no solutions at hand.

(15)

Methods | 6

Chapter 3

Methods

3.1 Protocols

Below are listed the materials and methods used involving cell culture and sample preparation.

Cell maintenance

PANC-1 cells were cultured at 37C, 5% CO2inDulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 1%

Glutamax, 0.5% Penicillin-Streptomycin (PS) and 0.1% phenol red as a pH indicator in 75cm2 tissue culture flasks. The cells were passaged at 80-100%

confluency by removing the medium and washing the cells with phosphate buffered saline (PBS)before adding trypsin-EDTA for 5 minutes in 37C, 5%

CO2. The trypsinization was stopped by adding new serum-containing (here, FBS) culture medium with which the flask was rinsed, the cells resuspended and split into a new flask. Split ratio was decided upon depending on when the cells were to be used, for basic propagation a split ratio of 1:7 was used.

Between passaging the medium was changed every 3-4 days.

Seeding of spheroids

Cells were trypsinized as above, washed in PBS and centrifuged at 500 rcf for 5 minutes, the supernatant was discarded and the pellet resuspended in 3 ml DMEM. The cell number density in this resuspension was then averaged over two measurements using a Countess™ Automated Cell Counter together with Countess™ cell counting chamber slides and Trypan blue stain (0.4%).

Cells were seeded in round-bottom Falcon 96-well plates with 2500 cells per

(16)

Methods | 7

well, together with DMEM and 0.24% methylcellulose (see appendix A) to support self-aggregation. The total volume of resuspended cells, DMEM and methylcellulose in each well was calculated to amount to 100 µl.

Seeding of 2D culture

Seeding of 2D cultures follows the protocol for seeding spheroids except no methylcellulose is used and the cells are seeded in flat-bottom 96-well plates with 100 µl medium per well.

Freezing cells

Cells were trypsinized and centrifuged at 500 rcf for 5 minutes, the supernatant was discarded and the pellet resuspended in 1 ml freezing medium (90%

FBS, 10%dimethyl sulfoxide (DMSO)) per aliquot and pipetted into cryovials.

Cryovials were well isolated for slow freezing at -80C for up to 1 week before being moved to -170C for permanent storage.

Cell lysis for Western Blot

Cells were passaged into 6-well plates and incubated until 80-100% confluent.

The 6-well plate was removed from incubation and placed on ice. The medium was removed and the cells were washed with PBS before adding 50µl ThermoFisher Scientific Pierce IP lysis buffer into each well. The cells were scraped from the wells with a cell scraper, pipetted into Eppendorf tubes and kept on ice for 30 minutes. The cells were then centrifuged in 4C and 10500 rcf for 5 minutes, the supernatant was then collected and placed in a new tube. The protein concentration was calculated by use of a BioRad Quick Start Bradford protein assay, and 30g of protein was aliquoted in solution with PBS and 5 µl Laemmli Buffer to a total volume of 20 µl and stored in -20C.

Seeding of 2D and 3D cultures for real time RT-PCR and gemcitabine treatment

One 96-well plate for each spheroid type was seeded for RNA extraction and measurements of gene expression by real time RT-PCR. For drug treatment the 96-well plates were divided into blocks so that every drug concentration treated each type of spheroid in three wells.

(17)

Methods | 8

Real time RT-PCR

The contents in each 96-well plate were collected and put on ice. The tubes were centrifuged at 500 rcf for 5 minutes, the supernatant was discarded and the pellet collected and washed with 3ml cold PBS before being centrifuged again. Again the supernatant was discarded and 350 µl RLT buffer from the Qiagen RNeasy kit was added to lyse the spheroids. The tubes could then be frozen in -80C for later use. Total RNA was retro-transcribed into cDNA using the Bio-Rad iScript cDNA Synthesis kit and the ThermoFisher Scientific Maxima SYBR Green/Fluorescein qPCR Master Mix was used for real time RT-PCR as following: initial denaturation 10 min at 95C, 40 cycles of 15 s at 95C and 1 min at 60C.

Viability assay

The entire volume in selected wells were pipetted into corresponding wells in 96-well assay plates, and 100 µl of Promega 3D CellTiter-GloT M solution was added to all wells. The assay plates were placed on a shaker for 5 minutes and then put in room temperature for 25 minutes. The luminescence was detected in a SpectraMax i3x.

M30 apoptosis assay

The 96-well plates containing spheroids seeded for the M30 apoptosis assay were sealed and frozen in -80C to enable simultaneous running of spheroids from different dates of seeding.

Imaging

The cell cultures for the drug treatment experiments and M30 assays were imaged using an inverted Olympus IX81 microscope. The total magnification used was 40x, with scale indicated in every image.

SDS-PAGE

Cell lysates for western blot were heated to 95C for 5 minutes and quickly centrifuged to gather contents in the bottom of the vial. A Bio Rad Mini PROTEAN system was used for Sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE), assembled according to manual, lysates of 6 unverified CCN1 cell lines was loaded into the Mini-PROTEAN TGX Stain- Free Precast gel lanes together with 3 µl PagerRuler Plus Prestained Protein

(18)

Methods | 9

Ladder as well as positive control (PC) consisting of proteins from wild-type cells and a previously confirmed negative control (NC) consisting of proteins from the cell line C3. The system was ran at 80V until the samples reached the bottom of the gel.

Western blot

The gel from the SDS-page was blotted on a membrane using the BioRad Trans-Blot Turbo Transfer system. After the blotting the membrane was placed in a solution of 5 wt% bovine serum albumin (BSA) intris-buffered saline tween (TBST) on a shaker for 2 hours to block non-specific antibody bindings. The membrane was then sealed in plastic along with 1 ml of primary antibody (rabbit anti-CCN1 and chicken anti-HSP60 1:1000 in 5 wt% BSA- TBST) and refrigerated over night on a tilting board. The following day the membrane was subject to a 3x quick-wash in TBST followed by 3x15min wash in TBST on a shaker. A pre-prepared solution of secondary antibody (anti- rabbit and anti-chicken, 1:5000 in TBST) was added to cover the membrane for 2 hours on a shaker, and from the point of secondary antibody addition the membrane was kept in the dark to avoid photobleaching. After incubation in secondary antibody the same washing procedure as above is repeated but TBST is exchanged for PBS for a 3x quick-wash after the first 15 minute wash, and for the final 2x15 minutes wash. The secondary antibodies are coupled with different fluorescent substrate, and are detected at 488 nm (CCN1) and 647 nm (HSP60).

3.2 Data Collection

2D and 3D cultures for the drug treatment and gene expression experiments were seeded on day 0. On day 1 the cultures for the drug treatment experiment was treated. On day 4 all cell cultures were imaged, and a viability assay was performed on the drug treatment cultures. On day 5 RNA was extracted from the gene expression cultures, and a real-time RT-PCR was performed.

For real time RT-PCR three biological repeats were performed, with an average of the three being used as final result. For viability assays values were averaged over three wells per plate while also performing three biological repeats with one plate per repeat.

(19)

Methods | 10

3.3 Assessing reliability and validity of the

data collected

3.3.1 Validity of method

Cell culture and maintenance was performed in accordance to standard procedures.

Choice of methods regarding immunodetection, viability and gene expressions are all standard methods in respective fields.

3.3.2 Reliability of method

Averaging over several measurements and repeats and presenting the data with error bars show how consistent the methods are in terms of displaying similar results under the same conditions at different times, ensuring the reproducibility of the experiment.

3.3.3 Data validity

Data is presented as relative values, e.g cell viability of treated cells is relative to untreated cells. As such the absolute values of cell viability which represent the amount of adenosine triphosphate (ATP) is not relevant until put into comparison with other measurements.

3.3.4 Reliability of data

Since a lot of the data acquired was available in real time there was room for instantaneous assessment of measurement errors, for example a misaligned plate in a plate reader would show obvious deviations from expected values.

While one should be careful with expecting certain values from measurements, repeat checks would clearly show whether the error is based on the measurement or natural reasons. Any causes for inconsistencies between biological repeats were investigated and if a possible cause was found to be external the results were discarded and the experiment repeated. An actual example of this was a power outage clearly skewing the results of a cell viability measurement.

By repeating the same measurement at different times and observing similar results the acquired data is deemed reliable.

(20)

Methods | 11

3.4 Data Analysis

3.4.1 Software Tools

• MATLAB

• Microsoft Excel

• SoftMax Pro

• ExpressionSuite Software

(21)

Results and Analysis | 12

Chapter 4

Results and Analysis

4.1 Results

4.1.1 Western blot

Figure 4.1 –

HSP60 (arrow) detected at 647 nm with 120s exposure time, greyscale inverted.

Lanes loaded with samples as indicated on the top.

The molecular weight of the HSP60 protein is 60 kDa, the PagerRuler Plus Prestained Protein Ladder indicates that HSP60 is detected in every lane, the blot is thus deemed successful but caution will be taken when drawing conclusions from lane 4, 7 and 9 since the signal is weaker there, implying

(22)

Results and Analysis | 13

that less protein has been loaded. The red color seen on the ladder originates from a software setting which highlights oversaturated pixels.

Figure 4.2 –

CCN1 (arrow) detected at 488 nm with 120s exposure time, greyscale inverted.

Lanes loaded with samples as indicated on the top.

The molecular weight of the CCN1 protein is 42 kDa, indicating that the bands showing between the 40 kDa and 55 kDa markers correspond to CCN1. We can see a clear band for CCN1 in the positive control lane while the negative control lane is empty. In this image however the signal is rather weak and the exposure time needed increasing.

(23)

Results and Analysis | 14

Figure 4.3 –

CCN1 (arrow) detected at 488 nm with 180s exposure time, greyscale inverted.

Lanes loaded with samples as indicated on the top.

Figure 4.3 shows the same filter as in figure 4.2, but with longer exposure time in order to increase the signal intensity.

Figure 4.4 –

CCN1 (arrow) detected at 488 nm with 300s exposure time, greyscale inverted.

Lanes loaded with samples as indicated on the top.

In figure 4.4 the signal-to-noise ratio is further decreased, the membrane

(24)

Results and Analysis | 15

is overexposed and it was decided that no more detection was meaningful.

However despite overexposure there are no signs of CCN1 protein detection in lanes 5,6 and 8 (cell lines E3, E4 and H8), a strong indication of successful knockout of the CCN1 gene.

Figures 4.3 and 4.4 are deemed to contain the clearest bands of CCN1 in terms of contrast and signal-to-noise ratio, lane 4 (cell line G9) is instantly rejected as a knockout cell line, while lanes 7 and 9 (H12 and H6 respectively) are interpreted as incomplete knockouts and should be subjects of another blot for confirmation. Lanes 5,6 and 8 (E3, E4 and H8) contain no visible signal and even through a complementary blot they were confirmed as successful knockouts. It should be noted that none of these lanes have weak signals for HSP60, meaning that insufficient protein loading is an unlikely explanation for the lack of a CCN1 signal.

The cell lines that were tested in this western blot were not the same cell lines that took part in the drug efficiency and gene expression studies. Since there were few cell lines remaining to test and the cell lines in question were not actively being cultured at the time, other previously tested cell lines with active cultures were selected to keep within the time frame of the project.

(25)

Results and Analysis | 16

4.1.2 2D culture and viability

(a) WT (b) B11

(c) C3 (d) C7

Figure 4.5 – Bright-field images of untreated 2D cultures, wild-type cells (a) and knockout cell lines (b-d). Total magnification is 40x.

(26)

Results and Analysis | 17

(a) WT (b) B11

(c) C3 (d) C7

Figure 4.6 –Bright-field images of 2D culture treated with 5 µM gemcitabine, wild- type cells (a) and knockout cell lines (b-d). Total magnification is 40x.

Typical images showcasing differences in cell viability for different cell lines and treatment concentrations. We can see a clear difference in viability for the cell lines treated with gemcitabine.

(27)

Results and Analysis | 18

Figure 4.7 –

Cell viability upon gemcitabine treatment relative to untreated cells. Cells were treated with concentrations between 1 µM and 50 µM and the viability assay was

used after 3 days.

The corresponding viability of each cell line when treated with gemcitabine is presented above, * corresponds to p<0.05 while ** and *** corresponds to p<0.01 and p<0.001 respectively. We observe a clear trend from the viability assay that the viability is higher for the knockout cell lines in 2D culture compared to the WT Panc1 cells. The difference in viability with increased concentration of gemcitabine is only mild, roughly 10% with a 50-fold increase in concentration.

(28)

Results and Analysis | 19

4.1.3 3D culture and viability

(a) WT (b) B11

(c) C3 (d) C7

Figure 4.8 – Bright-field images of untreated 3D cultures, wild-type cells (a) and knockout cell lines (b-d). Total magnification is 40x.

(29)

Results and Analysis | 20

(a) WT (b) B11

(c) C3 (d) C7

Figure 4.9 –Bright-field images of 3D culture treated with 5 µM gemcitabine, wild- type cells (a) and knockout cell lines (b-d). Total magnification is 40x.

Typical images showcasing differences in cell viability and aggregation for different cell lines and treatment concentrations. The difference in viability is not as obvious as in the 2D culture images, however we note a clear difference in cell aggregation of the untreated spheroids. The WT and C3 spheroids are more dense than B11 and C7, meaning the latter two should be more susceptible to treatment as the drug will have easier access to more cells.

(30)

Results and Analysis | 21

Figure 4.10 –

Cell viability upon gemcitabine treatment in 3D culture relative to untreated cells.

Cells were treated with concentrations between 1 µM and 50 µM and the viability assay was used after 3 days.

The corresponding viability of each cell line when treated with gemcitabine is presented above, * corresponds to p<0.05 while ** and *** corresponds to p<0.01 and p<0.001 respectively. We observe an increase in cancer cell viability from the viability assay for the knockout cell lines compared to the WT cells for all concentrations of gemcitabine, the trend being C3 having the highest viability followed by C7 and B11 respectively. This is what was predicted by visually analyzing the spheroids in figure 4.9 and arguing that the least dense spheroids are more susceptible to the drug. However a larger sample size is required to decide if this is a general effect or a coincidence.

(31)

Results and Analysis | 22

4.1.4 Gene expression

Figure 4.11 –

Relative gene expression ofdeoxycytidine Kinase (DCK)andsolute carrier family 29 member 1 (SLC29A1)in monospheroids.

Presented here is the expression of two genes, DCK and SLC29A1. DCK encodes an enzyme which activates gemcitabine, meaning, ceteris paribus, an increase in expression would lead to increased efficiency of gemcitabine treatment. We see an overall trend for a decrease inDCKexpression for the knockout cell lines, making it a possible contributor to the increase in viability seen in figure 4.11. SLC29A1encodes a transporter responsible for the uptake of cytotoxic substances such as gemcitabine, and studies have shown that a higher expression is associated with improved survival[11]. We see a large variation in SLC29A1expression between the knockout cell lines, almost a factor of 2 between the C7 and B11 lines. A larger sample size is required for any attempts of interpretation of this gene expression.

(32)

Results and Analysis | 23

4.2 Reliability Analysis

Cell lines were tested twice for knockout efficiency, knockout was confirmed when two separate western blots had consistent results. All results from 4.1.2 through 4.1.4 were accumulated over three methodologically identical technical repeats, and as the results were reproducible they are deemed reliable.

4.3 Validity Analysis

4.3.1 Western blot

As a well established immunoassay for selective protein detection the validity of the western blot comes down to the choice of antibodies. The antibodies used here had previously been picked out with certain prerequisites to ensure valid detection. The primary antibodies (anti-CCN1 and anti-HSP60) were chosen as good antibody-antigen matches, while the secondary antibodies were chosen to detect the primary antibodies and nothing else.

4.3.2 Viability assay

The assay kit used for viability analyses the amount of living cells in a space based on the amount of ATPpresent in the cell medium. As there are other factors capable of affecting the amount of ATP in the medium the validity of the assay is questioned and further discussed in 6.2.

(33)

Discussion | 24

Chapter 5

Discussion

5.1 Viability observations

5.1.1 Spheroid aggregation and physical traits

Examining figure 4.9 there are striking differences in the appearance of the spheroids, differences that were in fact general and not only observed for these specific images. We see disparities in the spheroids’ abilities to aggregate, spheroids consisting of certain cell lines become more dense than others.

This denseness is something mentioned in 1.2 as a contributing factor to a tumor’s chemoresistance, in this case we do not have a stromal component but the effect remains the same. The fact that the relative viability between the tested cell lines follows the visual sensation of how dense the spheroids are is something that should be further investigated with more cell lines, since there is a possibility that it is a correlation to be found. As such it is also advisable to adapt or develop a tool capable of quantifying the degree of aggregation in spheroids. The reason for the differences in aggregation is also something that should be investigated, suggestions at reasons to look into are for example off- target effects from the CRISPR/cas9 technique or genetic differences between the cell lines caused by mutations after keeping separate cultures through numerous passages.

As mentioned in 1.2 the effect of the dense stroma in PDAC causes an increase in interstitial fluid pressure inside the tumor leading to increased chemoresistance. Apart from this pressure increase the tumor is likely to display changes in other physical parameters relative to healthy tissue.[12]

These parameters are all either tumorigenic or detrimental to the tumor’s surrounding tissues and would thus together with cell proliferation (viability)

(34)

Discussion | 25

be relevant to investigate. There are examples of successful attempts to quantify some of these parameters on 3D cell culture models similar to the one described here [13][14][15][16][17], and if adapted to this research they could provide valuable information.

5.1.2 Off-target effects

As mentioned in chapter2the CRISPR/cas9 technique often comes with off- target effects. As such it is possible that the overall shift towards higher cell viability for knockout cell lines in figures 5.8 and 5.11 is due to these effects rather than the actual CCN1 knockout. Something that may suggest this is actually the case is a previous study showing the opposite in cell viability on CCN1 knockout[18]. Interestingly the method of knockout in the aforementioned study was shRNA gene silencing, a different approach than CRISPR/cas9. This strengthens the argument for off-target effects being responsible for the effects seen in viability measurements since in both cases CCN1 knockout is confirmed but the techniques to achieve this differ along with the result.

5.2 Assays

Since viability assay reagent is added to the cell medium at a 1:1 volume ratio it is imperative to keep the cell medium from evaporating during incubation, and if this were to happen it needs to be corrected by adding a smaller volume of reagent. It should also be noted that as the assay relies on indirect measurements, other factors affecting the amount ofATPin the cell medium is not considered. In order to partly circumvent these factors and more directly observe the parameter in question, i.e. cell viability (or death), an apoptosis assay is better adapted. Spheroid cultures for said assay was prepared, this is further discussed in6.2.1.

(35)

Conclusions and Future work | 26

Chapter 6

Conclusions and Future work

6.1 Conclusions

The observations regarding CCN1 knockout Panc1 cell viability when treated with gemcitabine indicate an increase in viability relative to the cells’ wild- type counterpart. These results contradict earlier similar work[18], however some methodological differences exist which are discussed in chapter 5that could be responsible for the discrepancy. A larger sample size and further analysis could as previously mentioned provide understanding not only about the discrepancy but also generally about the role of crosstalk in the high chemoresistance of PDAC.

6.2 Future work

6.2.1 Unfinished work

As previously mentioned cell culture plates with spheroids which were meant to be probed with the M30 apoptosis assay were prepared, however the measurements did not fall within the time and scope of this project. It should be mentioned that this apoptosis assay has the major advantage of only detecting cell death in epithelial cells, and thus can be used to measure apoptosis of cancer cells also in the presence of stromal cells.

6.2.2 Possibilities of future work

The knockout of of CCN1 seemingly increases the viability of Panc1 cells treated by gemcitabine as seen in figures 4.8 and 4.11. If these results are

(36)

Conclusions and Future work | 27

confirmed when testing more cell lines it would be of great relevance to see if the opposite is indicated on upregulation of CCN1. This would also help with understanding whether the changes in viability is due to the knockout itself or due to off-target effects. With the case of off-target effects from the CRISPR/cas9 technique further studies like whole exome or whole genome sequencing should be made to investigate what the actual effects are in terms of changes in the cells’ genome. If and when these effects are identified, a method of exclusion could be adapted in order to find factors relevant to the performed measurements. If for example an off-target effect is identified as an unwanted alteration on the cells’ genome, this alteration and its possible effects can in first hand be analyzed based on what is known about the function of that part of the genome. Should it be suspected that this function has direct or indirect consequences to the measured parameters, its corresponding gene could be used as a new target for CRISPR/cas9. With this approach it is possible to not only investigate off-target effects but also identify new target genes and work towards the purpose of further understanding crosstalk as stated in 1.3.

6.3 Reflections

Since I had no substantial previous experience in working with living cells the first period of time consisted of observing procedures and performing simpler supervised tasks. This included cell maintenance such as changing cell medium as well as splitting and freezing cells, all the while memorizing protocols and procedures. For the next few weeks I learned to create cell lysates for upcoming western blots, and later also the western blot itself. When enough cell lines had been analyzed it was time to start 3D cultures and specifically to grow spheroids, which as usual was done by first observing and denoting protocols before attempting it supervised and eventually being able to do it unsupervised. This led to growing spheroids in bulk over a few weeks, which obviously also involved treating said spheroids, imaging, performing assays and preparing lysates for real time RT-PCR. During this period of time I also had ongoing cell cultures in a hypoxia chamber, passaging cells while gradually decreasing the oxygen level to better correspond to the oxygen-poor microenvironment found in pancreatic cancer. Results from this have however not been included in this report.

I take many positive experiences from the work, including the satisfaction of having learned so much from a new field of science. But perhaps mostly the realization that my education has not only taught me first and foremost specific topics rather the ability to quickly tackle previously unknown subjects

(37)

Conclusions and Future work | 28

and adapt to them.

(38)

REFERENCES | 29

References

[1] F. Bray, J. Ferlay, I. Soerjomatara, R. Siegel, L. Torre, and A. Jemal, “Global cancer statistics 2018: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries,”

CA: A Cancer Journal for Clinicians, vol. 2018 Nov;68(6):394- 424., Sep. 2018. doi: 10.3322/caac.21492. [Online]. Available:

https://pubmed.ncbi.nlm.nih.gov/30207593/

[2] J. Schnittert, R. Bansal, and J. Prakash, “Targeting pancreatic stellate cells in cancer,” Trends in cancer, vol. 2019 Feb;5(2):128-142, Feb. 2019. doi: 10.1016/j.trecan.2019.01.001. [Online]. Available:

https://pubmed.ncbi.nlm.nih.gov/30755305/

[3] D. Thomas and P. Radhakrishnan, Pancreatic Stellate Cells: The Key Orchestrator of The Pancreatic Tumor Microenvironment. Cham: Springer International Publishing, 2020, pp. 57–70. ISBN 978-3-030-37184-5. [Online]. Available: https://doi.org/10.1007/

978-3-030-37184-5_5

[4] X. Liu, B. Gündel, X. Li, J. Liu, A. Wright, M. Löhr, G. Arvidsson, and R. Heuchel, “3d heterospecies spheroids of pancreatic stroma and cancer cells demonstrate key phenotypes of pancreatic ductal adenocarcinoma,” Translational Oncology, vol. 14, no. 7, p. 101107, 2021. doi: https://doi.org/10.1016/j.tranon.2021.101107.

[Online]. Available: https://www.sciencedirect.com/science/article/pii/

S1936523321000991

[5] K. P. Olive, M. A. Jacobetz, C. J. Davidson, A. Gopinathan, D. McIntyre, D. Honess, B. Madhu, M. A. Goldgraben, M. E. Caldwell, D. Allard, K. K. Frese, G. DeNicola, C. Feig, C. Combs, S. P.

Winter, H. Ireland-Zecchini, S. Reichelt, W. J. Howat, A. Chang, M. Dhara, L. Wang, F. Rückert, R. Grützmann, C. Pilarsky,

(39)

REFERENCES | 30

K. Izeradjene, S. R. Hingorani, P. Huang, S. E. Davies, W. Plunkett, M. Egorin, R. H. Hruban, N. Whitebread, K. McGovern, J. Adams, C. Iacobuzio-Donahue, J. Griffiths, and D. A. Tuveson, “Inhibition of hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer,” Science, vol. 324, no. 5933, pp.

1457–1461, 2009. doi: 10.1126/science.1171362. [Online]. Available:

https://science.sciencemag.org/content/324/5933/1457

[6] X.-H. Zhang, L. Y. Tee, X.-G. Wang, Q.-S. Huang, and S.-H.

Yang, “Off-target effects in crispr/cas9-mediated genome engineering,”

Molecular Therapy - Nucleic Acids, vol. 4, p. e264, 2015. doi:

https://doi.org/10.1038/mtna.2015.37. [Online]. Available: https://www.

sciencedirect.com/science/article/pii/S216225311630049X

[7] “Ccn1 cellular communication network factor 1 [ homo sapiens (human) ],”https://www.ncbi.nlm.nih.gov/gene/3491, accessed: 2021-05-18.

[8] P. Longati, X. Jia, J. Eimer, A. Wagman, M.-R. Witt, S. Rehnmark, C. Verbeke, R. Toftgård, M. Löhr, and R. L. Heuchel, “3d pancreatic carcinoma spheroids induce a matrix-rich, chemoresistant phenotype offering a better model for drug testing,” BMC Cancer, vol. 13, no. 1, Feb. 2013. doi: 10.1186/1471-2407-13-95. [Online]. Available:

https://doi.org/10.1186/1471-2407-13-95

[9] L.-B. Weiswald, D. Bellet, and V. Dangles-Marie, “Spherical cancer models in tumor biology,” Neoplasia, vol. 17, no. 1, pp. 1–15, 2015. doi:

https://doi.org/10.1016/j.neo.2014.12.004. [Online]. Available: https:

//www.sciencedirect.com/science/article/pii/S1476558614001948 [10] R. L. Siegel, K. D. Miller, and A. Jemal, “Cancer statistics, 2015,”

CA: A Cancer Journal for Clinicians, vol. 65, no. 1, pp. 5–29, 2015.

doi: https://doi.org/10.3322/caac.21254. [Online]. Available: https:

//acsjournals.onlinelibrary.wiley.com/doi/abs/10.3322/caac.21254 [11] B. Mohelnikova-Duchonova and B. Melichar, “Human equilibrative

nucleoside transporter 1 (hent1): Do we really have a new predictive biomarker of chemotherapy outcome in pancreatic cancer patients?” Pancreatology, vol. 13, no. 6, pp. 558–563, 2013. doi:

https://doi.org/10.1016/j.pan.2013.09.005. [Online]. Available: https:

//www.sciencedirect.com/science/article/pii/S1424390313008119

(40)

REFERENCES | 31

[12] H. T. Nia, L. L. Munn, and R. K. Jain, “Physical traits of cancer,” Science, vol. 370, no. 6516, 2020. doi: 10.1126/science.aaz0868. [Online].

Available: https://science.sciencemag.org/content/370/6516/eaaz0868 [13] D. Jaiswal, N. Cowley, Z. Bian, G. Zheng, K. P. Claffey, and K. Hoshino,

“Stiffness analysis of 3d spheroids using microtweezers,” PLOS ONE, vol. 12, no. 11, pp. 1–21, 11 2017. doi: 10.1371/journal.pone.0188346.

[Online]. Available: https://doi.org/10.1371/journal.pone.0188346 [14] ——, “Stiffness analysis of 3d spheroids using microtweezers,”

PLOS ONE, vol. 12, no. 11, p. e0188346, 2017.

doi: 10.1371/journal.pone.0188346. [Online]. Available:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5699838/andhttps:

//journals.plos.org/plosone/article/file?id=10.1371/journal.pone.

0188346&type=printable

[15] S. Monnier, M. Delarue, B. Brunel, M. E. Dolega, A. Delon, and G. Cappello, “Effect of an osmotic stress on multicellular aggregates,” Methods, vol. 94, pp. 114–119, 2016. doi:

https://doi.org/10.1016/j.ymeth.2015.07.009 Experimental Approaches in Mechanotransduction. [Online]. Available: https://pubmed.ncbi.nlm.

nih.gov/26210402/

[16] W. Lee, N. Kalashnikov, S. Mok, R. Halaoui, E. Kuzmin, A. Putnam, S. Takayama, M. Park, L. Mccaffrey, R. Zhao, R. Leask, and C. Moraes,

“Dispersible hydrogel force sensors reveal patterns of solid mechanical stress in multicellular spheroid cultures,” Nature Communications, vol. 10, 01 2019. doi: 10.1038/s41467-018-07967-4. [Online].

Available: https://www.nature.com/articles/s41467-018-07967-4 [17] M. Dolega, M. Delarue, F. Ingremeau, J. Prost, A. Delon,

and G. Cappello, “Cell-like pressure sensors reveal increase of mechanical stress towards the core of multicellular spheroids under compression,” Nature Communications, vol. 8, p. 14056, 01 2017. doi: 10.1038/ncomms14056. [Online]. Available: https:

//www.nature.com/articles/ncomms14056

[18] G. Maity, A. Ghosh, V. Gupta, I. Haque, S. Sarkar, A. Das, K. Dhar, S. Bhavanasi, S. Gunewardena, D. Hoff, S. Mallik, S. Kambhampati, S. Banerjee, and S. Banerjee, “Cyr61/ccn1 regulates dck and ctgf and causes gemcitabine resistant phenotype in pancreatic

(41)

REFERENCES | 32

ductal adenocarcinoma,” Molecular Cancer Therapeutics, vol. 18, p.

molcanther.0899.2018, 02 2019. doi: 10.1158/1535-7163.MCT-18- 0899. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/30787177/

(42)

Appendix A: Methylcellulose stock solution | 33

Appendix A

Methylcellulose stock solution

The preparation of methocel stock solution is very critical. If the concentration of methocel is too low or the solution is containing methylcellulose debris, single cells will stick to the wall and several small spheroids are formed in each well. We use methylcellulose from sigma (cat: m-0512, Sigma Aldrich; 4000 centipoises). The methocel stock solution should have an extremly high viscosity. We autoclave the pure powder (6g) in a 500ml flask containing a magnetic stirrer (the methylcellulose powder is resistant to this procedure). The autoclaved methylcellulose is dissolved in preheated 250ml basal medium (DMEM-F12; 60°C) for 20min (using the magnetic stirrer). Thereafter, 250ml basal medium including twice the amount of FBS, Pen/Strep (room temperature) is added to a final volume of 500ml and the whole solution is mixed/stirred slowly over night at 4°C.

The final stock solution (1.2%) is aliqoted and cleared by centrifugation (5000g, 2h, room temperature). Only the clear highly viscous supernatant should be used for the spheroid assay (about 90-95% of the stock solution).

Keep this stocks solution at 4 °C until use. For spheroid generation we use 20% of the stock solution and 80% culture medium resulting in a final 0.24% methylcellulose concentration.

(43)

Flerartiga 3D-heterosfäroider av bukspottkörtelns stroma- och adenokarcinomceller uppvisar inflammatorisk fenotyp av Panc1- celler under näringsfattig miljö

Abstrakt

Vi karaktäriserade en avancerad flerartig samodlingsmodell i 3D av PDAC, som efterliknade både tumör och CAF- överhörning vid dåligt näringstillstånd in vitro. Vi fann att biologin hos PDAC-celler och stromala bukspottskörtelceller inte bara påverkas av 3D-samodling utan även av tillgång till näring.

Introduktion

Duktal adenokarcinom i bukspottskörteln (PDAC) är en av de mest dödliga formerna av cancer, och kännetecknas av tät desmoplasi, näringsfattig miljö och nästan fullständig terapiresistens.

Material/Metoder

Vi odlade monosfäroider och heterosfäroider bestående av murina stellatceller från bukspottskörtel (mPSC) och humana PDAC-celler (Panc1) i DMEM- F12, kompletterat med antingen 10% fetalt bovint serum (FBS) eller 0.1% FBS / 0.3% bovint serumalbumin (BSA) som respektive motsvarigheter till de näringsrika och näringsfattiga miljöerna (NRM/NFM). RNA isolerades från intakta sfäroider för RNA-sekvensering, följt av en in silico, artspecifik separation av RNA-sekvensläsningar och undersökning av transkription (GSEA och analys av molekylär subtyp för PDAC).

Målet med postern

Vi utvecklade en heterosfäroidmodell i 3D för att studera kommunikationen mellan PDAC-celler och bukspottkörtelns stromaceller (pancreatic stellate cells, PSCs) i en näringsfattig miljö och jämförde med den i en näringsrik miljö.

Resultat

I NFM var generna som var till övervägande del uppreglerade i Panc1, odlade som heterosfäroider jämfört med monosfäroider, relaterade till inflammatorisk signalering, inklusive interferon alfa- och TNF-signalering via NF-κB. Gener som indikerar ökad celldelning var emellertid huvudsakligen uppreglerade i Panc1 från heterosfäroider under NRM.

Anrikningsanalys av genuppsättning, sammanfattade av genuppsättningarna som anrikades i Panc1 från monosfäroider (NES-) och heterosfäroider (NES+) I NFM (A) och NRM (B) med FDR q-värde

<0.05

Att upptäcka den molekylära mekanismen för en aggressiv subtyp av Panc1 vid samodling med mPSC, och den molekylära mekanismen för iCAF / myCAF- bildning.

Karolinska Institutet Xinyuan Liu PhD student

Hälsovägen 7, Novum, KFC, Plan 5F, Room 5-21, PaCaRes Lab

141 86 Huddinge Sweden

E-post: xinyuan.liu@ki.se

Logo för Sjukhus/Universitet Xinyuan Liu, Beate Gündel, Xidan Li, Jianping Liu, Martin Sundin, Anthony Wright, Matthias Löhr, Gustav Arvidsson and Rainer Heuchel

Representativ SEM bild av Panc1/mPSC heterosfäroidmodellen i NFM

Resultat

Oberoende av serumkoncentrationen observerade vi en förskjutning av Panc1 från den klassiska till den skvamösa/basala fenotypen vid samodling med mPSC.

Jämförelser av DEG för Panc1 odlade i hetero- vs monosfäroider med PDAC-stratifieringsgensignaturerna från Moffitt et al. (A, P = 8.45×10-3) och Bailey et al. (B, P = 1.52×10-5) I NFM, såväl som I NRM (C, P = 4.13×10-5; D, P = 5.46×10-8).

A B

B A

Resultat

Intressant nog antog mPSCs en mer inflammatorisk cancerassocierad fibroblastisk (iCAF) fenotyp vid samodling med Panc1 i NFM, medan de uppvisade en mer myofibroblastisk CAF-fenotyp i NRM.

A B

Jämförelse av DEG för mPSC odlade i hetero- vs monosfäroider med my- / iCAF-signaturer I NFM (A, P = 3.15×10-12) och NRM (B, P = 0.073).

C D

Appendix B

Poster

Figure B.1 –

A poster I was asked to give advice about.

(44)

For DIVA

{"Author1": {

"Last name": "Sundin",

"First name": "Martin",

"E-mail": "marsundi@kth.se",

"organisation": {"L1": "School of Engineering Sciences ", }, }

"Degree": {"Educational program": "Master’s Programme, Engineering Physics, 120 credits"},

"Title": {

"Main title": "Knockout studies of Panc1 cells",

"Subtitle": "Effect of CCN1 gene knockout on Panc1 cell viability",

"Language": "eng" },

"Alternative title": {

"Main title": "Knockout studier av Panc1-celler",

"Language": "swe"

},"Supervisor1": {

"Last name": "Heuchel",

"First name": "Rainer",

"E-mail": "rainer.heuchel@ki.se",

"Other organisation": "Karolinska institutet, H9 Klinisk vetenskap, intervention och teknik"}

"Examiner1": {},

"Last name": "Brismar",

"First name": "Hjalmar",

"E-mail": "brismar@kth.se",

"organisation": {"L1": "School of Engineering Sciences ",

} },

"Cooperation": { "Partner_name": "Karolinska institutet"},

"Other information": {

"Year": "2021", "Number of pages": "vii,35"}

}

References

Related documents

Stöden omfattar statliga lån och kreditgarantier; anstånd med skatter och avgifter; tillfälligt sänkta arbetsgivaravgifter under pandemins första fas; ökat statligt ansvar

46 Konkreta exempel skulle kunna vara främjandeinsatser för affärsänglar/affärsängelnätverk, skapa arenor där aktörer från utbuds- och efterfrågesidan kan mötas eller

Data från Tyskland visar att krav på samverkan leder till ökad patentering, men studien finner inte stöd för att finansiella stöd utan krav på samverkan ökar patentering

For the study, we used knockout mice that lack receptors of GLP-1R and compared the data collected to the estimated numbers of the cells in brain from wild type mice.. The

Re-examination of the actual 2 ♀♀ (ZML) revealed that they are Andrena labialis (det.. Andrena jacobi Perkins: Paxton &amp; al. -Species synonymy- Schwarz &amp; al. scotica while

common than the social incentives. The outcome space shows an overlap between students and teachers as regards conceptions of the nature of learning. This overlap is

Kraven på att differentiera de studerande mer än för närvarande leder oavsett vilken form av examination som tillämpas till en ökad arbetsbörda.. Detta måste rimligen betalas

För det tredje har det påståtts, att den syftar till att göra kritik till »vetenskap», ett angrepp som förefaller helt motsägas av den fjärde invändningen,