Purpose: Juvenile polyposis syndrome (JPS) can be caused by a germline defect of the SMAD4 gene. Somatic inactivation of SMAD4 occurs in pancreatic and colorectal cancers and is reflected by loss of SMAD4 immunohistochemistry. Here, SMAD4 immunohistochemistry as a marker of SMAD4 gene status and the role of SMAD4 in the adenoma-carcinoma sequence in neoplastic progression in JPS are studied.

Experimental Design: Twenty polyps with a SMAD4 germline defect and 38 control polyps were studied by SMAD4 immunohistochemistry. Inactivation of the SMAD4 wild-type allele was studied in dysplastic epithelium and in areas with aberrant SMAD4 expression. APC, β-catenin, p53, and K-ras were studied to evaluate the adenoma-carcinoma sequence.

Results: Nine of 20 polyps with a SMAD4 germline defect showed loss of epithelial SMAD4 expression. Loss of heterozygosity of SMAD4 was found in five polyps and a somatic stop codon mutation was found in two polyps without loss of heterozygosity. Remarkably, somatic inactivation of epithelial SMAD4 did not always coincide with dysplasia and aberrant p53 staining was found in four of six dysplastic polyps with normal SMAD4 staining. One K-ras mutation was found in nine juvenile polyps with dysplasia. No evidence for Wnt activation was found.

Conclusions: SMAD4 immunohistochemistry mirrors genetic status and provides a specific adjunct in the molecular diagnosis of JPS. However, epithelial SMAD4 inactivation is not required for polyp formation and is not obligatory for neoplastic progression in JPS. Instead, different routes to neoplasia in JPS caused by germline SMAD4 mutation seem to be operative, including somatic loss of SMAD4 and p53 inactivation without somatic loss of SMAD4. Clin Cancer Res; 16(16); 4126–34. ©2010 AACR.

Translational Relevance

The current study shows that SMAD4 immunohistochemistry mirrors genetic status and can be used as a first screening method in the molecular diagnosis of juvenile polyposis syndrome (JPS). A germline SMAD4 mutation is likely if absent or reduced SMAD4 expression is found in a juvenile polyp.

In addition, this study increases our understanding of JPS pathogenesis. It is shown that biallelic SMAD4 inactivation is not required for polyp formation and is not obligatory for neoplastic progression in juvenile polyps. Moreover, different routes to neoplasia in JPS caused by germline SMAD4 mutation seem to be operative, including somatic loss of SMAD4 and p53 inactivation without somatic loss of SMAD4.

Juvenile polyposis syndrome (JPS) is an autosomal dominant disorder characterized by the presence of distinct juvenile polyps in the gastrointestinal tract and an increased colorectal cancer risk (13). On histology, juvenile polyps have a prominent stromal compartment containing distorted and cystically dilated crypts often lined by reactive epithelium (4). A germline mutation in the SMAD4 or BMPR1A gene is found in 50% of patients (5, 6). Both genes are involved in the transforming growth factor-β/bone morphogenic protein (BMP) signaling pathway, which regulates cell proliferation and differentiation. SMAD4 is a cytoplasmic co-mediator that forms heteromeric complexes with various receptor-dependent SMADs. These complexes are translocated to the nucleus and regulate DNA transcription (7, 8). Somatic inactivation of the SMAD4 tumor suppressor gene occurs in up to 55% of pancreatic cancers and in other malignancies including colorectal cancer. This occurs through either somatic intragenic mutation with loss of the second allele [loss of heterozygosity (LOH)] or deletion of both alleles (homozygous deletion; refs. 911).

In JPS, the mechanism leading to polyp formation and the role of SMAD4 and BMPR1A are poorly understood. One hypothesis is that juvenile polyps develop through a “landscaper defect” in which the defective cell population lies in the stromal compartment. Neoplasia of the epithelial cells may take place as a result of an abnormal microenvironment (12, 13). Others suggest that inactivation of the second allele in the epithelial cell compartment is likely to initiate polyp formation (1416). Different mechanisms of polyp formation may exist for individuals with either a SMAD4 or a BMPR1A germline mutation (12).

In pancreatic cancer, somatic inactivation of SMAD4 is accurately mirrored by loss of immunohistochemical staining (17). Similarly, SMAD4 immunohistochemistry may prove a valuable tool in the molecular diagnosis of JPS. Also, this analysis could clarify the role of this gene in juvenile polyp development and disease progression. This understanding has been hampered by lack of studies systematically showing a correlation between SMAD4 immunohistochemistry and SMAD4 gene status in JPS. Therefore, we investigated SMAD4 protein expression by immunohistochemistry and correlated this result with SMAD4 gene status in juvenile polyps carrying a SMAD4 germline defect. In addition, we addressed the role and timing of somatic loss of the wild-type SMAD4 allele and the conventional adenoma-carcinoma sequence in neoplastic progression in JPS.

Patients and tissue samples

Archival material from patients with one or more juvenile polyps was collected from The Johns Hopkins Polyposis Registry and clinic (Baltimore, MD) and two academic hospitals in the Netherlands (Academic Medical Center, Amsterdam, and University Medical Center, Utrecht). The study was carried out according to the guidelines of the ethical committees of these institutions and with their approval. Clinical and family history data were examined and polyps were carefully reviewed by an experienced gastrointestinal pathologist (G.J.A.O.) to confirm the diagnosis of JPS or sporadic juvenile polyps. All JPS patients previously underwent genetic analysis through direct sequencing and multiplex ligation-dependent probe amplification analysis (5).

Forty-one patients were included in this study, including 8 patients with a SMAD4 germline defect, 6 with a BMPR1A germline defect, and 27 with sporadic juvenile polyps. Polyp tissue was formalin-fixed and paraffinized according to standard procedures.

Immunohistochemistry

Immunohistochemistry was done using a monoclonal antibody against SMAD4 (Santa Cruz Biotechnology, Inc.; 1:400), p53 (Neomarkers, DO7+BP53-12; 1:2,000) and β-catenin (BD Transduction Laboratories clone 14; 1:5,000). Briefly, 4-μm sections were deparaffinized and blocked for endogenous peroxidase activity by immersion in 0.3% H2O2 in methanol for 20 minutes. Antigen retrieval was done in Tris-EDTA buffer (10 mmol/L/1 mmol/L, pH 9.0) for 10 minutes at 120°C. Nonspecific binding sites were blocked in PBS with 10% normal goat serum for SMAD4 and in PBS with 5% normal goat serum for p53 and β-catenin for 10 minutes. This was followed by a 1-hour antibody incubation for SMAD4 and p53 at room temperature and overnight incubation for β-catenin at 4°C. Antibody binding was visualized using the Powervision+poly-HRP detection system (ImmunoVision Technologies, Co.) and PowerDAB (Immunologic) for SMAD4 and the Powervision+poly-HRP detection system and 3,3-diamino-benzidine (DAB, Sigma) for p53 and β-catenin. Sections were counterstained with hematoxylin.

Scoring of immunohistochemistry

On examination, slides were scored as having normal, reduced, or loss of SMAD4 expression. Normal nuclear staining in the epithelial cells lining normal crypts or inflammatory cells in the mesenchymal stroma on the same section served as an internal control (i.e., normal expression refers to the same expression as seen in these control cells). Loss of expression was defined as absence of nuclear staining. Reduced expression was graded when a weaker expression, but not a complete absence of nuclear staining, was noted compared with the control cells (Fig. 1). p53 immunohistochemistry was scored as either normal constitutive immunoreactivity or as a staining pattern suggesting mutation of the p53 gene, which can be reflected by either very intense immunostaining suggesting a stabilizing p53 mutation or total absence of p53 immunoreactivity consistent with a stop codon mutation in the p53 gene (18). β-Catenin immunohistochemistry was scored as either normal membranous or nuclear staining, indicating activation of Wnt signaling (19). Also, all sections were reviewed for dysplasia (G.J.A.O. and F.J.W.K.) using standard H&E-stained reference slides. Dysplasia was graded according to the standard criteria (20).

Fig. 1.

SMAD4 immunohistochemical scoring. SMAD4 immunohistochemistry was scored as normal (A), reduced (B), or loss of expression (C). Nuclear staining in the epithelial cells lining normal crypts or inflammatory cells in the mesenchymal stroma on the same section served as internal control. Note loss of SMAD4 expression in nonneoplastic epithelium in C. Magnification, ×20.

Fig. 1.

SMAD4 immunohistochemical scoring. SMAD4 immunohistochemistry was scored as normal (A), reduced (B), or loss of expression (C). Nuclear staining in the epithelial cells lining normal crypts or inflammatory cells in the mesenchymal stroma on the same section served as internal control. Note loss of SMAD4 expression in nonneoplastic epithelium in C. Magnification, ×20.

Close modal

Laser microdissection and DNA isolation

Epithelium of interest was isolated by laser capture microdissection using the PALM Laser Microbeam Microdissection System (Microlaser Technologies) on 8-μm sections counterstained with hematoxylin. DNA was obtained using TK buffer [400 μg/mL of proteinase K and 0.5% Tween 20, 50 mmol/L Tris (pH 9), 1 mmol/L NaCl, 2 mmol/L EDTA]. After overnight incubation in 50 μL of TK buffer at 56°C, tubes were incubated at 95°C for 10 minutes to inactivate proteinase K (21).

LOH analysis

LOH was assessed using fluorescently labeled primers for the following microsatellites: D18S46, D18S474, D18S858, and D18S64 (16, 22, 23). Epithelium with aberrant SMAD4 expression was separated from normal SMAD4-stained epithelium using laser capture microdissection. After PCR amplification, the products were separated using an ABI Prism 310 genetic analyzer (Applied Biosystems). One microliter of the PCR product was mixed with 23 μL of formamide and 0.5 μL of GeneScanTM ROX-500 (Invitrogen Corporation) as a size marker.

Samples with two distinctly sized alleles of a particular marker were termed informative. For all informative markers, the allelic imbalance factor was calculated as described by Cawkwell et al. (24). LOH was assumed if the allelic imbalance factor was >1.6 or <0.6. Observed losses were confirmed to exclude induced LOH. If retention of heterozygosity was found, microdissected material was sequenced to establish whether a somatic point mutation of the SMAD4 gene had occurred.

Mutation analysis

Sequencing of SMAD4 was done as described previously (5). For APC and K-ras mutation analysis, DNA was isolated from polyps with dysplasia and PCR amplified using PlatinumTaq DNA Polymerase (Invitrogen). Four primer sets covering the mutation cluster region (MCR) in exon 15 of the APC gene (25) were used (1For-GAAATAGGATGTAATCAGACG, 1Rev-CGCTCCTGAAGAAAATTCAAC; 2For-ACTGCAGGGTTCTAGTTTATC, 2Rev-GAGCTGGCAATCGAACGACT; 3For-TACTTCTGTCAGTTCACTTGATA, 3Rev-ATTTTTAGGTACTTCTCGCTTG; 4For-AAACACCTCCACCACCTCC, 4Rev-GCATTATTCTTAATTCCACATC). Two primer sets were used for K-ras mutation analysis for exon 1 and exon 2 containing mutational hotspot codons 12, 13, and 61 (Exon1For-CTGGTGGAGTAT TTGATAGT, Exon1Rev-ATG GTCCTGCACCAGTAATA; Exon2For-GTGCACTGTAATAATCCAGAC, Exon2Rev-CCACCTATAATGGTGAATATCT). PCR products were subsequently sequenced using an ABI Prism 3130 genetic analyzer.

SMAD4 immunohistochemistry

A total of 58 polyps, including 20 polyps from 8 patients with a SMAD4 germline defect, 11 polyps from 6 patients with a BMPR1A germline defect, and 27 sporadic juvenile polyps from 27 patients, were assessed for SMAD4 protein expression using immunohistochemistry (Fig. 1). Of 20 polyps with a SMAD4 germline defect, 9 showed focal reduction or loss of nuclear SMAD4 protein expression in the epithelium (Table 1). In contrast, none of the 11 polyps carrying a BMPR1A germline mutation and none of the 27 sporadic juvenile polyps had aberrant SMAD4 expression (data not shown).

Table 1.

SMAD4 immunohistochemistry in juvenile polyps from patients with a SMAD4 germline mutation

graphic
 
graphic
 

LOH and mutation analyses

To assess the implication of aberrant epithelial SMAD4 protein expression, we investigated whether reduction or loss of SMAD4 expression correlated with occurrence of a somatic event in SMAD4 (i.e., LOH or a somatic point mutation in polyps with a SMAD4 germline mutation). LOH analysis of the SMAD4 locus was done using four microsatellite markers. Nine polyps were assessed, all carrying a germline mutation in SMAD4, and all had aberrant SMAD4 expression. Results are summarized in Table 2.

Table 2.

SMAD4 LOH and mutation analyses in juvenile polyps from patients with a SMAD4 germline mutation

graphic
 
graphic
 

Polyps 2.3, 3.1, 8.1, 8.2a, and 8.4a with reduction or loss of nuclear SMAD4 expression showed LOH in two or more markers surrounding SMAD4, including at least one of two markers closest to the SMAD4 locus. Retention of heterozygosity was found in polyps 1.1 and 7.3 even though SMAD4 expression was reduced or lost. Subsequent sequence analysis revealed a somatic stop codon mutation in exon 1 (1.1) and exon 2 (7.3) of SMAD4, likely resulting in truncation of the protein. In polyps 4.1 and 5.1 with a hemizygous germline deletion of SMAD4 and immunohistochemical loss of the SMAD4 protein, LOH markers closest to SMAD4 were noninformative, although more distant markers showed LOH.

Dysplasia and genetic status of SMAD4

With aberrant epithelial SMAD4 protein expression reflecting the occurrence of a somatic event in the SMAD4 tumor suppressor gene, we investigated the association of these phenomena to neoplastic change in juvenile polyps by reviewing all corresponding H&E slides for dysplasia. In 9 of 20 polyps with a SMAD4 germline defect, foci of low-grade dysplasia were found, two of which contained focal high-grade dysplasia. Four polyps were graded indefinite for dysplasia and seven negative for dysplasia (Table 3).

Table 3.

Dysplasia in juvenile polyps and SMAD4, APC, β-catenin, K-ras, and p53 status

PatientPolypDysplasiaSMAD4 IHCAPC mutationK-ras mutationβ-Catenin IHCp53 IHC
1.1 LGD Reduced NM failed 
1.2 Neg. Normal — — 
2* 2.1 Indef. Normal — — 
2.2 Neg. Normal — — 
2.3 Indef. Reduced — — 
3* 3.1 LGD Reduced NM NM 
3.2 Indef. Normal — — 
4 4.1 Neg. Loss — — 
5 5.1 Neg. Loss — — 
5.2 Neg. Normal — — 
6.1 Neg. Normal — — 
6.2 HGD Normal NM NM 
6.3 HGD Normal NM NM 
7.1 LGD Normal NM NM 
7.2 LGD Normal NM NM 
7.3 LGD Loss NM Codon 12 
8.1 Indef. Loss — — 
8.2a Neg. Loss — — 
8.2b LGD Normal NM NM 
8.3 Neg. Normal — — 
8.4a Neg. Loss — — 
8.4b LGD Normal NM NM 
PatientPolypDysplasiaSMAD4 IHCAPC mutationK-ras mutationβ-Catenin IHCp53 IHC
1.1 LGD Reduced NM failed 
1.2 Neg. Normal — — 
2* 2.1 Indef. Normal — — 
2.2 Neg. Normal — — 
2.3 Indef. Reduced — — 
3* 3.1 LGD Reduced NM NM 
3.2 Indef. Normal — — 
4 4.1 Neg. Loss — — 
5 5.1 Neg. Loss — — 
5.2 Neg. Normal — — 
6.1 Neg. Normal — — 
6.2 HGD Normal NM NM 
6.3 HGD Normal NM NM 
7.1 LGD Normal NM NM 
7.2 LGD Normal NM NM 
7.3 LGD Loss NM Codon 12 
8.1 Indef. Loss — — 
8.2a Neg. Loss — — 
8.2b LGD Normal NM NM 
8.3 Neg. Normal — — 
8.4a Neg. Loss — — 
8.4b LGD Normal NM NM 

Abbreviations: LGD, low-grade dysplasia; HGD, high-grade dysplasia; Indef., indefinite for dysplasia; Neg., negative for dysplasia; NM, no mutation found; M, membranous β-catenin staining pattern; N, normal p53 staining pattern; A, aberrant p53 staining pattern (i.e., overexpression of absent staining).

*Patients 2 and 3 were from the same family.

Patients 4 and 5 were from the same family.

Intriguingly, the presence of dysplasia did not consistently correlate with reduction or loss of nuclear SMAD4 protein expression in juvenile polyps (Tables 3 and 4). Polyps 6.2, 6.3, and 7.1 showed dysplasia even though nuclear SMAD4 expression of the epithelium was normal (Fig. 2A), whereas polyps 4.1 and 5.1 showed loss of epithelial SMAD4 expression but had no dysplasia (Fig. 1C). Polyps 1.1 and 3.1 had foci of low-grade dysplasia within a larger area of reduced epithelial nuclear SMAD4 expression (Fig. 2B), but in polyp 7.3, areas of low-grade dysplasia extended beyond the area showing loss of expression of SMAD4 (Fig. 2C).

Table 4.

Correlation between dysplasia and SMAD4 immunostaining in juvenile polyps

Normal SMAD4 IHC, n (%)Reduced/lost SMAD4 IHC, n (%)Total
No dysplasia 5 (56) 4 (44) 
Indefinite for dysplasia 2 (50) 2 (50) 
Dysplasia (LGD and HGD) 6 (66) 3 (33) 
Normal SMAD4 IHC, n (%)Reduced/lost SMAD4 IHC, n (%)Total
No dysplasia 5 (56) 4 (44) 
Indefinite for dysplasia 2 (50) 2 (50) 
Dysplasia (LGD and HGD) 6 (66) 3 (33) 
Fig. 2.

SMAD4 immunohistochemistry and dysplasia. A, dysplasia with normal epithelial SMAD4 expression. B, dysplasia within area of reduced SMAD4 expression. C, dysplasia extending beyond area of SMAD4 loss. D, dysplasia with normal SMAD4 expression and adjacent nonneoplastic epithelium with loss of SMAD4 expression. Magnification: left, ×10 (D, ×5); right, ×20 (zoomed images of the boxed areas on the left).

Fig. 2.

SMAD4 immunohistochemistry and dysplasia. A, dysplasia with normal epithelial SMAD4 expression. B, dysplasia within area of reduced SMAD4 expression. C, dysplasia extending beyond area of SMAD4 loss. D, dysplasia with normal SMAD4 expression and adjacent nonneoplastic epithelium with loss of SMAD4 expression. Magnification: left, ×10 (D, ×5); right, ×20 (zoomed images of the boxed areas on the left).

Close modal

Remarkably, polyps 8.2 and 8.4 both showed loss of SMAD4 expression in nonneoplastic epithelium (8.2a and 8.4a), but the same sections also contained low-grade dysplasia with normal SMAD4 expression (8.2b and 8.4b; Fig. 2D).

To confirm that SMAD4 immunohistochemistry accurately mirrors the molecular status of SMAD4, we aimed to exclude somatic inactivation of SMAD4 in juvenile polyp tissue with dysplasia and a normal SMAD4 staining pattern. Therefore, dysplastic epithelium with normal nuclear SMAD4 expression was microdissected and analyzed for LOH and somatic mutation using nondysplastic epithelium with normal nuclear SMAD4 expression as a reference. As shown in Table 2, polyps 6.2, 7.1, 8.2b, and 8.4b all had retention of heterozygosity of the SMAD4 locus and no somatic mutations were found.

Role of APC, β-catenin, K-ras, and p53 in neoplastic progression in JPS

To investigate whether mutations in the conventional adenoma-carcinoma sequence underlie neoplastic change in juvenile polyps without loss of the wild-type SMAD4 allele (i.e., with normal SMAD4 protein expression), APC (MCR) and K-ras mutation analysis, as well as p53 and β-catenin immunohistochemistry, was done. Results are summarized in Table 3. One somatic K-ras mutation was found in codon 12 (GGT→GAT) in polyp 7.3 with low-grade dysplasia and loss of SMAD4 immunostaining. Nonneoplastic areas from the same polyp did not show this K-ras mutation. Aberrant p53 staining, suggesting a p53 mutation, was found in six polyps. Five of these polyps contained dysplasia and one was graded indefinite for dysplasia. Interestingly, four of these polyps (66%) showed normal SMAD4 expression. No mutations were found in the MCR of the APC gene and no aberrant β-catenin expression was found.

SMAD4 is one of two known genes responsible for JPS when mutated in the germline. SMAD4 is a tumor suppressor gene and is frequently inactivated in advanced stages of pancreatic and colorectal cancers. In pancreatic cancer, loss of immunohistochemical labeling in tumor cells reflects genetic status of SMAD4 with high accuracy (17).

The role of SMAD4 in JPS polyp formation is poorly understood. Investigators supporting the landscaper theory postulate that juvenile polyps arise primarily due to a stromal defect. The abnormal stroma causes disruption of normal development and regeneration of the overlying epithelium (12, 13). In contrast, other studies provide evidence that LOH of SMAD4 in the epithelium initiates polyp growth, suggesting that SMAD4 acts as a classic tumor suppressor protein in JPS polyps (1416). In fact, it is deemed likely that a second hit of the wild-type allele initiates the growth and neoplastic progression of JPS polyps, which fits the classic tumor suppressor model (16).

In this study, we illuminate the role of SMAD4 in juvenile polyp formation by investigating SMAD4 protein expression and SMAD4 gene status in juvenile polyps from eight patients with a germline SMAD4 mutation. In almost half of all polyps from patients with a SMAD4 germline defect, focal reduction or loss of nuclear SMAD4 expression in the epithelium was seen. In contrast, no aberrant SMAD4 expression was noted in polyps from patients with a BMPR1A mutation or in any of the sporadic juvenile polyps.

Aberrant SMAD4 immunostaining in JPS showed clear correlation with somatic inactivation of the SMAD4 gene. A second hit of the wild-type SMAD4 allele was found in seven of nine polyps with aberrant SMAD4 expression. This included LOH in five polyps and a somatic stop codon mutation resulting in truncation of the SMAD4 protein in two others. It proved difficult to assess LOH status using the microsatellite technique in two polyps (4.1 and 5.1) from two patients with a hemizygous germline deletion of all 11 exons of SMAD4. LOH analysis gave unreliable or noninformative results because the full extent of the germline deletion was not known. However, markers located further away from the SMAD4 gene locus showed LOH in these polyps.

These results clearly show that aberrant nuclear SMAD4 protein expression in JPS patients is indicative of somatic inactivation through LOH or somatic mutation, as has previously been shown in pancreatic cancer (17). Furthermore, reduction or loss of epithelial SMAD4 expression in the polyps of individuals with JPS is specific for the presence of a SMAD4 germline defect, ranging from missense mutations to hemizygous deletions. Therefore, SMAD4 immunohistochemistry can be used as a first screening method in the molecular diagnosis of JPS. An underlying germline SMAD4 mutation is likely if reduced (compared with surrounding stroma) or absent SMAD4 expression is found in the epithelial component of a juvenile polyp. However, normal SMAD4 expression is less predictive of germline status.

Moreover, because focal loss of epithelial SMAD4 expression was found only in a subset of juvenile polyps with a SMAD4 germline mutation, inactivation of the wild-type SMAD4 allele in the epithelium is not required for polyp initiation and formation, but rather occurs as a late event during polyp growth and neoplastic progression. This concurs with previous observations in mouse models of JPS reporting that haploinsufficiency is sufficient for polyp initiation (26, 27).

One study by Kim et al. reported that targeted inactivation of Smad4 in stromal T cells leads to a JPS-like phenotype and epithelial cancers of the gastrointestinal tract in mice, whereas inactivation in the epithelium does not (28). Although our results argue that inactivation of SMAD4 occurs in the epithelium and not in the stroma of juvenile polyps, we cannot eliminate the concept that haploinsufficiency of SMAD4 in cells of the stromal compartment contributes to juvenile polyp initiation as per the landscaper theory. In fact, our finding that epithelial inactivation of SMAD4 is not required for polyp initiation suggests that this may indeed be the case.

Surprisingly, we found that the majority of polyps with dysplasia showed normal SMAD4 protein expression (66%), whereas loss of SMAD4 expression was slightly more common in nonneoplastic polyps than in juvenile polyps with dysplasia (44% versus 33%; Table 4). With regard to SMAD4 in neoplastic progression, this finding suggests that neoplastic change of the epithelium in juvenile polyps with a SMAD4 germline defect is not necessarily initiated by inactivation of the wild-type SMAD4 allele (Fig. 2A), conflicting with the proposed gatekeeper function of SMAD4 in JPS (16). Rather, these results suggest an alternative pathway leading to neoplasia in JPS with somatic inactivation of SMAD4 as a late event during neoplastic progression, in accordance with its role in the conventional adenoma-carcinoma sequence in colorectal cancer (29). Although in our study evidence for (early) Wnt pathway activation was not found, p53 accumulation occurred in four of six polyps with dysplasia and normal SMAD4 immunostaining. Others reported somatic APC mutations in dysplastic juvenile polyps (30), and our study used only APC MCR mutation analysis and β-catenin immunohistochemistry; however, also others did not find support for a major role for Wnt pathway activation in early neoplastic development in JPS (14, 27, 31).

On the other hand, somatic inactivation of SMAD4 also occurred in epithelium without morphologic features of dysplasia in 44% of juvenile polyps (Table 4; Fig. 1C). In some cases, this was observed on the very same section containing areas of low-grade dysplasia with normal SMAD4 expression (Fig. 2D).

Consequently, the role of SMAD4 in neoplastic progression of juvenile polyps remains unclear. Although SMAD4 inactivation is seen in a clonal pattern, it occurs seemingly independent of microscopically evident neoplastic change. Perhaps the most likely scenario is that two pathways causing neoplasia occur. In juvenile polyps carrying a SMAD4 germline defect, an increased selective pressure leading to early stage inactivation of this gene may exist. This molecular marker of neoplasia can be visualized by loss of SMAD4 immunohistochemical staining but may on microscopy of the H&E section not yet be recognizable as dysplasia. Alternatively, selective pressure may also be increased on other genes capable of initiating neoplastic change, such as p53. This could be a direct result of the SMAD4 germline defect or from the abnormal microenvironment present in juvenile polyps. In addition, epigenetic silencing of genes may be important in this model. Somatic inactivation of SMAD4 may then occur at a later stage, possibly leading to acceleration of the neoplastic progression. Alternatively, retention of a wild-type SMAD4 allele may also enable polyps to benefit from the tumor-promoting actions by the transforming growth factor-β signaling pathway (8).

Although this study was performed with a limited number of patients, several important conclusions can be drawn. First, we found that SMAD4 immunohistochemistry accurately reflects SMAD4 status in polyps of the JPS and can be a useful and specific adjunct to the molecular diagnosis of JPS. Second, somatic inactivation of SMAD4 occurs in the epithelium but is presumably not a prerequisite for neoplastic change. Our results suggest that various pathways can lead to neoplasia in JPS caused by germline mutation of SMAD4. One pathway is initiated by somatic loss of SMAD4 and another is characterized by p53 inactivation with retention of the wild-type SMAD4 allele.

No potential conflicts of interest were disclosed.

Grant Support: The Netherlands Digestive Disease Foundation (MLDS WS 04-06), The John G. Rangos, Sr. Charitable Foundation, The Clayton Fund, and NIH grants CA 53801, 63721, 51085, and P50 CA 93-16. The study sponsors were not involved in study design; collection, analysis, and interpretation of data; writing of the report; or the decision to submit the paper for publication.

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.

1
Giardiello
FM
,
Offerhaus
GJ
,
Krush
AJ
, et al
. 
Risk of hepatoblastoma in familial adenomatous polyposis
.
J Pediatr
1991
;
119
:
766
8
.
2
Jass
JR
,
Williams
CB
,
Bussey
HJ
,
Morson
BC
. 
Juvenile polyposis—a precancerous condition
.
Histopathology
1988
;
13
:
619
30
.
3
Brosens
LA
,
van Hattem
A
,
Hylind
LM
, et al
. 
Risk of colorectal cancer in juvenile polyposis
.
Gut
2007
;
56
:
965
7
.
4
Aaltonen
LA
,
Jass
JR
,
Howe
JR
. 
Juvenile polyposis
. In:
Hamilton
SR
,
Aaltonen
LA
, editors.
Pathology and genetics of tumours of the digestive system
.
Lyon
:
IARC Press
; 
2000
, p.
130
2
.
5
van Hattem
WA
,
Brosens
LA
,
de Leng
WW
, et al
. 
Large genomic deletions of SMAD4, BMPR1A and PTEN in juvenile polyposis
.
Gut
2008
;
57
:
623
7
.
6
Aretz
S
,
Stienen
D
,
Uhlhaas
S
, et al
. 
High proportion of large genomic deletions and a genotype phenotype update in 80 unrelated families with juvenile polyposis syndrome
.
J Med Genet
2007
;
44
:
702
9
.
7
Massague
J
,
Blain
SW
,
Lo
RS
. 
TGFβ signaling in growth control, cancer, and heritable disorders
.
Cell
2000
;
103
:
295
309
.
8
Akhurst
RJ
,
Derynck
R
. 
TGF-β signaling in cancer—a double-edged sword
.
Trends Cell Biol
2001
;
11
:
S44
51
.
9
Schutte
M
,
Hruban
RH
,
Hedrick
L
, et al
. 
DPC4 gene in various tumor types
.
Cancer Res
1996
;
56
:
2527
30
.
10
Thiagalingam
S
,
Lengauer
C
,
Leach
FS
, et al
. 
Evaluation of candidate tumour suppressor genes on chromosome 18 in colorectal cancers
.
Nat Genet
1996
;
13
:
343
6
.
11
Hahn
SA
,
Schutte
M
,
Hoque
AT
, et al
. 
DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1
.
Science
1996
;
271
:
350
3
.
12
Haramis
AP
,
Begthel
H
,
van den Born
M
, et al
. 
De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine
.
Science
2004
;
303
:
1684
6
.
13
Kinzler
KW
,
Vogelstein
B
. 
Landscaping the cancer terrain
.
Science
1998
;
280
:
1036
7
.
14
Takaku
K
,
Miyoshi
H
,
Matsunaga
A
, et al
. 
Gastric and duodenal polyps in Smad4 (Dpc4) knockout mice
.
Cancer Res
1999
;
59
:
6113
7
.
15
Woodford-Richens
K
,
Bevan
S
,
Churchman
M
, et al
. 
Analysis of genetic and phenotypic heterogeneity in juvenile polyposis
.
Gut
2000
;
46
:
656
60
.
16
Woodford-Richens
K
,
Williamson
J
,
Bevan
S
, et al
. 
Allelic loss at SMAD4 in polyps from juvenile polyposis patients and use of fluorescence in situ hybridization to demonstrate clonal origin of the epithelium
.
Cancer Res
2000
;
60
:
2477
82
.
17
Wilentz
RE
,
Su
GH
,
Dai
JL
, et al
. 
Immunohistochemical labeling for dpc4 mirrors genetic status in pancreatic adenocarcinomas: a new marker of DPC4 inactivation
.
Am J Pathol
2000
;
156
:
37
43
.
18
Baas
IO
,
Mulder
JW
,
Offerhaus
GJ
,
Vogelstein
B
,
Hamilton
SR
. 
An evaluation of six antibodies for immunohistochemistry of mutant p53 gene product in archival colorectal neoplasms
.
J Pathol
1994
;
172
:
5
12
.
19
Iwamoto
M
,
Ahnen
DJ
,
Franklin
WA
,
Maltzman
TH
. 
Expression of β-catenin and full-length APC protein in normal and neoplastic colonic tissues
.
Carcinogenesis
2000
;
21
:
1935
40
.
20
Hamilton
SR
,
Vogelstein
B
,
Kudo
S
, et al
. 
Carcinoma of the colon and rectum
. In:
Hamilton
SR
,
Aaltonen
LA
, editors.
Pathology and genetics of tumours of the digestive system
.
Lyon
:
IARC Press
; 
2000
, p.
101
19
.
21
Brosens
LA
,
Iacobuzio-Donahue
CA
,
Keller
JJ
, et al
. 
Increased cyclooxygenase-2 expression in duodenal compared with colonic tissues in familial adenomatous polyposis and relationship to the −765G→C COX-2 polymorphism
.
Clin Cancer Res
2005
;
11
:
4090
6
.
22
Carvalho
R
,
Milne
AN
,
van Rees
BP
, et al
. 
Early-onset gastric carcinomas display molecular characteristics distinct from gastric carcinomas occurring at a later age
.
J Pathol
2004
;
204
:
75
83
.
23
Hirata
H
,
Matsuyama
H
,
Matsumoto
H
, et al
. 
Deletion mapping of 18q in conventional renal cell carcinoma
.
Cancer Genet Cytogenet
2005
;
163
:
101
5
.
24
Cawkwell
L
,
Bell
SM
,
Lewis
FA
, et al
. 
Rapid detection of allele loss in colorectal tumours using microsatellites and fluorescent DNA technology
.
Br J Cancer
1993
;
67
:
1262
7
.
25
Miyoshi
Y
,
Nagase
H
,
Ando
H
, et al
. 
Somatic mutations of the APC gene in colorectal tumors: mutation cluster region in the APC gene
.
Hum Mol Genet
1992
;
1
:
229
33
.
26
Alberici
P
,
Jagmohan-Changur
S
,
De Pater
E
, et al
. 
Smad4 haploinsufficiency in mouse models for intestinal cancer
.
Oncogene
2006
;
25
:
1841
51
.
27
Xu
X
,
Brodie
SG
,
Yang
X
, et al
. 
Haploid loss of the tumor suppressor Smad4/Dpc4 initiates gastric polyposis and cancer in mice
.
Oncogene
2000
;
19
:
1868
74
.
28
Kim
BG
,
Li
C
,
Qiao
W
, et al
. 
Smad4 signalling in T cells is required for suppression of gastrointestinal cancer
.
Nature
2006
;
441
:
1015
9
.
29
Kinzler
KW
,
Vogelstein
B
. 
Lessons from hereditary colorectal cancer
.
Cell
1996
;
87
:
159
70
.
30
Wu
TT
,
Rezai
B
,
Rashid
A
, et al
. 
Genetic alterations and epithelial dysplasia in juvenile polyposis syndrome and sporadic juvenile polyps
.
Am J Pathol
1997
;
150
:
939
47
.
31
Fogt
F
,
Brown
CA
,
Badizadegan
K
,
Zimmerman
RL
,
Odze
R
. 
Low prevalence of loss of heterozygosity and SMAD4 mutations in sporadic and familial juvenile polyposis syndrome-associated juvenile polyps
.
Am J Gastroenterol
2004
;
99
:
2025
31
.