[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.167.156.247. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Observation
May 2000

Sporadic Trichoepithelioma Demonstrates Deletions at 9q22.3

Author Affiliations

From the Department of Dermatology, University Hospital of Zurich, Zurich, Switzerland (Drs Matt, Xin, Burg, and Böni); and the Laboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, Md (Drs Vortmeyer and Zhuang).

Arch Dermatol. 2000;136(5):657-660. doi:10.1001/archderm.136.5.657
Abstract

Background  Trichoepithelioma (TE) is a benign cutaneous tumor that originates from hair follicles and occurs either in multiple or solitary lesions. Multiple TE is transmitted as an autosomal dominant trait, and a region at 9p21 is thought to be involved in the tumorigenesis. Solitary TE occurs more commonly than multiple TE and is not inherited. Histologically, TE tumors contain horn cysts and abortive hair papillae. A basal cell carcinoma appearance in some or all regions of a TE tumor can happen. In sporadic basal cell carcinoma, frequent deletions at 9q22.3 (Drosophila patched gene) have occurred. The objective of this study is to test whether loss of heterozygosity (LOH) on either 9p21 or on chromosome 9q22.3 could be detected in archival sporadic TE.

Observations  We studied 29 randomly selected cases of sporadic TE by microdissection and polymerase chain reaction using paraffin-embedded, formalin-fixed tissue specimens on glass slides. Analysis was performed with the polymorphic markers IFNA and D9S171 (9p21) as well as D9S15, D9S303, D9S287, and D9S252 (9q22.3).

Results  The LOH at 9q22.3 was identified in 14 (48%) of 29 cases with at least 1 marker, while LOH could not be demonstrated using the markers IFNA and D9S171 (9p21).

Conclusions  The results show that the Drosophila patched gene LOH can be frequently identified in paraffin-embedded sporadic TE after routine processing and indicates a common gatekeeper mechanism for both TE and basal cell carcinoma.

TRICHOEPITHELIOMA (TE) is a benign skin tumor originating from hair follicles. It may occur as a solitary nonfamilial and a multiple-familial type. Multiple-familial TE (McKusick No. 132700)1 is an autosomal dominant disease characterized by the presence of many skin-colored small tumors located predominantly in the nasolabial folds, but also on the nose, forehead, upper lip, and occasionally on the scalp, neck, and upper trunk.2 The tumor suppressor gene believed to be involved in the tumorigenesis of multiple-familial TE has been located in a region at 9p21.3 Several known tumor suppressor genes, including p15, p16, and p19 have been assigned to this region.4

Histologically, TE tumors contain horn cysts, abortive hair papillae, and, uncommonly, areas with the appearance of basal cell carcinoma (BCC) from which it is sometimes almost or completely indistinguishable.2,5,6 The gene for hereditary BCC has been identified and located at 9q22.3-q31, the human homologue of Drosophila patched gene (PTCH), and allelic deletions of the PTCH gene are found in hereditary BCCs.79 In addition, a substantial subset of sporadic BCC also shows PTCH loss of heterozygosity (LOH).10,11 Furthermore, recent studies indicate potential involvement of the PTCH locus in the pathogenesis of TE.12,13 We therefore analyzed 29 tumors for LOH at both 9q22.3 and 9p21.

MATERIALS AND METHODS
HISTOLOGIC SPECIMEN

Formalin-fixed, paraffin-embedded histologic specimens of 29 sporadic TE tumors from 26 different patients (18 men and 8 women) were investigated. None of the patients had a known family history of TE. All TE tumors were nonulcerating, nonbleeding lesions stable for many years.

The histologic specimens were obtained from the archives of the Department of Dermatology, University Hospital of Zurich, Zurich, Switzerland. From each specimen, 1 section was used for hematoxylin-eosin staining and microdissection. To assure correct diagnosis, all histologic slides were reviewed.

MICRODISSECTION TECHNIQUE

In each case, a 5-µm tissue section was obtained for hematoxylin-eosin staining and microdissection. From each section (N=29) we microdissected between 50 and 100 cells of dermal aggregates of basaloid cells with connection to or differentiation toward hair follicles (Figure 1).

Figure 1.
Histologic study of a solitary trichoepithelioma showing tumor islands composed of basophilic cells surrounded by a stroma before (left) and after (right) microdissection (hematoxylin-eosin, original magnification ×200).

Histologic study of a solitary trichoepithelioma showing tumor islands composed of basophilic cells surrounded by a stroma before (left) and after (right) microdissection (hematoxylin-eosin, original magnification ×200).

Microdissection was performed under light microscope visualization (magnification ×200) using a 30-gauge needle. Samples from normal cells apart from TE cell structures (eg, inflammatory cells, sebaceous glands) were also obtained in all cases from the same slide for comparison with the tumor samples.

DNA EXTRACTION

Procured cells (dermal aggregates of basaloid cells as well as samples obtained from normal cell structures apart from TE tumors) were immediately suspended in 30 µL of buffer containing 0.05-mol/L Tris-hydrochloride, 1-mmol/L EDTA, 1% Tween 20, 1-g/L proteinase K (pH 8.0), and incubated 2 days at 37°C. The mixture was boiled for 10 minutes at 94°C to inactivate proteinase K, and 1.5 µL of this solution was used for polymerase chain reaction (PCR). Analysis of LOH was carried out by PCR amplification of microsatellite polymorphisms.

PRIMERS AND PCR CONDITIONS

Two polymorphic DNA markers at the short arm of chromosome 9 (IFNA [9p21] and D9S171 [9p21 ]) and 4 markers at chromosome 9q22.3 (D9S303 [9q 13-9q22.3], D9S15 [9q 13-9q21.1], D9S252 [9q 13-9q22.1], and D9S287 [9q22.3-9q31]) (Research Genetics, Huntsville, Ala) were used in this study. The PCR was performed in 10 µL units that contained 1 µL of 10 × PCR buffer (Boehringer Mannheim GmbH, Mannheim, Germany), 50 pmol of each primer per liter; 20 nmol/L each of dCTP, dGTP, dTTP, dATP; 0.2 µL [32P] of dCTP [22,200 × 1010 Bq/mmol]; and 0.1 U of Taq DNA polymerase). Reactions were cycled in a thermal cycler (Gene Amp PCR System 9600; Perkin Elmer, Zurich), and amplification consisted of 35 cycles of 1 minute at 94°C, 1 minute at 60°C, and 1 minute at 72°C with a final 10-minute extension at 72°C.

Labeled amplified DNA was mixed with an equal volume of formamide loading dye (95% formamide; 20 mmol/L of EDTA; 0.05% bromophenol blue; and 0.05% xylene cyanol). The samples were denatured for 5 minutes at 94°C and loaded onto a gel consisting of 6% acrylamide (49:1 acrylamide-bis). Samples were electrophoresed at 1600 V for 2 hours. Gels were transferred to 3-mm Whatman paper (Merck & Co Inc, Zurich), dried, and subjected to autoradiography with Typon X-Ray DX 41 film (Typon, Burgdorf, Switzerland). A case was considered informative for a polymorphic marker if normal tissue DNA showed 2 different alleles (heterozygosity). Loss of heterozygosity was defined as absence or substantial reduction of one allele in the tumor DNA evaluated by autoradiography.

RESULTS

The LOH on 9q22.3 was identified with at least 1 marker in 14 (48%) of 29 cases. D9S252 and D9S303 showed LOH in 7 (35%) of 20 cases; D9S15 in 4 (33%) of 12; and D9S287 in 2 (11%) of 18 (Figure 2 and Table 1). Results proved homozygous or noninformative for D9S15 in 17 of 29 cases, for D9S303 in 9 of 29, for D9S287 in 11 of 29, and for D9S252 in 9 of 29; these markers were therefore excluded from the calculation. There was consistently no LOH for IFNA or D9S171.

Figure 2.
Sequence gel analysis results obtained after subjection of microdissected cells to single-step DNA extraction and amplification using the markers D9S252 and D9S303. Loss of an allele was scored as a greater than 50% reduction of the band intensity (arrowheads). N indicates normal; T, trichoepithelioma.

Sequence gel analysis results obtained after subjection of microdissected cells to single-step DNA extraction and amplification using the markers D9S252 and D9S303. Loss of an allele was scored as a greater than 50% reduction of the band intensity (arrowheads). N indicates normal; T, trichoepithelioma.

Results of Loss of Heterozygosity (LOH) Analysis in 29 Solitary Trichoepitheliomas Using the Microsatellite Markers IFNA and D9S171 (Both 9p21) and D9S15, D9S303, D9S287, and D9S252 (9q22.3)*
Results of Loss of Heterozygosity (LOH) Analysis in 29 Solitary Trichoepitheliomas Using the Microsatellite Markers IFNA and D9S171 (Both 9p21) and D9S15, D9S303, D9S287, and D9S252 (9q22.3)*
COMMENT

A region at 9p21 has been suspected to be involved in tumorigenesis in multiple familial TE, and deletions within this area harbor known tumor suppressor genes.3 With this study we tried to elucidate whether genetic changes at 9p21 play a role in the tumorigenesis of sporadic TE. In 29 cases of sporadic TE, tumors have been microdissected and examined using microsatellite markers at 9p21 (IFNA and D9S171), and there was no LOH for either marker. These data suggest that sporadic TE (or at least most cases of sporadic TE) has a different pathogenesis from hereditary counterparts.

Although TE and BCC represent 2 different clinical entities, several lines of evidence suggest common pathogenetic features: (1) Trichoepithelioma and BCC share histopathologic features and may be difficult to differentiate.57 (2) They occur together.2,3,1315 (3) Several genetic studies suggest a common origin of TE and BCC: in a subset of TE, somatic mutations in the overexpressed PTCH gene have occurred,16 and in a solitary TE, LOH was only demonstrated at the BCC locus, while LOH could not be demonstrated using markers at 9p21.12

Based on these facts, dermal aggregates of basaloid cells with connection to or differentiation toward hair follicles have been microdissected in 29 sporadic TE tumors, and genetic analysis of the PTCH locus (9q22) was performed using a panel of microsatellite markers. The results showed LOH in 14 of the 29. The rate of 48% LOH of the PTCH gene is strikingly similar to that of sporadic BCC.11 In addition to PTCH alterations, additional genetic changes may be present accounting for the distinct biological potential of these 2 tumor types. This might be further supported by the findings in 2 patients (Nos. 3 and 15), where losses occurred that are inconsistent with a single target region.The demonstrated loss rate at 9q22 in most studied cases is supportive of a putative tumor suppressor gene being common to the development of BCC and TE.

Accepted for publication December 3, 1999.

This study was supported in part by the Lydia Hochstrasser Foundation, Zurich, Switzerland.

Reprints: Roland Böni, MD, Department of Dermatology, University Hospital of Zurich, Gloriastrasse 31, 8091 Zurich, Switzerland (e:mail: rboeni@derm.unizh.ch).

References
1.
McKusick  VA Mendelian Inheritance in Man. 11th ed. Baltimore, Md Johns Hopkins University Press1994;
2.
Lever  WFedSchaumburg-Lever  Ged Tumors of fibrous tissue. Histopathology of the Skin 7th ed. Philadelphia, Pa JB Lippincott Co1990;660- 668
3.
Harada  HHashimoto  KKo  MSH The gene for multiple familial trichoepithelioma maps to chromosome 9p21. J Invest Dermatol. 1996;10741- 43Article
4.
Hussussian  CJStruewing  JPGoidstein  AM  et al.  Germline pl6 mutations in familial melanoma. Nat Genet. 1994;815- 21Article
5.
Brooke  JDFitzpatrick  JEGolity  LE Papillary mesenchymal bodies: a histologic finding useful in differentiating trichoepitheliomas from basal cell carcinomas. J Am Acad Dermatol. 1989;21523- 528Article
6.
Kirchmann  TTPrieto  VGSmoiler  BR CD34 staining pattern distinguishes basal cell carcinoma from trichoepithelioma. Arch Dermatol. 1994;130589- 592Article
7.
Farndon  PADel Maestro  RGEvans  DGRKilpatrick  MW Location of the gene for Gorlin syndrome. Lancet. 1992;339581- 582Article
8.
Gailani  MRBale  SJLeffell  DJ  et al.  Developmental defects in Gorlin syndrome related to putative tumor suppressor gene on chromosome 9. Cell. 1992;69111- 117Article
9.
Wicking  CBerkman  JWainwright  BChenevix-Trench  G Fine genetic mapping of the gene for nevoid basal cell carcinoma syndrome. Genomics. 1994;22505- 511Article
10.
Shanley  SMDawkins  HWainwright  BJ  et al.  Fine deletion mapping on the long arm of chromosome 9 in sporadic and familial basal cell carcinomas. Hum Mol Genet. 1995;4129- 133Article
11.
Shen  TPark  WSBöni  RPham  TVortmeyer  AOZhuang  Z Detection of loss of heterozygosity on chromosome 9q22.3 in microdissected sporadic basal cell carcinoma. Hum Pathol. 1999;30284- 287Article
12.
Takata  MQuinn  AGHashimoto  KRees  JL Low frequency of loss of heterozygosity at the nevoid basal cell carcinoma locus and other selected loci in appendageal tumors. J Invest Dermatol. 1996;1061141- 1144Article
13.
Böni  RFogt  FVortmeyer  AOTronic  BZhuang  Z Genetic analysis of a trichoepithelioma and associated basal cell carcinoma [letter]. Arch Dermatol. 1998;1341170- 1171Article
14.
Pariser  RJ Multiple hereditary trichoepitheliomas and basal cell carcinomas. J Cutan Pathol. 1986;13111- 117Article
15.
Johnson  SCBennett  RG Occurrence of basal cell carcinoma among multiple trichoepithelioma. J Am Acad Dermatol. 1993;28322- 326Article
16.
Vorechovsky  IUnden  ABSandstedt  BToftgard  RStahle-Bäckdahl  M Trichoepitheliomas contain somatic mutations in the overexpressed PTCH gene: support for a gatekeeper mechanism in skin tumorigenesis. Cancer Res. 1997;574677- 4678
×