Mutations in NOD2 have been shown to associate with increased susceptibility to Crohn’s disease. A recent Polish study linked the truncating NOD2 3020insC variant with an increased risk of colorectal cancer (CRC) at an older age (>50 years) of disease onset, with an odds ratio of 2.23. We studied the possible contribution of the 3020insC variant to CRC risk in a series of 1,042 Finnish population-based patients from which 926 samples were successfully analyzed and in 348 anonymous cancer-free controls. The frequency of the 3020insC mutation was 3.7% in both CRC patients (34 of 926, including 1 homozygote) and cancer-free controls (13 of 348; odds ratio, 0.98; confidence interval, 0.51–1.88). Contrary to the Polish study, there was no significant difference in the mutation rates between CRC patients > 50 years of age (25 of 576; 4.3%) and controls in the present series. We studied respective tumor tissue DNAs of all patients displaying heterozygous 3020insC changes for loss of heterozygosity. Loss of heterozygosity at NOD2 was observed in only 1 of the 33 CRC samples. Our results suggest that NOD2 3020insC alone does not contribute to CRC risk. If this variant predisposes to CRC, additional factors not present in the Finnish population need to be involved.

Genetic variation of NOD2 has been shown to associate with increased susceptibility to Crohn’s disease in several independent studies, although in some populations no association has been observed (1, 2, 3, 4, 5, 6, 7, 8, 9). Three common mutations, two missense changes (Arg702Trp and Gly908Arg) and one frameshift change (3020insC→1007fs), located in the leucine-rich repeat (LRR) domain of the NOD2 protein have been reported to be linked with the disease (1, 4). Association of the 3020insC mutation with Crohn’s disease has also been detected in the Finnish population (8). The NOD2 protein shows structural homology with both the apoptosis regulators and plant disease resistance gene products (10). However, the exact molecular function of NOD2 is not thoroughly established. The protein consists of 2 NH2-terminal caspase recruitment domains, 1 nucleotide-binding domain, and 10 COOH-terminal LRRs; the caspase recruitment domains are involved in activating nuclear factor-κB pathways, and the LRRs are involved in the recognition of bacterial components (10, 11, 12). The NOD2 3020insC variant, resulting in a Leu1007→Pro substitution in the 10th LRR followed by a premature stop codon, has been shown to be less active in response to bacterial lipopolysaccharides, which might generate an increased inflammatory response by the adaptive immune system in patients with Crohn’s disease (1).

A recent study suggested an association of the NOD2 3020insC variant with colorectal cancer (CRC) predisposition in a consecutive series of Polish CRC patients (13). The mutation frequency was, however, significantly higher only in a group of 250 patients >50 years of age at diagnosis, whereas in CRC patients ≤50 years of age, no difference was discerned when compared with healthy controls. Here we studied the frequency of the NOD2 3020insC mutation in a population-based series of 1,042 Finnish CRC patients and in 348 anonymous cancer-free blood donors. We also investigated possible loss of the wild-type NOD2 allele in the respective cancer samples of all heterozygous cases.

Patients.

A population-based series of 1,042 Finnish CRC patients was analyzed for NOD2 3020insC. Both normal and tumor tissue samples were collected prospectively at nine Finnish central hospitals between 1994 and 1998. This extensively documented material is described in more detail in previous works (14, 15). DNAs from 348 anonymous cancer-free Finnish blood donors were obtained from the Finnish Red Cross Blood Transfusion Service. Complete clinical data, including patient’s age at disease onset, family history, pathology reports, grade, and molecular data, including microsatellite instability status, and MLH1 and MSH2 mutation screening results of all microsatellite instability cases were available from all CRC patients.

Mutation and Loss of Heterozygosity Analysis.

Detection of the 3020insC mutation was performed by amplifying normal tissue DNA by allele-specific polymerase chain reaction (PCR) as described by Ogura et al.(1). A positive control was used in each PCR. The results were confirmed by amplifying the samples again with the primers framing a 533-bp region around the insertion spot (1) and by sequencing the product using Applied Biosystems (Foster City, CA) 3730 BD3.1 sequencing chemistry and 5.1 sequencing analysis software (Applied Biosystems). Respective histologically verified tumor tissue DNA of all 33 samples heterozygous for the 3020inC variant was sequenced using the aforementioned primers to assess possible loss of heterozygosity (LOH). NOD2 mutation 3020insC was used as a polymorphism to determine possible LOH (Fig. 1).

Statistical Methods.

χ2 test with Yates correction was used in analyzing the results. P < 0.05 was considered significant.

A total of 1,042 CRC samples were analyzed for the NOD2 3020insC variant, and results were successfully derived from 926 samples. The mutation frequencies were identical in CRC patients (34 of 926; 3.7%) and population controls (13 of 348; 3.7%). One CRC patient was homozygous; the rest of the NOD2 mutation-positive cases were heterozygotes for the 3020insC variant.

To assess the possible contribution of loss of the wild-type NOD2 allele to 3020insC-related tumorigenesis, we sequenced the respective tumor tissue DNAs of all of the patient samples displaying a heterozygous NOD2 mutation. LOH was detected in only 1 of the 33 samples; thus, evidence to support involvement of NOD2 in the genesis of these cancers was not obtained through this effort (Fig. 1).

To analyze the impact of CRC family history on mutation frequency, the 926 CRC cases were divided into two groups: patients with no first-degree relatives with CRC were classified as nonfamilial cases (n = 798); whereas patients with one to five first-degree relatives with CRC were classified as familial cases (n = 128). The occurrence of the mutation was also analyzed according to microsatellite instability status (Table 1). Finally, the CRC patients were divided into three subgroups similar to those described by Kurzawski et al.(13). At this point, 25 mutation-positive hereditary nonpolyposis CRC patients were excluded from the analysis, resulting in a total of 901 patients. In group 1 (n = 576), all patients were >50 years of age at the time of diagnosis and had, at the most, one first-degree relative with cancer. Group 2 (n = 68) was identical to group 1, except that all patients were ≤50 years old at diagnosis of the disease. Group 3 (n = 256) represented CRC patients with familial cancer aggregation; i.e., patients from families in which at least two other family members were diagnosed with CRC or other malignancies (ref. 13; Table 2).

There was no difference between the mutation rates in the case and the control groups [P = 0.91; odds ratio (OR), 0.98; confidence interval (CI), 0.51–1.88]. Allele frequency in the control individuals (1.9%) coincided with those reported previously (2, 3, 4, 8, 16). The frequency of the variant was similar in familial (2.3%) and nonfamilial (3.9%) groups (P = 0.77; OR, 0.98; CI, 0.29–3.35) and when each individual group was compared with the controls (P = 0.64 and P = 0.97, respectively; values not corrected for multiple testing). There was no significant difference in mutation rates between microsatellite stable (3.7%) and instable (3.6%) CRCs (P = 0.82; Table 1). Significant differences were not observed when groups 1 (4.3%), 2 (4.4%), and 3 (2.3%) were compared with healthy controls (P = 0.78, P = 0.78, and P = 0.46, respectively) or when group 1 was compared with group 2 (P = 0.77; Table 2).

Similar frequencies of the NOD2 3020insC variant in a population-based series of individuals with CRC and in healthy controls indicate that either the mutation does not predispose to CRC or the risk of CRC conferred by the mutation is, at the most, modest. The Polish study observed an OR of 2.23 in CRC patients older than 50 years (13). Despite sufficient power, our study was also negative for this patient group. The frequency of the 3020insC mutation was reported to be somewhat higher in Polish CRC patients, who match the criteria of hereditary nonpolyposis CRC (the exact criteria were not specified) but display no mutations in the mismatch repair genes, although the difference did not reach statistical significance (13). However, in our series, there was no aggregation of the 3020insC mutation to CRC families. Whereas our results speak strongly against NOD2 involvement in CRC predisposition in Finns, we cannot exclude a role for NOD2 as a low penetrance CRC susceptibility gene in other populations, such as the Polish population. There might also be modifying factors other than the NOD2 3020insC variant conferring an elevated NOD2-related risk of CRC in the Polish population, but not in the Finnish population. These might include, for instance, effects of other as yet unknown predisposition alleles or environmental factors.

In summary, we have analyzed the incidence of the NOD2 3020insC variant in 926 CRC patients and 348 healthy controls. Mutation frequencies were identical in CRC patients and control individuals. No evidence of silencing of the second allele by LOH was obtained. Our results suggest that NOD2 has no major contribution to CRC risk in the Finnish population. However, it has been proposed that there is regional heterogeneity within Europe in the contribution of NOD2 variants to Crohn’s disease susceptibility, reflecting effects of differing founder populations (16, 17). The observed overall mutation frequency indeed appears to be higher in Poland than in many other Western populations (2, 3, 4, 8, 16). Also, environmental factors and additional genetic factors vary greatly between populations. Thus, additional data from different populations are needed to determine whether the NOD2 3020insC mutation predisposes to CRC and, if so, to identify the additional determinants necessary for the increased susceptibility.

Fig. 1.

Mutation screening was performed by allele-specific PCR. All samples with mutations were subsequently sequenced to confirm the results. Tumor DNA of all of the CRC patients harboring heterozygous 3020insC mutation was sequenced to assess the possible contribution of loss of wild-type NOD2 to tumorigenesis. LOH was observed in only 1 of 33 tumor samples. A. Normal tissue sequence trace shows the heterozygous 3020insC mutation. B. Cancer tissue DNA from the same patient shows some evidence of LOH. The wild-type allele versus mutated allele peak ratio was reduced approximately 45% in cancer tissue. The content of malignant tissue in the sample was 80%.

Fig. 1.

Mutation screening was performed by allele-specific PCR. All samples with mutations were subsequently sequenced to confirm the results. Tumor DNA of all of the CRC patients harboring heterozygous 3020insC mutation was sequenced to assess the possible contribution of loss of wild-type NOD2 to tumorigenesis. LOH was observed in only 1 of 33 tumor samples. A. Normal tissue sequence trace shows the heterozygous 3020insC mutation. B. Cancer tissue DNA from the same patient shows some evidence of LOH. The wild-type allele versus mutated allele peak ratio was reduced approximately 45% in cancer tissue. The content of malignant tissue in the sample was 80%.

Close modal

Grant support: Grants from the Academy of Finland (grant 44870, Finnish Center of Excellence Program 2000–2005), Finnish Cancer Society, Sigrid Juselius Foundation, and European Commission (QLG2-CT-2001-01861).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Lauri A. Aaltonen, Department of Medical Genetics, Biomedicum Helsinki, Haartmaninkatu 8, University of Helsinki, Helsinki 00014, Finland. Phone: 358-9-19125595; E-mail: lauri.aaltonen@helsinki.fi

Table 1

Frequency of the NOD2 3020insC mutation in controls and in CRC patients

Cohorts and SubgroupsNo. of subjectsNo. with mutation (%)Allele frequency (%)P              *OR (CI)*
Controls 348 13 (3.7) 1.9   
CRC patients 926 34 (3.7) 1.9 0.91 0.98 (0.51–1.88) 
 Nonfamilial CRC 798 31 (3.9) 2.0 0.97 1.04 (0.53–2.01) 
 Familial CRC 128 3 (2.3) 1.2 0.64 0.62 (0.17–2.21) 
 MSI CRC 111 4 (3.6) 2.3 0.82 0.96 (0.31–3.02) 
 MSS CRC 815 30 (3.7) 1.8 0.90 0.98 (0.51–1.91) 
Cohorts and SubgroupsNo. of subjectsNo. with mutation (%)Allele frequency (%)P              *OR (CI)*
Controls 348 13 (3.7) 1.9   
CRC patients 926 34 (3.7) 1.9 0.91 0.98 (0.51–1.88) 
 Nonfamilial CRC 798 31 (3.9) 2.0 0.97 1.04 (0.53–2.01) 
 Familial CRC 128 3 (2.3) 1.2 0.64 0.62 (0.17–2.21) 
 MSI CRC 111 4 (3.6) 2.3 0.82 0.96 (0.31–3.02) 
 MSS CRC 815 30 (3.7) 1.8 0.90 0.98 (0.51–1.91) 

Abbreviations: MSI, microsatellite instable; MSS, microsatellite stable.

*

χ2 with Yates correction factor; group versus controls.

Table 2

NOD2 3020insC mutation in subgroups according to Kurzawski et al.(13) 

CRC subgroupsNo. of subjectsNo. with mutation (%)Allele frequency (%)P              *OR (CI) by χ2 test
Group 1 577 25 (4.3) 2.3 0.78 1.17 (0.59–2.32) 
Group 2 68 3 (4.4) 2.2 0.78 0.78 (0.17–3.54) 
Group 3 256 6 (2.3) 1.2 0.46 0.62 (0.23–1.65) 
CRC subgroupsNo. of subjectsNo. with mutation (%)Allele frequency (%)P              *OR (CI) by χ2 test
Group 1 577 25 (4.3) 2.3 0.78 1.17 (0.59–2.32) 
Group 2 68 3 (4.4) 2.2 0.78 0.78 (0.17–3.54) 
Group 3 256 6 (2.3) 1.2 0.46 0.62 (0.23–1.65) 

NOTE. Groups are defined as follows: Group 1, CRC patients > 50 years of age with one other malignancy, at most, in the family; Group 2, CRC patients ≤ 50 years of age with one other malignancy, at most, in the family; and Group 3, CRC patients of all ages with familial cancer aggregation (i.e., two or more other malignancies in the family).

*

χ2 with Yates correction factor; group versus controls.

We thank Sini Marttinen, Tuula Lehtinen, and Kirsi Pylvänäinen for collecting the clinical data.

1
Ogura Y, Bonen DK, Inohara N, et al A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease.
Nature (Lond)
2001
;
411
:
603
-6.
2
Hugot JP, Chamaillard M, Zouali H, et al Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease.
Nature (Lond)
2001
;
411
:
599
-603.
3
Vermeire S, Wild G, Kocher K, et al CARD15 genetic variation in a Quebec population: prevalence, genotype-phenotype relationship, and haplotype structure.
Am J Hum Genet
2002
;
71
:
74
-83.
4
Lesage S, Zouali H, Cezard JP, et al CARD15/NOD2 mutational analysis and genotype-phenotype correlation in 612 patients with inflammatory bowel disease.
Am J Hum Genet
2002
;
70
:
845
-57.
5
Hampe J, Cuthbert A, Croucher PJ, et al Association between insertion mutation in NOD2 gene and Crohn’s disease in German and British populations.
Lancet
2001
;
357
:
1925
-8.
6
Guo QS, Xia B, Jiang Y, Qu Y, Li J NOD2 3020insC frameshift mutation is not associated with inflammatory bowel disease in Chinese patients of Han nationality.
World J Gastroenterol
2004
;
10
:
1069
-71.
7
Inoue N, Tamura K, Kinouchi Y, et al Lack of common NOD2 variants in Japanese patients with Crohn’s disease.
Gastroenterology
2002
;
123
:
86
-91.
8
Helio T, Halme L, Lappalainen M, et al CARD15/NOD2 gene variants are associated with familially occurring and complicated forms of Crohn’s disease.
Gut
2003
;
52
:
558
-62.
9
Yamazaki K, Takazoe M, Tanaka T, Kazumori T, Nakamura Y Absence of mutation in the NOD2/CARD15 gene among 483 Japanese patients with Crohn’s disease.
J Hum Genet
2002
;
47
:
469
-72.
10
Ogura Y, Inohara N, Benito A, et al Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-kappaB.
J Biol Chem
2001
;
276
:
4812
-8.
11
Inohara N, Koseki T, del Peso L, et al Nod1, an Apaf-1-like activator of caspase-9 and nuclear factor-kappaB.
J Biol Chem
1999
;
274
:
14560
-7.
12
Inohara N, Nunez G The NOD: a signaling module that regulates apoptosis and host defense against pathogens.
Oncogene
2001
;
20
:
6473
-81.
13
Kurzawski G, Suchy J, Kladny J, et al The NOD2 3020insC mutation and the risk of colorectal cancer.
Cancer Res
2004
;
64
:
1604
-6.
14
Salovaara R, Loukola A, Kristo P, et al Population-based molecular detection of hereditary nonpolyposis colorectal cancer.
J Clin Oncol
2000
;
18
:
2193
-200.
15
Aaltonen LA, Salovaara R, Kristo P, et al Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease.
N Engl J Med
1998
;
338
:
1481
-7.
16
Arnott ID, Nimmo ER, Drummond HE, et al NOD2/CARD15, TLR4 and CD14 mutations in Scottish and Irish Crohn’s disease patients: evidence for genetic heterogeneity within Europe?.
Genes Immun
2004
;
5
:
417
-25.
17
Croucher PJ, Mascheretti S, Hampe J, et al Haplotype structure and association to Crohn’s disease of CARD15 mutations in two ethnically divergent populations.
Eur J Hum Genet
2003
;
11
:
6
-16.