Mlh1 deficiency in normal mouse colon mucosa associates with chromosomally unstable colon cancer
Mlh1 deficiency in normal mouse colon mucosa associates with chromosomally unstable colon cancer
Marjaana Pussil a 3
Petri Törönen 1
Elisabet Einarsdott 0
Shintaro Katayam 0
Kaarel Krjutško 7
Liisa Holm 1 3
Juha Kere 0 5 7
Päivi Peltomäki 4
Markus J.Mäkinen 8 9
Jere Linden 2
Minna Nyström 3
0 Dde,partment of Biosciences and Nutrition, Karolinska Institutet , SE-14183 Huddinge , Sweden
1 Institute of Biotechnology, University of Helsinki , FI-00014 Helsinki, Finla n
2 dDepartment of Basic Veterinary Sciences, FCLAP, University of Helsinki , FI-00014 Helsinki , Finland
3 Faculty of Biological and Environmental Sciences, Molecular and Integrative Biosciences Research Programme
4 MK,edicum, Department of Medical and Clinical Genetics, University of Helsinki , FI-00014 Helsinki, Finla
5 Department of Genetics and Molecular Medicine, King's College London , London SE1 9RT, U
6 C, ompetence Centre on Health Technologies , 50410 Tartu , Estonia
7 F,olkhälsan Institute of Genetics, Molecular Neurology Research Program, University of Helsinki , FI-00014 Helsinki , Finland
8 Md,edical Research Center Oulu, Oulu University Hospital, University of Oulu , FI-90014 Oulu , Finland an
9 nCda,ncer and Translational Medicine Research Unit, Department of Pathology, University of Oulu , FI-90014 Oulu, Finla n
Colorectal cancer (CRC) genome is unstable and different types of instabilities, such as chromosomal instability (CIN) and microsatellite instability (MSI) are thought to reflect distinct cancer initiating mechanisms. Although 85% of sporadic CRC reveal CIN, 15% reveal mismatch repair (MMR) malfunction and MSI, the hallmarks of Lynch syndrome with inherited heterozygous germline mutations in MMR genes. Our study was designed to comprehensively follow genome-wide expression changes and their implications during colon tumorigenesis. We conducted a long-term feeding experiment in the mouse to address expression changes arising in histologically normal colonic mucosa as putative cancer preceding events, and the effect of inherited predispositionM(lh1+/−) and Western-style diet (WD) on those. During the 21-month experiment, carcinomas developed mainly in WD-fed mice and were evenly distributed between genotypes. Unexpectedly, the heterozygote (B6.129M-lh1tm1Rak) mice did not show MSI in their CRCs. Instead, both wildtype and heterozygote CRC mice showed a distinct mRNA expression profile and shortage of several chromosomal segregation gene-specific transcripts (Mlh1, Bub1, Mis18a, Tpx2, Rad9a, Pms2, Cenpe, Ncapd3, Odf2 and Dclre1b) in their colon mucosa, as well as an increased mitotic activity and abundant numbers of unbalanced/atypical mitoses in tumours. Our genome-wide expression profiling experiment demonstrates that cancer preceding changes are already seen in histologically normal colon mucosa and that decreased expressions of Mlh1 and other chromosomal segregation genes may form a field-defect in mucosa, which trigger MMR-proficient, chromosomally unstable CRC.
Colorectal cancer (CRC) is the third most common cancer including diet, are suggested to play a critical role in its etiology
and the fourth most common cause of cancer-related deaths (
). Cancer development always includes lack of genomic
worldwide. The incidence rates increase significantly with age integrity in cells and different types of genomic instability, such
and interactions between genetic and environmental factors, as chromosomal instability (CIN) and microsatellite instability
shown to cause CRCs in mice even without any predisposing
mutation or carcinogen treatment14(–17). The mouse model
provided a valuable tool to study the process of carcinogenesis
from the earliest changes in colon mucosa until tumour de-vel
opment and characterization. Moreover, the use of an animal
model enabled to distinguish gene expression changes caused
by different risk factors, such as age, inherited predisposition
and diet and sort out the ones that signal carcinogenesis.
Materials and methods
Mice, experimental study and diets
Heterozygote B6.129-Mlh1tm1Rak mice (Mlh1+/−) strain 01XA2 (18) and the
C57BL/6 strain were obtained from NCI-MMHCC; National Institutes of
(MIN, also called MSI) are thought to reflect distinct initiatingHealth, Mouse Repository, NCI-Frederick, MD. Altogether 12 animals (equal
mechanisms in cancer (3). Three different pathways leading to numbers of sexes), the Mlh1+/− mice and their wild-type C57BL/6 mates,
genomic instability in colon cancer have been described. Most formed six breeder pairs which produced the mouse colony used in our
CRCs represent CIN, where chromosomes fail to trigger the study. Mice were genotyped F(igure 1C) using genomic DNA extracted from
spindle assembly checkpoint (SAC) leading to aberrant chro m-o earmarks according to the protocol published in our previous wo1r9k).(The
some segregation. In recent years, many new genes have been mice were bred and treated according to the study protocol approved by the
reported, whose mutations and expression changes disturb National Animal Experiment Board in Finland (ESLH-2008-06502/Ym-23).
chromosomal stability causing aneuploidy and/or compreh-en The Mlh1 heterozygote mice and their homozygote wild-type lit-ter
sive loss of heterozygosity (LOH) and alterations in chromosome mates were divided into two dietary groups at the age of 5 weeks. The
structure4(,5). About 15% of sporadic CRCs and over 95% of CRCs mWiDcemwoedrifeiefdedfrowmithAINeit(hHearrlhaenaTltehklyardo,dMeandtiscoonn,tWroI)l1(9d)ietto ArIeNse-9m3bG2l0e)(, oorn
in Lynch syndrome (LS), the most common inherited colon ca-n the nutritional level, the diet consumed in human Western population
cer syndrome, represent MSI caused by a defective DNA mi-s (high fat and energy content, low amounts of fiber, calcium and vitamin
match repair (MMR) mechanism 6(). MMR deficiency causes D3) (Supplementary Table )1(19) Twelve mice per each group M(lh1+/+ AIN,
accumulation of point mutations in the genome and especially Mlh1+/− AIN, Mlh1+/+ WD, Mlh1+/− WD) with equal representation of sexes, at
in short repeat sequences called microsatellites, and is thought time point (tp) 0 (5 weeks of age,Mlh1+/−, Mlh1+/+), tp1 (12 months of age),
to be the driver defect in MSI carcinomas7(). The third path- tp2 (18 months of age) and tp3 (21 months of age), 168 mice in total were
way, CpG Island Methylator Phenotype (CIMP), characterized by sacrificed and sampled.
global genome hypermethylation and tumour suppressor gene
silencing, is seen in 20–30% of CRCs (
). Collection of tumours and normal colon mucosa
There are different ways to induce MMR malfunction and samples
consequently lead to MSI. In sporadic tumours, the most co-m All observed colon tumours were collected under dissecting microscope
mon mechanism is the MLH1 (mutL homolog 1) promoter hyper- and preserved as FFPE blocks. If a tumour was large enough (3–5 mm in
methylation often associated with BRAF V600E mutation8).( diameter), approximately half of it was embedded in O.C.T compound
Also, in CIMP tumours theMLH1 is often inactivated by promoter (iVngWsRa,nRdadtnhoerg,rPaedninnsgylovfanneioap) lfoarsicarsywoesraemcpalrirnige.dHoiusttoaltogTihcealFisntnuidsihesC, esnt-atirne
hypermethylation but usually caused by aging and environm-en for Laboratory Animal Pathology (FCLAP), University of Helsinki, Finland.
tal factors9(). In LS, the first mutation is inherited, mainly in The neoplasias were graded as hyperplasias, adenomas and carcinomas
MLH1 (40%), MSH2 (34%) or MSH6 (18%) (10) and generally fo-l according to criteria based on consensus rodent intestinal cancer no m-en
lowed by LOH and somatic inactivation of the MMR gene8)(. In clature 2(
), (Supplementary Table )2. Longitudinal pieces (excluding the
LS, the second hit leading to MSI rarely happens through p-ro previously harvested tumorous sections), representing approximately one
). The fact that inactivation oMf LH1 third of the proximal mouse colon were collected for normal mucosa. The
is associated with most MSI tumours irrespective of the ina-cti mucosa was separated from the underlying submucosa and muscu-la
vating mechanism emphasizesMLH1 deficiency as a key defect ture under a dissecting microscope and samples for RNA extraction were
shared by sporadic and inherited CRC. stored in RNAlater (Qiagen, Hilden, Germany) at −80°C13().
Colon cancer research focuses mainly on tumour char-ac Transcriptome analysis of normal mucosa
teristics, such as genomic instability, which can be utilized in
treatment design. Recent findings have revealed that CIN and (TRrNanA-sscerqip).tToomtael aRnNaAlywsaiss pwraespapreerdfofrrommed0,u1s2inagndR1N8A m-soenquthenscoilndgmmiceeth(1o4d,
MSI pathways are not mutually exclusiv4e,1(2), suggesting that 40 and 40 mice, respectively) using the RNeasy Plus Kit (Qiagen) with an
also tumours with distinct features and instabilities may share extra DNase treatment (Qiagen). The RNA concentration was measured by
initiative genomic aberrations while different tumour cha-rac Qubit 1.0 (Thermo Fisher Scientific, Waltham, MA) and RNA integrity with
teristics reflect subsequent alterations during cancer deve-lop the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Only
ment. Since MLH1 deficiency seems to be a key defect shared high quality RNA (RNA integrity number RIN ≥ 8) qualified for expression
by sporadic and inherited microsatellite unstable CRC, our aim analysis (13).
was to follow howMLH1 would contribute to colon cancer dev-el RNA-seq method followed the single-cell tagged reverse transcription
opment. Here, we used a mouse model to study cancer prece-d (STRT) (22) protocol with modifications23(). Briefly, 10 ng of total RNA was
ing expression changes in colon mucosa, Mlh1 phenotype in converted to cDNA and amplified to form an Illumina-compatible library.
tumours and the effect of inherited predispositionM(lh1+/−) and iIdnetnottifaile,r2s5wPeCrRe caypcplleisedw, eornelyustehde, baubtsoalsutfoeunrubmasbee-rpoafiruunniqiquueermeaodlsecwuelraer
Western-style diet (WD) on those1(3). We conducted a long- included in the subsequent analysis. The samples were sequenced on a
term feeding experiment with either a healthy rodent diet AIN- total of six lanes of Illumina HiSeq2000, further processed to fastq files by
93G (AIN) or WD modified from AIN. WD was used to ensure the Casava 1.8.2 (both Illumina, San Diego, CA). Quality control was performed
development of colon carcinomas, since it has previously been using the STRTprep pipeline h(ttps://github.com/shka/STRTpre)p(
Normalizing the RNA-seq data
processed reads were aligned by TopHat2 to the mouse RefSeq mm9 r-ef multidimensional data while trying to preserve the pair-wise distances
erence genome. STRT captures sequences at the ′5-end of poly(A)+ RNAs of samples from the multi-dimensional data. PlotMDS distributed in the
and the aligned reads therefore tend to be distributed close to th′-e 5 Limma package was used as a basis of the analysis2(
), although modified
end (start site) of genes. STRTprep counts only the aligned reads at the so that we were able to use any selected score to pick the genes that were
5′-untranslated region of protein-coding genes, or within the proximal used to calculate pair-wise distances13().
(500 bp) upstream region 1(3). The activity ofMlh1 was visualized with ComBat normalized data.
Samples were grouped based on the sample types (genotype, diet and
time-point) to highlight the sample differences.
STRTprep pipeline generated a read count matrix, with genes as rows Pathway analysis
and samples as columns. Different sample library sizes were normalized
using DESeq-style normalization24(). Next shifted log transformation To study the biological functions and pathways enriched among the top
(xlog = log(x +1)) was done to generate more Gaussian like data and the separating genes, we used QIAGEN’s Ingenuity Pathway Analysis (IPA
ComBat program was used to filter batch effects. These preprocessing steps Software 7.0, Qiagen). Here, we analysed both the top 100 and top 300
and alternative pipelines were evaluated by looking at the hierarchical-clus genes, which were found to separate the normal mucosa expression p-at
tering of samples and by plotting quantiles of expression values for each terns in carcinoma mice from the others. The settings for a core analysis
sample (Supplementary Informatio,navailable atCarcinogenesis Online) (13). were as follows: ingenuity knowledge base (genes and endogenous chem- i
cals) with both direct and indirect relationships, default network i-nter
action settings (include endogenous chemicals, 35 molecules per network
Tests for differential gene expression and 25 networks per analysis). Data sources were used with stringent c-on
Since the analyzed data was not any more integer count values after fidence (experimentally observed and high predicted) and data obtained
ComBat normalization, we tested threte-test based methods, Voom- in all species was selected with a relaxed filte1r3)(.
Limma, Cyber-T and Shrinkage-T for analysis of differential gene expr-es
sion. All these methods add a prior to variance estimate. Shrinkage-T MSI and LOH analyses
is the only method here that allows also testing with unequal variance. The MSI status was analysed from all the seven carcinomas (twMolh1+/+
This turned out to be important, as the genes with strongest separation WD, four Mlh1+/− WD and one Mlh1+/+ AIN mice, Table 1) from which a
between the sample groups had small variance in the analysed subset and cryo-preserved sample was available, using four dinucleotide (D18Mit15,
medium variance among remaining samples. D14Mit15, D10Mit2, D7Mit91) and two mononucleotide (JH104, U12235)
The three methods were evaluated by viewing the separation of cancer markers (27). Tumour DNA samples were extracted from the cryo-p-re
samples from the remaining samples in the multidimensional Scaling (MDS) served colon carcinomas using laser micro-dissection for cutting (Zeiss
plots with top-k genes, which were selected using the evaluated statistic. PALM MicroBeam, Carl Zeiss Microscopy GmbH, Jena, Germany) and no-r
Parameter k was varied from 25 to few hundreds. Shrink-T showed the best mal DNA control samples from the tails of the same mice with QIAamp
separation in the generated plots across all values of k. Each methods’-abil DNA micro Kit, and DNeasy Blood & Tissue Kit (Qiagen), respectively. The
ity to find correlations with Gene Ontology classes was also tested. We used genomic DNA was amplified with 6-FAM labeled primers in 11.×1 PCR
t-test scores from each method separately as an input to enrichment analysis master mix (28) using the following PCR protocol: 1 min at 96°C, 30 cycles
tool called GSZ (Gene Set Z-score)25(). Shrink-T again generated strongest of 20 s at 96°C, 30 s at 62°C and 15 s at 70°C, and 7 min at 70°C. The fra-g
results. We therefore applied only Shrink-T in the subsequent analys1e3s)(. ments were analysed with ABI3730xl capillary electrophoresis (Thermo
Fisher Scientific) and visualized with PeakScanner v1.0 (Thermo Fisher
Visualization of sample differences Scientific) (13).
To detect similarities and differences between the samples, we used The four colon carcinomas found in the heterozygotMelh1+/− mice, of
MDS that generates a small-dimensional visualisation from the which the cryo sample was available, were also studied for lossMoflh1.
Immunohistochemical analysis of Mlh1 protein Differential expression analysis (DEA) used modifiedt-tests (Limma,
cyber-T and shrinkage-T). With Limma and cyber-T, we used their own
expression in carcinomas P-value estimates. Shrinkage-T does not provide Pa-value estimate, which
Formalin-fixed, paraffin embedded cancer tissue blocks were studied for were estimated by re-calculating Shrinkage-T with 1000 permutations for
Mlh1 expression. The 4 µm thick sections were deparaffinized and reh-y each gene separately. Normal distribution was fitted to the permutations
drated and heat induced antigen retrieval was performed with 10 mM-cit and a one-tailedP-value was obtained from the cumulative distribution.
rate buffer (pH 6). To detect Mlh1, the slides were incubated overnight at Multiple testing correction was performed using false discovery rate.
4°C with the rabbit monoclonal antibody ab92312 (1:1500) (clone EPR3894, Importantly, we used DEA mainly to order the genes to most differentially
Abcam, Cambridge, UK). Stainings were visualized using UltraVision regulated genes. All analysis was performed within the R-environment.
Detection System anti-rabbit HRP/DAB (ThermoFisher Scientific, Waltham, Pathway enrichment analysis was done using IPA which uses Fisher’s
MA) by manufacturer´s instructions. Analysis of staining patterns was exact test to analyse over-representation of genes from the analysed gene
conducted at The Finnish Centre for Laboratory Animal Patholog13y)(. groups. Here, multiple testing correction was done using the Benjamin–
Hochberg method (13).
Analysis of mitoses in carcinomas
A Feulgen with Midori green background stain was used to visualise Results
nuclear material and mitoses in six carcinoma samples (E249, E314, E329,
E333, E338 and E347). The samples were deparaffinized and rinsed in 1 Carcinomas developed mainly and earlier in WD
M HCl. Mild acid hydrolysis was accomplished by using 60°C 1 M HCl fed mice
and DNA was stained purple in Schiff´s reagent for 45 min. After several
bisulfite washes the samples were counterstained briefly with 1% Midori The feeding study was done with offspring produced by crossing
light green, dehydrated through alcohol series to xylene and mounted two isogenic strains, the heterozygotMe lh1+/− (B6.129-Mlh1tm1Rak)
and the wild-typeMlh1+/+ (C57BL/6) mice, and selecting an equal the feeding experiment (at time point 0), thMelh1 heterozygote
number of both genotypes to the study. Half of the mice fed WD mice showed exactly 50% lowerMlh1 expression than theMlh1+/+
and half the control diet, AIN-93G. In all 168 mice, 24 mice at mice. Contrary to varyingMlh1 expression levels in mice in
gentime point 0 and 48 mice at time points 12, 18 and 21 months, eral, 5/6 mice (E249, E314, E329, E333, E338) who developed ca-r
were operated. Approximately 80% of all colon tumours, 10 out cinoma showed remarkably lowMlh1 expression in their normal
of 13 colon adenocarcinomas and 14 out of 20 adenomas and colon mucosa (P = 0.03) (Figure 2) (13). The mouse E347 whose
hyperplasias, developed in WD-fed mice T(able 1) (13). carcinoma had serrated histology had, however, higheMr lh1
At time points 12, 18 and 21 months, 80, 78 and 64% of all expression level as non-carcinoma mice on averageT(able 1;
tumours and 100, 80 and 72% of CRCs were found in WD-fed Figure 2).
mice, respectively. The overall number of colon tumours
increased significantly with timeSu(pplementary Figure )1, Expression profiles in normal mucosa formed a
being 5 at 12 months (one adenocarcinoma, two adenomas, one distinct cluster for CRC mice
hyperplasia), 9 at 18 months (five adenocarcinomas, two a-de After finding that carcinoma mice had extremely low levels
nomas, two hyperplasias) and 19 at the 21 months time point of Mlh1 transcripts in their mucosa, we next compared their
(seven adenocarcinomas, five adenomas, seven hyperplasias). genome-wide expression profiles with profiles of all other 12
Tumours were approximately evenly distributed between d-if and 18 months old mice. The normal mucosa expression pr-o
ferent genotypes since heterozygoteMlh1+/− mice showed 0, 40 files of the six CRC mice were strikingly different from the p-ro
and 43% of carcinomas and 50, 75 and 42% of adenomas and files of the other mice and formed a distinct cluster as visualized
hyperplasias at different time pointsTa( ble 1). All the 13 carc-i by an MDS plot created with the 100 most altered/differentially
nomas were found in the proximal part of colon and the ma-jor regulated genes F(igure 3, Supplementary Table ,3available at
ity of them were either tubular (54%) or mucinous (31%), two Carcinogenesis Online). Altogether 86% of the top 300 differe-n
were tubulovillous and one carcinoma had serrated histology tially regulated genes in CRC mice were down-regulated and
(Supplementary Table ,2 Supplementary Figure ,2 available at 14% were up-regulated S(upplementary Table ,3 available at
Carcinogenesis Online). Carcinogenesis Online) (13).
Mlh1 mutation carriers did not show MSI, LOH and Pathway analysis and shortage of chromosomal loss of MMR protein in tumours segregation gene-specific transcripts suggest
To check for the typical LS characteristics, seven carcinomas problems in cell cycle regulation and mitosis
found in 18 and 21 months old mice, four in theMlh1+/− mice To further understand the biological functions and pathways
(E338, E347, E437, E444) and three in theMlh1+/+ mice (E402, E410, enriched among the top separating genes in CRC mice, the
E421), were analysed for MSI status and all the 13 carcinomas for expression data were analysed with Ingenuity Pathway Analysis.
Mlh1 expression. Surprisingly, all CRCs showed Mlh1 expression According to Ingenuity Pathway Analysis, chromosome segreg-a
(Figure 1A), indicating that irrespective of the inherited mu-ta tion (P = 2.92 × 10–5), aneuploidy of fibroblastsP( = 5.31 × 10–4),
tion in one Mlh1 allele in the heterozygote mice, the normal checkpoint controlP( = 1.10 × 10–4), DNA replication checkpoint
allele was still present in the tumours. The presence of the -nor (P = 1.88 × 10–4) and morphology of mitotic spindleP (= 6.45 × 10–5)
mal allele was further confirmed with LOH study in all the four were among the most affected biological functions. In network
Mlh1+/− carcinomas (Figure 1B). To study whether the detected analysis, the most affected molecular and cellular functions
Mlh1 protein was functional and MMR proficient, we analysed included cell cycleP( = 9.24 × 10–5), cellular assembly and orga-n
the stability of six polymorphic microsatellite regions in the ization (P = 9.24 × 10–5), DNA replication, recombination and
mouse genome. The markers and their amplified fragment sizes repair (P = 9.24 × 10–5), cell death and survivalP (= 7.30 × 10–5) and
were as follows: D14Mit15 (148 bp, 150 bp), D18Mit15 (151 bp, cellular growth and proliferatioPn =( 3.07 × 10–3) (Supplementary
157 bp), D7Mit91 (139 bp, 147 bp), D10Mit2 (117 bp, 122 bp), JH104 Table 4, available atCarcinogenesis Online). The analysis was
(178 bp, 181 bp) and U12235 (79 bp, 83 bp). Altogether six out also repeated with different RNA-seq data preprocessing (all
of seven CRCs were studied (E410 could not be amplified) and mouse samples without ComBat normalizatioSnu,pplementary
shown to be microsatellite stable, since no differences in the Table 4, available atCarcinogenesis Online). These results
confragment lengths were observed between the tumour and c-or firmed our findings on chromosome segregationP (= 1.03 × 10–5),
responding normal DNA S(upplementary Figure ,3available at aneuploidy of fibroblastsP( = 4.57 × 10–4) and checkpoint control
Carcinogenesis Online) (13). (P = 4.29 × 10–4) (13).
The Ingenuity Pathway Analysis results strongly indicated
Mlh1 RNA expression was significantly decreased in that there are severe problems in cell cycle regulation and mitosis
normal mucosa of CRC mice already in colon mucosa. In the six mice who developed carc-in
After finding that irrespective of the mouse genotype, fu-nc oma up to 18 months, the most altered/differentially expressed
tional Mlh1 was still expressed in carcinomas, our interest was genes that pointed to chromosome segregation and SAC were
to look for potential early drivers of tumorigenesis on a genome- Bub1 (BUB1, mitotic checkpoint serine/threonine kinaseM), is18a
wide scale. Altogether 71 out of 80 normal colon mucosa sa-m (MIS18 kinetochore protein A),Tpx2 (TPX2 microtubule assoc-i
ples operated from 12- and 18-month old mice were qualified ated),Rad9a (RAD9 checkpoint clamp component A),Pms2
(postfor genome-wide transcriptome analysis. Analysis was done meiotic segregation increased 2)M,lh1 (mutL homolog 1, along
with RNAseq using the single-cell tagged reverse transc-rip with MMR function also triggers checkpoint activatioCne),npe
tion method (STRT) (
). The 21-month old mice were left (centromere protein E)N,capd3 (non-SMC condensing II complex
out from the RNAseq study due to many health problems most subunit D3),Odf2 (outer dens fiber of sperm tails 2) andDclre1b
probably because of their old age. Altogether 12 216 expressed (DNA cross-link repair 1B). Five of these 10 genes,Bub1, Mis18a,
transcripts were identified in the samples. First, we analyzed Tpx2, Rad9a and Pms2, were strongly down regulated in all of
the Mlh1 gene expression levels from the STRT dataF(igure 2). the six carcinoma mice F(igure 4). Mlh1, Cenpe, Ncapd3, Odf2 and
In our previous study 1(9), we showed that in the beginning of Dclre1b showed variable level of expression in two CRC mice
(E347 and E249). In E249, Mlh1 and Dclr1b showed approximately mice, E325 and E332, showed similar low expression of all the
50% down regulation when compared to the average expression 10 genes (Figure 4). Although, no colonic tumours were found
level in non-carcinoma mice F(igure 4). In E347, whose carcin- in those mice, E325 had bloody feces and anemia, suggesting
oma was histologically different from the others and showed undefined mucosal pathology1(3).
typical serrated phenotypeC, enpe, Ncapd3 and Odf2 expression
levels were equal andMlh1 expression level higher than in the Abnormal mitoses and CIN in carcinomas
non-carcinoma mice. Importantly, among all the 74 mice that Undisturbed mitosis is a central requirement of the normal cell
did not develop colon carcinoma up to 18 months, only two cycle and division. In cancer cells, mitoses are often aberrant,
showing aneuploidy caused by unequal segregation of chrom-o Discussion
somes and/or structural changes in chromosomes, both of which
lead to CIN. To validate the RNA sequencing results, which s-ug The instability status and other characteristics of colon tumours
gested impaired cell cycle regulation and mitosis in CRC mice, all have been used to define the mechanisms behind colon ca
nthe 13 carcinomas were stained with feulgen and analyzed for cer formation 3( ). Our present genome-wide expression profi-l
mitotic aberrations. Although all the carcinomas were wel-l-dif ing experiment demonstrates that cancer preceding changes
ferentiated early cancers with limited submucosal invasion and occur and can be detected already in normal colon mucosa. We
relatively lenient cytological changes, they exhibited increased show that these changes may form a field-defect in histolo-gi
mitotic activity and abundant numbers of unbalanced/atypical cally normal mucosa and trigger MMR-proficient, chromos-om
mitoses in contrast to normal tissue sampleFsig(ure 5) (13). ally unstable CRC both inMlh1+/+ and Mlh1+/− mice.
One common event linked to both sporadic and inherited with causal implication of Bub1 deficiency drives tumorig-en
MSI cancers is the inactivation ofMLH1, although the inact-i esis through tumour-suppressor gene LOH 3(7), perfectly in
vation mechanism varies from genetic mutations in LS to e-pi line with our findings that the majority of the carcinoma mice
genetic silencing in sporadic CRC7(
). Our mouse study was distinguishing genes were tumour suppressor genes, which
designed to comprehensively clarify the role oMflh1 expression were down regulated S(upplementary Table ,3 available at
during colon tumorigenesis and the effects of genetic pred-is Carcinogenesis Online). Cenpe, a kinesin-like motor protein which
position and other risk factors such as WD and aging on it. Mlh1 is an efficient stabilizer of microtubule capture at kinetochores
protein expression was studied in colon tumours anMdlh1 gene and hence essential for metaphase chromosome alignment
expression in histologically normal mucosa. Approximately (38), was strongly down regulated in the mice with carcinoma.
70% of all tumours and 80% of colon carcinomas developed in While it plays an important role in the movement of chro-mo
WD-fed mice indicating a strong diet effect on cancer pred-is somes toward the metaphase plate during mitosis, it is also
position. 33 % of CRCs and 75% of adenomas and hyperplasias necessary for the mitotic checkpoint signal at the kinetochore
found in mice up to 18 months of age was found inMlh1+/− mice. to prevent chromosome loss3(9). Dclre1b has a central role in
Surprisingly, Mlh1 protein was present and there was no MSI telomere maintenance and protection during S-phase through
in their cancers. Genome-wide expression profiling of histolo-gi its 5′–3′ exonuclease activity. Moreover, in case of spindle stress,
cally normal mucosa however showed that 5/6 mice who dev-el Dclre1b like Bub1 is involved in prophase checkpoint40(,41).
oped CRC up to 18 months had significantly decreased mucosal RAD9A, a component of the 9-1-1 cell cycle checkpoint response
Mlh1 RNA expression. Only in the carcinoma mouse E347 the complex, plays a major role in DNA repair and participates in
Mlh1 expression level was higher than the average level of 65 multiple cell cycle checkpoints and apoptosis and its aberrant
mice without cancer. expression has been linked to tumorigenesis of multiple tissues
Low Mlh1 expression, although a prominent signal, seemed (42). Interestingly, Rad9 also physically interacts with the MMR
not to be an absolute requirement or sufficient alone to cause protein MLH1 (43). The MMR mechanism is so essential for no-r
colon cancer since several mice without CRC had lowMlh1 mal cell function that it may explain why even a small amount
expression as well. In order to identify other genes and pa-th of MLH1 appears to be sufficient for MMR function, whereas its
ways involved in CRC development, we compared the genome checkpoint activation role seems to require a full complement
wide expression profiles in the six CRC mice with the profiles of of the protein 4(4). It has been argued that theMLH1
heterozy65 mice without CRC 1(3). Remarkably, the expression profiles of gosity/haploinsufficiency may drive the development of cancer
CRC mice formed a clearly distinct clusteFrig(ure 3), indicating by accumulation of insertion/deletion mutations in other g-ate
a field-defect in normal colon mucosa31( ,32). By network ana- keeper genes prior to MSI 4(5). Indeed, cells with diminished
lysis of top 100 CRC mice separating genesM,lh1 expression in amount of MLH1 protein may still be MMR proficient, although
normal mucosa from CRC mice was found to be low together they show defects in DNA damage signalling4(4). Consequently,
with significant down regulation of several cancer related genes the damaged cells may not activate cell cycle checkpoints and
and pathways (Supplementary Table 3and Tables 4, available at enter apoptosis. Our observation that low mRNA expression of
Carcinogenesis Online) and especially of chromosomal segregation Mlh1 in carcinoma mice together with down regulation of s-ev
genes, Bub1, Mis18a, Tpx2, Rad9a, Pms2, Cenpe, Ncapd3, Odf2 and eral other genes related to chromosome segregation and ch-eck
Dclre1b. Only two (E325 and E332) of the 65 mice without cancer point control supports the proposition that already decreased
shared the expression profile of the CRC mice related to chro m-o amount of Mlh1, when MMR is still functional, may have an
somal segregationF(igure 4). Although no colonic tumours were important role in tumorigenesis13().
found in those mice, carcinogenesis might been happening in Low expression of Tpx2, Mis18a, Ncapd3 and Odf2 reflects
their mucosa. For example, E325 had bloody feces and anemia problems in formation of the nuclear spindle and chro m-o
suggesting pathological problems in mucosa. Differing from some segregation.Tpx2, Ncpd3 and Odf2, a general scaffolding
the other CRC mice, E347, which did not show decrease inMlh1 protein (46), are all involved in microtubules related p-ro
expression, showed decrease only in the expressions ofBub1, cesses in spindle formation.Tpx2 plays a role in microtubule
Mis18a, Tpx2, Rad9a, Pms2 and Dclre1b, suggesting their remar-k organization and is involved in centrosome maturation
able importance in serrated carcinogenesis. Furthermore, in the (47). In fact, TPX2-depleted cells fail to form proper mitotic
mouse E249 the Mlh1 and Dclre1b genes showed approximately spindles (48). Recent findings suggest that TPX2 also plays
50% lower expression than was detected in the non-carcinoma an important role in promoting colon tumorigenes4is9)(. In
mice on average. Here, the milder decrease may reflect young this study results support a driver role fToprx2, since it was
age of onset. strongly down regulated in colon mucosa in all carcinoma
During cell division the SAC, which is the major target of mice. Ncapd3 functions in the condensin II complex and is
mitotic alterations, maintains the genome stability by de-lay needed to establish the chromosomal architecture necessary
ing cell division until all chromosomes are accurately aligned for proper spindle assembly and chromosome segregation.
in the spindle (33). Aberrant expression of mitotic genes leads Chromosome condensation and resolution are compr-o
to mitotic abnormalities including centrosome defects and mised when condensin is depleted (50). The MIS18 complex
improper spindle checkpoint leading to CIN34() and tumour accumulates at the centromere during anaphase to early G1
formation in multiple tissues including colo3n5)(. In this study, phase, slightly ahead of the histone H3 variant CENPA and
the mRNA expression was significantly decreased in five SAC is an absolute requirement for the localization of CENPA at
associated genes,Mlh1, Bub1, Rad9a, Dclre1b and Cenpe. Of those, centromeres. Importantly, Mis18a knockout causes severe
Bub1 is a major player and activator in SAC and its haploins-uf chromosomal missegregation, lack of CENPA and ultimately
ficiency (heterozygosity) is known to be responsible for chrom- o cell death 5(
). Here, along withMis18a, Cenpa was
signifisome segregation defects and aneuploidy3(5). During mitosis, cantly down regulated in the normal colon mucosa of the
Bub1 is required for the recruitment of other checkpoint and CRC mice (Supplementary Table ,3available atCarcinogenesis
motor proteins, such as Cenpe, to a kinetochore36(). There is Online) supporting the finding of improper chromosome
evidence suggesting that inaccurate chromosome segregation segregation.
Our RNA sequencing results suggest that cancer preceding 13 . Pussila , M. ( 2017 ) Cancer-Preceding Gene Expression Changes in Mouse changes occur already in histologically normal mucosa and Colon Mucosa . University of Helsinki, Helsinki. trigger MMR-proficient chromosomally unstable CRC . Problems 14. Newmark , H.L. et al. ( 2009 ) Western-style diet-induced colonic tumors in cell cycle regulation and mitosis were confirmed by tumour and their modulation by calcium and vitamin D in C57Bl/6 mice: a p-re analysis, which showed increased mitotic activity and abu-n clinical model for human sporadic colon cancer . Carcinogenesis , 30 ,
88- 92 . dant numbers of atypical mitoses in carcinomas . Altogether 71 15. Newmark, H.L. et al. ( 2001 ) A Western-style diet induces benign mice were included in genome-wide expression profiling and and malignant neoplasms in the colon of normal C57Bl/6 mice. only two of the 65 mice without cancer shared the expression Carcinogenesis , 22 , 1871 - 1875 . profile of the CRC mice related to chromosomal segregation, 16 . Yang , K. et al. ( 2008 ) Dietary induction of colonic tumors in a mouse indicating that the aberrant expression of this gene set could be model of sporadic colon cancer . Cancer Res. , 68 , 7803 - 7810 . used as signalling carcinogenesis in colon mucosa . In addition, 17 . Yang , K. et al. ( 2005 ) Dietary components modify gene expression: the finding of unexpected tumour phenotypes inMlh1+/− mice implications for carcinogenesis . J. Nutr ., 135 , 2710 - 2714 . implies that stable microsatellites and normal MLH1 expression 18 . Edelmann , W. et al. ( 1996 ) Meiotic pachytene arrest in MLH1-deficient in human tumours may not necessarily exclude inheriteMdLH1 mice . Cell , 85 , 1125 - 1134 . deficiency. 19 . cPoulsosinliac, mMu.ecto asla. (o2f0W13e)sCtearnncedri-eptrfeeddicMtlinhg1+g/e-nmeiceex .pPrLeosSsioOnnec, h8a , en7g6e8s6i5n .
20. Reeves , P.G. et al. ( 1993 ) AIN-93 purified diets for laboratory rodents: Supplementary material final report of the American Institute of Nutrition ad hoc writing -com
mittee on the reformulation of the AIN-76A rodent diet . J. Nutr ., 123 , Supplementary data can be found aCtarcinogenesis online . 1939- 1951 .
21. Washington, M.K .et al. ( 2013 ) Pathology of rodent models of intestinal Funding cancer: progress report and recommendations . Gastroenterology , 144 ,
705- 717 . This study was supported by grants from Jane and Aatos Erkko 22 . Islam , S. et al. ( 2012 ) Highly multiplexed and strand-specific single-cell Foundation (M.N .), and Helsinki Institute of Life Science (P.T.). RNA 5' end sequencing . Nat. Protoc ., 7 , 813 - 828 .
23. Krjutškov , K. et al. ( 2016 ) Single-cell transcriptome analysis of en-do
metrial tissue . Hum . Reprod., 31 , 844 - 853 . Acknowledgements 24. Anders , S. et al. ( 2010 ) Differential expression analysis for sequence The authors wish to thank Laura Sarantaus for help in planning count data . Genome Biol ., 11 , R106. the feeding study and sample collection , Satu Valo and Denis 25 . Mishra , P. et al. ( 2014 ) Gene set analysis: limitations in popular existing Dermadi-Bebek for help in sample collection, Marja Mutanen for methods and proposed improvements . Bioinformatics , 30 , 2747 - 2756 .
26. Ritchie , M.E. et al. ( 2015 ) Limma powers differential expression analyses help with the diet planning, Mikael Niku for help with immu-no for RNA-sequencing and microarray studies . Nucleic Acids Res ., 43 , e47. histochemistry, Liisa Kauppi for help in planning the MSI ma-rk 27 . Edelmann , W. et al. ( 1999 ) Tumorigenesis in Mlh1 and Mlh1/Apc1638N ers and Ingegerd Fransson for help in the sequencing library mutant mice . Cancer Res. , 59 , 1301 - 1307 . preparation. 28. Kauppi , L. et al. ( 2009 ) Analysis of meiotic recombination products from Conflict of Interest Statement: None declared. human sperm . Methods Mol. Biol ., 557 , 323 - 355 .
29. Islam , S. et al. ( 2011 ) Characterization of the single-cell transcriptional References landscape by highly multiplex RNA-seq . Genome Res. , 21 , 1160 - 1167 .
30. Ollikainen , M. et al. ( 2007 ) Mechanisms of inactivation of MLH1 in 1 . Bingham, S. et al. ( 2004 ) Diet and cancer-the European prospective hereditary nonpolyposis colorectal carcinoma: a novel approach .
investigation into cancer and nutrition . Nat. Rev. Cancer , 4 , 206 - 215 . Oncogene, 26 , 4541 - 4549 . 2. Fearon , E.R. ( 2011 ) Molecular genetics of colorectal cancer . Annu. Rev. 31 . Park , S.K. et al. ( 2016 ) Field cancerization in sporadic colon cancer . Gut
Pathol., 6 , 479 - 507 . Liver, 10 , 773 - 780 . 3. Abbas , T. et al. ( 2013 ) Genomic instability in cancer . Cold Spring Harb . 32 . Hawthorn , L. et al. ( 2014 ) Evidence for field effect cancerization in col-o
Perspect. Biol. , 5 , a012914. rectal cancer. Genomics , 103 , 211 - 221 . 4. Al-Sohaily , S. et al. ( 2012 ) Molecular pathways in colorectal cancer . J . 33 . Lara-Gonzalez , P. et al. ( 2012 ) The spindle assembly checkpoint . Curr.
Gastroenterol. Hepatol. , 27 , 1423 - 1431 . Biol., 22 , R966 - R980 . 5. Pino , M.S. et al. ( 2010 ) The chromosomal instability pathway in colon 34 . Perez de Castro, I.et al. ( 2007 ) A census of mitotic cancer genes: new
cancer. Gastroenterology , 138 , 2059 - 2072 . insights into tumor cell biology and cancer therapy . Carcinogenesis , 28 , 6 . Colussi , D. et al. ( 2013 ) Molecular pathways involved in colorectal c-an 899-912 .
cer: implications for disease behavior and prevention . Int. J. Mol. Sci ., 35 . de Voer, R.M. et al. ( 2013 ) Germline mutations in the spindle assembly
14 , 16365 - 16385 . checkpoint genes BUB1 and BUB3 are risk factors for colorectal cancer. 7 . Boland , C.R. et al. ( 2010 ) Microsatellite instability in colorectal cancer . Gastroenterology , 145 , 544 - 547 .
Gastroenterology , 138 , 2073 - 2087 . e3 . 36 . Johnson , V.L. et al. ( 2004 ) Bub1 is required for kinetochore localization 8 . Parsons , M.T. et al. ( 2012 ) Correlation of tumour BRAF mutations and of BubR1, Cenp-E, Cenp-F and Mad2, and chromosome congression . J.
MLH1 methylation with germline mismatch repair (MMR) gene mu-ta Cell Sci ., 117(Pt 8) , 1577 - 1589 .
tion status: a literature review assessing utility of tumour features for 37 . Sotillo , R. et al. ( 2009 ) Very CIN-ful: whole chromosome instability p-ro
MMR variant classification . J. Med. Genet ., 49 , 151 - 157 . motes tumor suppressor loss of heterozygosity . Cancer Cell , 16 , 451 - 452 . 9. Ahuja , N. et al. ( 2000 ) Aging, methylation and cancer . Histol . 38 . Putkey , F.R. et al. ( 2002 ) Unstable kinetochore-microtubule capture and
Histopathol., 15 , 835 - 842 . chromosomal instability following deletion of CENP-E. Dev . Cell, 3 , 10 . Thompson , B.A. et al. ( 2014 ) Application of a 5-tiered scheme for stan-d 351-365 .
ardized classification of 2,360 unique mismatch repair gene variants in 39 . Weaver , B.A. et al. ( 2003 ) Centromere-associated protein-E is essential
the InSiGHT locus-specific database . Nat. Genet ., 46 , 107 - 115 . for the mammalian mitotic checkpoint to prevent aneuploidy due to 11 . Valo , S. et al. ( 2015 ) DNA hypermethylation appears early and shows single chromosome loss . J. Cell Biol ., 162 , 551 - 563 .
increased frequency with dysplasia in Lynch syndrome-associated 40 . Liu , L. et al. ( 2009 ) SNM1B/Apollo interacts with astrin and is required
colorectal adenomas and carcinomas . Clin. Epigenetics , 7 , 71. for the prophase cell cycle checkpoint . Cell Cycle , 8 , 628 - 638 . 12 . Goel , A. et al. ( 2003 ) Characterization of sporadic colon cancer by p-at 41 . Yan , Y. et al. ( 2010 ) The multifunctional SNM1 gene family: not just
terns of genomic instability . Cancer Res. , 63 , 1608 - 1614 . nucleases. Future Oncol., 6 , 1015 - 1029 . 42 . Lieberman , H.B. et al. ( 2011 ) The role of RAD9 in tumorigenesis . J. Mol. 47 . Karsenti , E. et al. ( 2001 ) The mitotic spindle: a self-made machine .
Cell Biol ., 3 , 39 - 43 . Science, 294 , 543 - 547 . 43 . He , W. et al. ( 2008 ) Rad9 plays an important role in DNA mismatch repair 48 . De Luca, M. et al. ( 2006 ) A functional interplay between Aurora-A, Plk1
through physical interaction with MLH1 . Nucleic Acids Res ., 36 , 6406 - 6417 . and TPX2 at spindle poles: Plk1 controls centrosomal localization of 44 . Cejka , P. et al. ( 2003 ) Methylation-induced G(2)/M arrest requires a full co-m Aurora-A and TPX2 spindle association . Cell Cycle , 5 , 296 - 303 .
plement of the mismatch repair protein hMLH1 . EMBO J ., 22 , 2245 - 2254 . 49 . Wei , P. et al. ( 2013 ) TPX2 is a novel prognostic marker for the growth and 45 . Wang , L. et al. ( 2012 ) Whole-exome sequencing of human pancreatic metastasis of colon cancer . J. Transl. Med ., 11 , 313 .
cancers and characterization of genomic instability caused by MLH1 50 . Wignall , S.M. et al. ( 2003 ) The condensin complex is required for
haploinsufficiency and complete deficiency . Genome Res. , 22 , 208 - 219 . proper spindle assembly and chromosome segregation in Xenopus egg 46 . Ishikawa , H. et al. ( 2005 ) Odf2-deficient mother centrioles lack distal/ extracts . J. Cell Biol ., 161 , 1041 - 1051 .
subdistal appendages and the ability to generate primary cilia . Nat. 51 . Baumann , K. ( 2012 ) Keeping centromeric identity . Nat. Rev. Mol. Cell
Cell Biol ., 7 , 517 - 524 . Biol., 13 , 340 .