- Email: [email protected]
In vitro hyperdiploidy in a family with nasopharyngeal cancer
In Vitro Hyperdiploidy in a Family with Nasopharyngeal Cancer B. Shannon Danes
ABSTRACT: In this in vitro study of 28 members of a family with nasopharyngeal cancer (NPCJ and nasopharyngea! angiofibromas (NPA) in consecutive generations, in vitro hyperdiploidy, (WH) in dermal fbroblast cultures was associated with NPC in 2 affected members, NPA in 4 affected members, and 4 o f 16 members considered to be at increased genetic risk for these tumors. None of the six controls, all spouses, had IVH. As/VH has been shown to be associated with high specificity (~ 99%) with other heritable single tumors [3], IVH appeared to be a potential biomarker also for genetic predisposition for NPC. As NPC and NPA did not occur in a single affected family member, any association other than with IVH has not been established. IVH may be the in vitro expression of a gene/genes conveying tumor proneness per se, or it may determine tissue specificity of tumor pathology. As cellular hyperdiploidy has been considered to convey genomic instability [15], chromosome instability in numerically normal (diploid) and altered (tetraploidy and hyperdiploid) dermal fbroblasts from two NPC-affected members and two control spouses were evaluated by a parameter of chromosome stability-the frequency of sister chromatid exchanges (SCE) per cell. Irrespective of donor source, the diploid and tetraploid fbroblasts exhibited little evidence of chromosome instability, whereas there was a statistically significant difference in the mean frequency of SCE/cell between hyperdiploid and diploid cells in cultures derived from NPC-affected members (p < 0.01). INTRODUCTION The value of biomarkers for solid tumors is p r o b a b l y most justified in families w i t h the same t u m o r in consecutive generations. A l t h o u g h such families represent o n l y a m i n u t e fraction of the total clinical group for each specific tumor, identification of risk status w o u l d p r o v i d e (a) relevant information for family m e m b e r s a n d (1o) an o p p o r t u n i t y for evaluating the influence of e n v i r o n m e n t a l factors on cells w i t h a n d w i t h o u t genetic p r e d i s p o s i t i o n for a specific tumor. This w o u l d be particularly pertinent for a t u m o r a s s u m e d to result from such an interaction; such a group of tumors are the n a s o p h a r y n g e a l tumors (NPC). Through in vitro family studies on dermal m o n o l a y e r cultures d e r i v e d from skin biopsies, two in vitro n u m e r i c a l alterations in c h r o m o s o m e c o m p l e m e n t have been reported to s h o w an association w i t h the in r i v e expression of certain heritable tumors: increased t e t r a p l o i d y (in vitro tetraploidy, IVT) and h y p e r d i p l o i d y w i t h a normal occurrence of t e t r a p l o i d y (in vitro h y p e r d i p l o i d y , IVH). IVT has been identified w i t h h e r e d i t a r y colon cancer [1, 2], whereas IVI-I (defined as a m e t a p h a s e
From the Laboratoryfor Cell Biology,Department of Medicine, Cornell University Medical College,New York, IVY. Address requests for reprints to Dr. B. Shannon Danes, Laboratory for Cell Biology, Department of Medicine, Cornell University Medical College, New York, NY 10021. Received April 19, 1985; accepted July 3, 1985.
107 Science Publishing Co., Inc. 52 Vanderbilt Ave., New York, NY 10017
© 1986 Elsevier
Cancer Genet Cytogenet 21:107--115 0165-4608/86/$03.50
{1986)
108
B.S. Danes with more than 46 chromosomes exclusive of 92, tetraploidy), has been seen with cancer proneness in families with hereditary single tumors, including breast, melanoma, ovary, and pancreas [3, 4]. The purpose of this article is to demonstrate the potential usefulness of IVH for the identification of genetic predisposition for NPC, using a research family with such tumors in consecutive generations.
MATERIALS AND METHODS
Family Members Studied Subepidermal biopsies were obtained from 28 family members: 6 clinically affected (2 with nasopharyngeal cancer and 4 with nasopharngeal angiofibromas), 16 at risk for either tumor, and 6 spouses who were not at risk. The family had a history of nasopharyngeal cancer (NPC) in three consecutive generations, as well as nasopharyngeal angiofibromas (NPA) in four consecutive generations (Fig. 1). The histopathology was established for both tumors from tissue specimens taken from affected members in generation III (NPC III:3 and III:10; NPA II:8, III:13, IV:8, and V:3). The squamous cell carcinoma was classified as nonkeratinizing carcinoma by the WHO nomenclature:J5]. On the basis of the pedigree (Fig. 1), no association was established between the t w o t u m o r s , as the two lesions did not occur in the same affected family member:'AI1 28 family members completed the same questionnaire for family and personal history, including social habits.
Figure 1 Pedigree of a family with heritable nasopharyngeal cancer (histopathology, squamous cell nonkeratinizing carcinoma) in three consecutive generations and heritable nasopharyngeal fibromas in four consecutive generations.
4
1
[] [] m ['~
2
5
3
S
4
5
9
10
11
12
6
Unaffected with nasophoryngeal cancer or nosophoryngeol ongiofibromo Nosophoryngeol cancer by history (histopotholoQy not established) Squomous ceil corcinomo of nasopharynx (histopothology nonkerotinizin9 corcinon'm) Nosophoryngeal ongioflbromo in vitro hyperdiplotdy defined as a numerically abnormal metaphose containing more than 46 chromosomesexctu$ive of 92, tetroploidy []*Positive if % heter®lold metophoue (heteroploid metophoses over total number metophaess counted) greater than 2% (range in study 11-33%) [--~-Negative if % heteroploid metaphases less than 3% (range in study 0-2%)
"~[~ Deceased
)~ Proband
•
Hyperdiploidy in Nasopharyngeal Cancer
In Vitro Methodology
109
~
Dermal monolayer cultures were established from these split-thickness biopsies by standard culture methods [6]. Cultures were grown in plastic Petri dishes in Eagle's minimum essential medium (EMEM) with 15% (vol) fetal bovine serum (non-heattreated) in an atmosphere of 5% COz in air, pH 7.0-7.4. Cultures were routinely checked for bacterial contamination and by the fluorescent DNA staining procedure of Barile et al. [7] for mycoplasma and were found to be uncontaminated. Cells were grown in culture for 8-15 weeks (3-6 subcultures by trypsinization after the primary explant culture) before assays were performed. A coded number was assigned to each culture, so that its identity would not be known until completion of the assays. To ensure that cells to be assayed were in logarithmic growth, a semiconfluent monolayer culture (approximately 106 cells in a 26 cm 2 growth area] for each individual studied was trypsinized; this cell suspension was plated into a single 100cm z plastic Petri dish (initial cell density 104 cells/cm2). After 48-hr growth, this culture was trypsinized to obtain a cell suspension. Aliquots of this cell suspension were plated into 6 Petri dishes (100 cm z) at an initial cell concentration of 5 x 104 (initial cell density 0.5 × 10 s cells/cm z growth area). The medium was changed in all 6 subcultures after 12 hr. Twenty-four hours after this medium change, cell counts were done on 2 Of the 6 subcultures to determine cell number and density at assay time. The same batch of medium and serum was used in all in vitro studies involved in this in vitro research on this family [8].
Assay for Numerical Alteratiens in Chromosome Complement Twenty-four hr after the medium change, a burst of mitosis was invariably observed. At that time, chromosome preparations were made, without the addition of colchicine, from two of the duplicate subcultures and stained with acetoorcein [6]. For this study, tetraploidy was defined as a metaphase with 92 chromosomes, or ~>88, ~92, with a random loss of no more than 3 chromosomes. Hyperdiploidy was defined as a metaphase containing more than 46 chromosomes, exclusive of 92. The incidence of tetraploid or hyperdiploid metaphases was expressed as the number of tetraploid or hyperdiploid metaphases divided by the total number of metaphases scored on a slide, having at least 100 countable metaphases. Only if more than 100 mitoses were counted per slide was the preparation considered to reflect the mitotic activity of that culture. This study did not include evaluation of structural chromosome aberrations, as reported in in vitro studies on dermal fibroblasts from familial polyposis coli patients [2, 9].
Assay for Frequency of Sister Chromatid Exchanges To determine if the hyperdiploid cells represented a dividing subpopulation within a culture and to assess the extent of chromosome instability in both the diploid and numerically altered (tetraploid and hyperdiploid) cell populations, the frequency of spontaneous chromatid exchanges was determined for cultures derived from two N-PC-affected members and two spouse controls Two of the six subcultures established for assays were used for this purpose. After the culture medium was completely changed to medium containing 0.25 ~zg 5-bromo-2'deoxyuridine (BrdU) per milliliter of medium, the cultures were grown
110
B.S. Danes for two days in complete darkness. Each monolayer was then washed with unused medium and grown for 12 hr in medium without BrdU [10] before fixation for chromosome preparations. Slides were made according to a conventional airdrying method and were stained for SCE by the method described by Chaganti et a1.[11]. The analysis of the preparations was done without knowing the origin of the cultures. The SCE metaphases were identified (position on slide noted), photographed with a Zeiss photomicroscope, and 8 x 10 inch prints made. Counts of SCE were scored on photoprints at first and then checked again under the microscope by two independent observers. There was less than a 5% difference between the SCE counts of the two observers for any cell line studied. Statistical analysis between the frequency of SCE in diploid and hyperdiploid metaphases in cultures derived from NPC-affected family members utilized the Student's t-test for pairwise comparison between these two groups.
RESULTS Generation time for all dermal fibroblast monolayer cultures, irrespective of their human donor source, was 19-21 hr in the density range 0.4-1.8 × 103 cells/cm 2 growth area. All metaphase preparations were done on cultures during documented logarithmic growth in this cell density range. The frequency (%) of tetraploidy was constant for replicate cultures of sublines of a primary culture and showed little variation through six passages (Table 1). None of the cultures showed IVT (over 7% tetraploidy).
Table 1
Incidence (percent) of hyperdiploidy a in metaphase preparations derived from lines established from skin biopsies from two NPC-affected members and two spouse controls from the family studied Subculture number b 1
2
3
Donor source of c e l l l i n e
Ic
2c
Ic
2c
Ic
2c
19 20
20 19
20 25
o 0
o 0
o 1
NPC CLINICALLY AFFECTED Ill:3 III:10
20 22
19 25
23 24
LOW-RISK SPOUSE In:6 III:ll
o 1
o 1
o 1
• lypardiploidy defined as a metaphase w i t h more than 46 chromosomes exclusive of 92, tetraploidy. At least 100 metaphases were counted on a single lhetaphase preparation slide to determine hyperdiploi.dy incidence (percent]. bSubculture by trypsinization once a week after establishment of first subculture from primary biopsy culture ~ u m b e r s refer to replicate culture dishes
111
Hyperdiploidy in Nasopharyngeal Cancer
Frequency of V I I The frequency {%) of IVH in duplicate subcultures from the same culture, irrespective of donor source, showed little variation (Table 1}. The IVH findings were not influenced by the age of the skin donor or the time the cells were in culture (8-15 weeks in culture after the primary explant culture, 3-6 subcultures by trypsinization}.
Clinically affected members. IVH was observed in cultures established from the two family members with a clinical NPC history (111:3 and III:10). There was a wide distribution of the total number of chromosomes per IVH metaphase (72 to >200/ metaphase}. The four members studied with a clinical history of NPA (11:8, 1ii:13, IV:8, V:3) had IVH (11%-25%} that appeared to be identical to that occurring in NPC. The percent of IVH showed no correlation with the chronologic age of the NPA biopsy donor, as 11:8 was 74 years old at the time of biopsy and had 18% IVH, which is midrange for NPA-derived cultures (Table 2). IVH showed an association with both NPC and NPA, but neither tumor occurred in the same family member. In generation IV, the offspring (IV:8) of the NPC patient (II1:7} had IVH in association with NPA, as did his nephew {V:3). Based on pedigree data, family members studied were divided into two risk groups--high and low. Increased risk status was assigned to first-degree relatives of a clinically affected member and to their offspring; low-risk status was given to family members by marriage, i.e., spouses. Increased risk group. Four of the 16 clinically normal members studied who were considered to be at possible increased risk for genetic cancer proneness showed IVH {Fig. 1). Three were considered to be at 50% risk status (IV:7 and IV:9 were offspring of NPC-affected members and IV:IO of an NPA-affected member}. The fourth
Table 2 Family members studied: Pedigree, clinical status, and incidence of numerical alterations {hyperdiploidy and tetraploidy} in dermal flbroblast cultures derived from 28 family members Percent metaphases d
showing Clinical status
Number studied
M/F
Age (yr)e
Clinically affected
NPC ° NPA °
2 4
1:1 2:2
60,66 20,30,56,74
26,33 11,18,19,25
1,3 (0-2}
At increased
NL
12
7:5
(15-68)
(0-2)
(0-1)
genetic risk b L o w risk ~
NL NL
4 6
1:3 4:2
16,36,40,42 (38-68)
18,22,28,30 {0-1)
(0--4) (0-3)
Genetic status
Hyperdiploidye
Tetraploidye
°NPC, nasopharyngeal cancer, histopathology squamous cell nonkeratinizing carcinoma; NPA, nasopharyngeal angiofibromas. bAt increased genetic risk based on pedigree: offspring of at increased risk family members {clinically affected or normal}. CLew-risk
spouses of family members.
dHyperdiploidy is defined as a metaphase with more than 46 chromosomes exclusive of 92, tet~aploidy. Normal incidence of hyperdiploidy is 0 % - 1 % and for tetraploidy 0 % - 7 % in dermal fibroblast cultures [3].At least 100 metaphases were counted on a single chromosome preparation slide to determine the incidence of numerical alterations in chromosome complements. ~Numbers in parentheses represent age range.
112
B.S. Danes IVH-positive family member was V:5, the offspring of disease-free IvH-positive IV:7. Low-r/sk group. None of the six spouse controls studied had IVH (range 0%-1%). (Table 1).
Frequency of SCE The yield of metaphases in all cultures grown in BrdU-containing medium was greatly reduced to that in duplicate subcultures grown in medium without BrdU. In order to increase the number of metaphases present after growth in BrdU-containin 8 medium, each monolayer was washed with unused medium and then grown in medium without BrdU for ~12 hr prior to preparation of metaphase preparations [10]. Such post-BrdU control growth resulted in sufficient metaphases to give a slide preparation containin8! at least 100 metaphases. There was no difference in the mean frequency of SCE in diploid metaphases from cultures derived from NPC-affected members and normal spouse controls (1.33 - 1.02 SCE/cell for NPC versus 1.25 _ 1.49 for spouse controls). Nor was there any difference in their mean frequency in tetraploid metaphases from the same two family groups (1.50 _ 1.45 aCE/cell for NPC versus 2.2 - 1.15 SCE/cell for controis]. However, there was a statistically significant difference in the mean frequency of SCE between diploid and hyperdiploid metaphases in cultures from NPC-affected Individuals, (1.33 ± 1.02 SCE/cell for diploids versus 12.42 ± 8.34 SCE/cell for hyperdiploids t -- 3.47, p ~ 0.01) (Table 3).
DISCUSSION In the majority of NPC cases, there is no Clear genetic pattern, as is seen in the autosomal dominant heritable cancer syndromes. However, on the basis of pedigree (family and clinical) data, there are families with histologically confirmed NPC in
Table 3 Numerical chromosome distribution and frequency of sister chromatid exchanges (SCE) in metaphase preparations derived from lines established from skin biopsies from two NPC-affected members and two spouse controls Chromosome number/ metaphase
Donor source of cell line
Number of SCE metaphases analyzeda
Mean
Range
SCE/cell
Number of SCE
Mean ± SD
Range
298 32 18
12.42 ± 8.34 1.33 ± 1.02 c 1.50 ± 1.45
1-32 0-4 0-4
30 20
1.25 ± 1.49 2.22 ± 1.15
NPC CLINICALLY AFFECTED
(I]I:3 a n d III:lO)
H D T
25 25 10
72 46 90
62-105 45-46 b 89-92
LOW-RISK SPOUSES (III:6 a n d III:11)
D T
25 10
46 91
46 90-92
0-4 0-5
~rypes of metaphases: H, hyperdiploid; D, diploid; T, tetraploid. b4S count due to a presumed broken metaphase. CFrequency of SCE/cell in hyperdiploid metapheses versus frequency of SCE/cell in diploid metapheses from dermal flbroblast cultures derived from NPC-affected family members, p ~ 0.01.
Hyperdiploidy in Nasopharyngeal Cancer
113
consecutive generations or in familial clusters (several members in different generations with NPC without evidence of vertical transmission of its clinical occurrence, i.e., segregation of genetic predisposition for NPC). Such families have been identified in the southern Chinese population [12], and other families, although rare, do exist in other geographical and racial groups. The family described (Fig. 1) in this article fell into the latter category. NPC-affected family member I:1 (by family history without a tissue diagnosis) and his wife (I:2) emigrated from eastern Norway to the United States around 1890. This family met the criteria for a hereditary NPC tumor on the basis of seven clinically affected members in three consecutive generations and vertical transmission from a clinically affected member to an offspring (I:1-->II:1,3,6; II:l--~III:3; II:6-~,III:7). In this family, NPA, a recognized hereditary benign tumor [13], also met the same criteria for a hereditary basis, as there were four affected members in four generations (II:8,III:13, IV:8, V:3) and vertical transmission (II:8->III:13). However, NPC and NPA did not occur in any single family member, and IV:8 with NPA was the offspring of III:7 with NPC. NPA had an earlier age of onset and a broader age range (mean 45 years, range 20-74 years) than NPC (mean 64 years, range 60-66 years). On the basis of pedigree data (Fig. 1) the hereditary association between NPC and NPA could not be clarified. An association between IVH and several hereditary single tumors (breast, melanoma, ovary, and pancreas) has been demonstrated in human dermal fibroblast monolayer cultures derived from families with these heritable single tumors [3, 4]. With the recognition that this numerical alteration in chromosome complement occurred rarely in such cultures from the general population without a family history of such single tumors (specificity was high, approximately 99%-100%) [3, 4], IVH appeared to be a potential marker for genetic predisposition for such hereditary tumors. On the basis of the present family study (Tables 1 and 2), NPC has been added to this list. However, IVH was not gene-specific (the in vitro expression of cancer proneness for a single hereditary tumor), but appeared to be the in vitro expression of cancer proneness for a group of hereditary tumors. So far, it has not been established whether IVH is the in vitro expression of a gene/genes conveying cancer proneness per se, or one determining tissue specificity of cancer expression. Through this in vitro family study, an association between IVH and either NPC or NPA has been demonstrated (Fig. 1; Tables 1 and 2). Sufficient numbers have not been studied from families with NPC, NPA, or both to calculate sensitivity. The biologic significance of IVH has not been established, nor has any relevancy to cancer proneness. Since the mutational theory of cancer was proposed by Boveri in 1914 [14], the possible relationship of cancer to chromosomal rearrangements resulting from chromosome instability has been entertained. Cellular hyperdiploidy has been considered to convey genetic instability [15]. Chromosomal instability was reported [2, 9] for numerical and structural aberrations in cultured human dermal fibroblasts from patients with such genetic colon cancer syndromes as familial polyposis colt and related syndromes. On these bases, it appeared relevant to assay for such chromosomal instability in the present family. Chromosomal instability in numerically normal (diploid) and altered (tetraploid and hyperdiploid) fibroblasts have been evaluated by a parameter of chromosome stability--the frequency of SCE/cell. This assay has been used to detect chromosomal instability in such genetic cancer disorders as Bloom's syndrome [11] and del(13)-retinoblastoma [16]. Two previous in vitro studies [17, 18] failed to agree on whether or not there was a significant difference in the mean SCE frequencies occurring in dermal fibroblast cultures from individuals with and without heritable colon cancer. Irrespective of donor source, the diploid and tetraploid fihroblasts exhibited little evidence of chromosome instability based on the mean frequency of SCE/ceI1, al-
114
B . S . Danes
though there was a statistically significant difference in the m e a n frequency of SCE/ cell b e t w e e n h y p e r d i p l o i d a n d d i p l o i d cells in cultures derived from NPC-affected family m e m b e r s (p ~ 0.01]. There a p p e a r e d to be a s u h p o p u l a t i o n in such fibroblast cultures that was h y p e r d i p l o i d , w i t h an increased m e a n frequency of SCE/cell. As IVH has been s h o w n to be associated w i t h hereditary single tumors [3, 4], it is suggested that such tumors are associated w i t h an increased frequency of hyperd i p l o i d cells w i t h a genomic instability detectable by the SCE assay (Table 3). As all fibroblasts w i t h the p r e s u m e d germinal m u t a t i o n for cancer p r o n e n e s s d i d not show increased SCE frequency/cell (range 0-32, Table 3), this instability might have been c o n v e y e d s e c o n d a r i l y b y a n u m e r i c a l alteration in c h r o m o s o m e complement. However, as the m e a n frequency of SCE/cell was not increased in tetraploid cells, irrespective of source, mere n u m e r i c a l alteration in the c h r o m o s o m e c o m p l e m e n t from d i p l o i d y c o u l d not be responsible for this instability. The h y p e r d i p l o i d y state a p p e a r e d to convey genomic instability detectable as an increase in the m e a n frequency of SCE/ceI1. Supported by Grant 0053 from the Smokeless Tobacco Research Council.
REFERENCES
1. Danes BS (1981): Occurrence of in vitro tetraploidy in the heritable colon cancer syndromes. Cancer 48:1596-1601. 2. Delhanty JDA, Davis MB, Wood J (1983): Chromosome instability in lymphocytes, fibroblasts, and colon epithelial-like cells from patients with familial polyposis coli. Cancer Genet Cytogenet 8:27-50. 3. Danes BS (1984): In vitro expressions of cancer genes for heritable human tumors: Numerical alterations in chromosome complement. Dis Markers 2:371-380. 4. Danes BS, Lynch HT (1982): A familial aggregation of pancreatic cancer. JAMA 247:27982802. 5. Batsakis JG (1979): Tumors of the Head and Neck. Clinical and Pathological Considerations (ed. 2). Williams & Wilkins, Baltimore. 6. Danes BS (1976): The Gardner syndrome: Increased tetraploidy in cultured skin fibroblast. J Med Genet 13:52-56. 7. Barile MF, Malizia WF, Riggs DB (1962): Incidence and detection of pleuro-pneumonialike organisms in cell culture by fluorescent antibody and culture procedures. J Bacterial 84:130-136. 8. Kato H, Sandberg AA (1977): The effect of sara on sister chromatid exchanges in vitro. Exp Cell Res 109:445-448. 9. Gardner EJ, Rogers SW, Woodward S (1982): Numerical and structural chromosome aberrations in cultured lymphocytes and cutaneous fibroblasts of patients with multiple adenomas of the colorectum. Cancer 49:1413-1419. 10. Chen TR (1981): Frequencies of sister chromatid exchanges in heteroploid cell lines of human melanoma origin. J Natl Cancer Inst 66:273-277. 11. Chaganti RSK, Schonberg S, German J (1974): A manyfold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc Natl Acad Sci USA 71:4508-4512. 12. Ho JHC (1975): Epidemiology of nasopharyngeal carcinoma. J R Coll Surg Edinburg 20:223-235. 13. Ryan TJ, Wells RS (1971): Hereditary benign telangiectasis. Trans St John's Dermatol Soc 57:148-156. 14. Boveri T (1914}: Zur Frage der Entstehung maligner Tumoren. Gustav Fischer Verlag, Jena. 15. Nowell PC (1976): The clonal evolution of tumor cell populations. Science 194:23-28. 16. Turleau C, Cabanis M-O, de Grouchy J (1980}: Augmentation des echanges de chromatides
Hyperdiploidy in Nasopharyngeal Cancer
115
dans les fibroblastes d'un enfant atteint de del (13] -retinoblastome. Ann Genet 23:169170. 17. Gardner EJ, Rogers SW, Woodward SR, Burt RW, Neff LK (1981): Diagnosis of familial polyposis coli, Gardner syndrome and hereditary discrete colorectal polyps. In: The Large Bowel Cancer Program: Its Achievments and Future Directions of Investigation. Program and Book of Abstracts of the 1981 Workshop on Large Bowel Cancer. University of Texas System Cancer Center, Houston, pp 67-71. 18. Helm S (1985): Sister chromatid exchange with and without in vitro mutagen induction in cultured skin fibroblasts from patients with adenomatosis of the colon and rectum. Clin Genet 27:51-58.
ABSTRACT: In this in vitro study of 28 members of a family with nasopharyngeal cancer (NPCJ and nasopharyngea! angiofibromas (NPA) in consecutive generations, in vitro hyperdiploidy, (WH) in dermal fbroblast cultures was associated with NPC in 2 affected members, NPA in 4 affected members, and 4 o f 16 members considered to be at increased genetic risk for these tumors. None of the six controls, all spouses, had IVH. As/VH has been shown to be associated with high specificity (~ 99%) with other heritable single tumors [3], IVH appeared to be a potential biomarker also for genetic predisposition for NPC. As NPC and NPA did not occur in a single affected family member, any association other than with IVH has not been established. IVH may be the in vitro expression of a gene/genes conveying tumor proneness per se, or it may determine tissue specificity of tumor pathology. As cellular hyperdiploidy has been considered to convey genomic instability [15], chromosome instability in numerically normal (diploid) and altered (tetraploidy and hyperdiploid) dermal fbroblasts from two NPC-affected members and two control spouses were evaluated by a parameter of chromosome stability-the frequency of sister chromatid exchanges (SCE) per cell. Irrespective of donor source, the diploid and tetraploid fbroblasts exhibited little evidence of chromosome instability, whereas there was a statistically significant difference in the mean frequency of SCE/cell between hyperdiploid and diploid cells in cultures derived from NPC-affected members (p < 0.01). INTRODUCTION The value of biomarkers for solid tumors is p r o b a b l y most justified in families w i t h the same t u m o r in consecutive generations. A l t h o u g h such families represent o n l y a m i n u t e fraction of the total clinical group for each specific tumor, identification of risk status w o u l d p r o v i d e (a) relevant information for family m e m b e r s a n d (1o) an o p p o r t u n i t y for evaluating the influence of e n v i r o n m e n t a l factors on cells w i t h a n d w i t h o u t genetic p r e d i s p o s i t i o n for a specific tumor. This w o u l d be particularly pertinent for a t u m o r a s s u m e d to result from such an interaction; such a group of tumors are the n a s o p h a r y n g e a l tumors (NPC). Through in vitro family studies on dermal m o n o l a y e r cultures d e r i v e d from skin biopsies, two in vitro n u m e r i c a l alterations in c h r o m o s o m e c o m p l e m e n t have been reported to s h o w an association w i t h the in r i v e expression of certain heritable tumors: increased t e t r a p l o i d y (in vitro tetraploidy, IVT) and h y p e r d i p l o i d y w i t h a normal occurrence of t e t r a p l o i d y (in vitro h y p e r d i p l o i d y , IVH). IVT has been identified w i t h h e r e d i t a r y colon cancer [1, 2], whereas IVI-I (defined as a m e t a p h a s e
From the Laboratoryfor Cell Biology,Department of Medicine, Cornell University Medical College,New York, IVY. Address requests for reprints to Dr. B. Shannon Danes, Laboratory for Cell Biology, Department of Medicine, Cornell University Medical College, New York, NY 10021. Received April 19, 1985; accepted July 3, 1985.
107 Science Publishing Co., Inc. 52 Vanderbilt Ave., New York, NY 10017
© 1986 Elsevier
Cancer Genet Cytogenet 21:107--115 0165-4608/86/$03.50
{1986)
108
B.S. Danes with more than 46 chromosomes exclusive of 92, tetraploidy), has been seen with cancer proneness in families with hereditary single tumors, including breast, melanoma, ovary, and pancreas [3, 4]. The purpose of this article is to demonstrate the potential usefulness of IVH for the identification of genetic predisposition for NPC, using a research family with such tumors in consecutive generations.
MATERIALS AND METHODS
Family Members Studied Subepidermal biopsies were obtained from 28 family members: 6 clinically affected (2 with nasopharyngeal cancer and 4 with nasopharngeal angiofibromas), 16 at risk for either tumor, and 6 spouses who were not at risk. The family had a history of nasopharyngeal cancer (NPC) in three consecutive generations, as well as nasopharyngeal angiofibromas (NPA) in four consecutive generations (Fig. 1). The histopathology was established for both tumors from tissue specimens taken from affected members in generation III (NPC III:3 and III:10; NPA II:8, III:13, IV:8, and V:3). The squamous cell carcinoma was classified as nonkeratinizing carcinoma by the WHO nomenclature:J5]. On the basis of the pedigree (Fig. 1), no association was established between the t w o t u m o r s , as the two lesions did not occur in the same affected family member:'AI1 28 family members completed the same questionnaire for family and personal history, including social habits.
Figure 1 Pedigree of a family with heritable nasopharyngeal cancer (histopathology, squamous cell nonkeratinizing carcinoma) in three consecutive generations and heritable nasopharyngeal fibromas in four consecutive generations.
4
1
[] [] m ['~
2
5
3
S
4
5
9
10
11
12
6
Unaffected with nasophoryngeal cancer or nosophoryngeol ongiofibromo Nosophoryngeol cancer by history (histopotholoQy not established) Squomous ceil corcinomo of nasopharynx (histopothology nonkerotinizin9 corcinon'm) Nosophoryngeal ongioflbromo in vitro hyperdiplotdy defined as a numerically abnormal metaphose containing more than 46 chromosomesexctu$ive of 92, tetroploidy []*Positive if % heter®lold metophoue (heteroploid metophoses over total number metophaess counted) greater than 2% (range in study 11-33%) [--~-Negative if % heteroploid metaphases less than 3% (range in study 0-2%)
"~[~ Deceased
)~ Proband
•
Hyperdiploidy in Nasopharyngeal Cancer
In Vitro Methodology
109
~
Dermal monolayer cultures were established from these split-thickness biopsies by standard culture methods [6]. Cultures were grown in plastic Petri dishes in Eagle's minimum essential medium (EMEM) with 15% (vol) fetal bovine serum (non-heattreated) in an atmosphere of 5% COz in air, pH 7.0-7.4. Cultures were routinely checked for bacterial contamination and by the fluorescent DNA staining procedure of Barile et al. [7] for mycoplasma and were found to be uncontaminated. Cells were grown in culture for 8-15 weeks (3-6 subcultures by trypsinization after the primary explant culture) before assays were performed. A coded number was assigned to each culture, so that its identity would not be known until completion of the assays. To ensure that cells to be assayed were in logarithmic growth, a semiconfluent monolayer culture (approximately 106 cells in a 26 cm 2 growth area] for each individual studied was trypsinized; this cell suspension was plated into a single 100cm z plastic Petri dish (initial cell density 104 cells/cm2). After 48-hr growth, this culture was trypsinized to obtain a cell suspension. Aliquots of this cell suspension were plated into 6 Petri dishes (100 cm z) at an initial cell concentration of 5 x 104 (initial cell density 0.5 × 10 s cells/cm z growth area). The medium was changed in all 6 subcultures after 12 hr. Twenty-four hours after this medium change, cell counts were done on 2 Of the 6 subcultures to determine cell number and density at assay time. The same batch of medium and serum was used in all in vitro studies involved in this in vitro research on this family [8].
Assay for Numerical Alteratiens in Chromosome Complement Twenty-four hr after the medium change, a burst of mitosis was invariably observed. At that time, chromosome preparations were made, without the addition of colchicine, from two of the duplicate subcultures and stained with acetoorcein [6]. For this study, tetraploidy was defined as a metaphase with 92 chromosomes, or ~>88, ~92, with a random loss of no more than 3 chromosomes. Hyperdiploidy was defined as a metaphase containing more than 46 chromosomes, exclusive of 92. The incidence of tetraploid or hyperdiploid metaphases was expressed as the number of tetraploid or hyperdiploid metaphases divided by the total number of metaphases scored on a slide, having at least 100 countable metaphases. Only if more than 100 mitoses were counted per slide was the preparation considered to reflect the mitotic activity of that culture. This study did not include evaluation of structural chromosome aberrations, as reported in in vitro studies on dermal fibroblasts from familial polyposis coli patients [2, 9].
Assay for Frequency of Sister Chromatid Exchanges To determine if the hyperdiploid cells represented a dividing subpopulation within a culture and to assess the extent of chromosome instability in both the diploid and numerically altered (tetraploid and hyperdiploid) cell populations, the frequency of spontaneous chromatid exchanges was determined for cultures derived from two N-PC-affected members and two spouse controls Two of the six subcultures established for assays were used for this purpose. After the culture medium was completely changed to medium containing 0.25 ~zg 5-bromo-2'deoxyuridine (BrdU) per milliliter of medium, the cultures were grown
110
B.S. Danes for two days in complete darkness. Each monolayer was then washed with unused medium and grown for 12 hr in medium without BrdU [10] before fixation for chromosome preparations. Slides were made according to a conventional airdrying method and were stained for SCE by the method described by Chaganti et a1.[11]. The analysis of the preparations was done without knowing the origin of the cultures. The SCE metaphases were identified (position on slide noted), photographed with a Zeiss photomicroscope, and 8 x 10 inch prints made. Counts of SCE were scored on photoprints at first and then checked again under the microscope by two independent observers. There was less than a 5% difference between the SCE counts of the two observers for any cell line studied. Statistical analysis between the frequency of SCE in diploid and hyperdiploid metaphases in cultures derived from NPC-affected family members utilized the Student's t-test for pairwise comparison between these two groups.
RESULTS Generation time for all dermal fibroblast monolayer cultures, irrespective of their human donor source, was 19-21 hr in the density range 0.4-1.8 × 103 cells/cm 2 growth area. All metaphase preparations were done on cultures during documented logarithmic growth in this cell density range. The frequency (%) of tetraploidy was constant for replicate cultures of sublines of a primary culture and showed little variation through six passages (Table 1). None of the cultures showed IVT (over 7% tetraploidy).
Table 1
Incidence (percent) of hyperdiploidy a in metaphase preparations derived from lines established from skin biopsies from two NPC-affected members and two spouse controls from the family studied Subculture number b 1
2
3
Donor source of c e l l l i n e
Ic
2c
Ic
2c
Ic
2c
19 20
20 19
20 25
o 0
o 0
o 1
NPC CLINICALLY AFFECTED Ill:3 III:10
20 22
19 25
23 24
LOW-RISK SPOUSE In:6 III:ll
o 1
o 1
o 1
• lypardiploidy defined as a metaphase w i t h more than 46 chromosomes exclusive of 92, tetraploidy. At least 100 metaphases were counted on a single lhetaphase preparation slide to determine hyperdiploi.dy incidence (percent]. bSubculture by trypsinization once a week after establishment of first subculture from primary biopsy culture ~ u m b e r s refer to replicate culture dishes
111
Hyperdiploidy in Nasopharyngeal Cancer
Frequency of V I I The frequency {%) of IVH in duplicate subcultures from the same culture, irrespective of donor source, showed little variation (Table 1}. The IVH findings were not influenced by the age of the skin donor or the time the cells were in culture (8-15 weeks in culture after the primary explant culture, 3-6 subcultures by trypsinization}.
Clinically affected members. IVH was observed in cultures established from the two family members with a clinical NPC history (111:3 and III:10). There was a wide distribution of the total number of chromosomes per IVH metaphase (72 to >200/ metaphase}. The four members studied with a clinical history of NPA (11:8, 1ii:13, IV:8, V:3) had IVH (11%-25%} that appeared to be identical to that occurring in NPC. The percent of IVH showed no correlation with the chronologic age of the NPA biopsy donor, as 11:8 was 74 years old at the time of biopsy and had 18% IVH, which is midrange for NPA-derived cultures (Table 2). IVH showed an association with both NPC and NPA, but neither tumor occurred in the same family member. In generation IV, the offspring (IV:8) of the NPC patient (II1:7} had IVH in association with NPA, as did his nephew {V:3). Based on pedigree data, family members studied were divided into two risk groups--high and low. Increased risk status was assigned to first-degree relatives of a clinically affected member and to their offspring; low-risk status was given to family members by marriage, i.e., spouses. Increased risk group. Four of the 16 clinically normal members studied who were considered to be at possible increased risk for genetic cancer proneness showed IVH {Fig. 1). Three were considered to be at 50% risk status (IV:7 and IV:9 were offspring of NPC-affected members and IV:IO of an NPA-affected member}. The fourth
Table 2 Family members studied: Pedigree, clinical status, and incidence of numerical alterations {hyperdiploidy and tetraploidy} in dermal flbroblast cultures derived from 28 family members Percent metaphases d
showing Clinical status
Number studied
M/F
Age (yr)e
Clinically affected
NPC ° NPA °
2 4
1:1 2:2
60,66 20,30,56,74
26,33 11,18,19,25
1,3 (0-2}
At increased
NL
12
7:5
(15-68)
(0-2)
(0-1)
genetic risk b L o w risk ~
NL NL
4 6
1:3 4:2
16,36,40,42 (38-68)
18,22,28,30 {0-1)
(0--4) (0-3)
Genetic status
Hyperdiploidye
Tetraploidye
°NPC, nasopharyngeal cancer, histopathology squamous cell nonkeratinizing carcinoma; NPA, nasopharyngeal angiofibromas. bAt increased genetic risk based on pedigree: offspring of at increased risk family members {clinically affected or normal}. CLew-risk
spouses of family members.
dHyperdiploidy is defined as a metaphase with more than 46 chromosomes exclusive of 92, tet~aploidy. Normal incidence of hyperdiploidy is 0 % - 1 % and for tetraploidy 0 % - 7 % in dermal fibroblast cultures [3].At least 100 metaphases were counted on a single chromosome preparation slide to determine the incidence of numerical alterations in chromosome complements. ~Numbers in parentheses represent age range.
112
B.S. Danes IVH-positive family member was V:5, the offspring of disease-free IvH-positive IV:7. Low-r/sk group. None of the six spouse controls studied had IVH (range 0%-1%). (Table 1).
Frequency of SCE The yield of metaphases in all cultures grown in BrdU-containing medium was greatly reduced to that in duplicate subcultures grown in medium without BrdU. In order to increase the number of metaphases present after growth in BrdU-containin 8 medium, each monolayer was washed with unused medium and then grown in medium without BrdU for ~12 hr prior to preparation of metaphase preparations [10]. Such post-BrdU control growth resulted in sufficient metaphases to give a slide preparation containin8! at least 100 metaphases. There was no difference in the mean frequency of SCE in diploid metaphases from cultures derived from NPC-affected members and normal spouse controls (1.33 - 1.02 SCE/cell for NPC versus 1.25 _ 1.49 for spouse controls). Nor was there any difference in their mean frequency in tetraploid metaphases from the same two family groups (1.50 _ 1.45 aCE/cell for NPC versus 2.2 - 1.15 SCE/cell for controis]. However, there was a statistically significant difference in the mean frequency of SCE between diploid and hyperdiploid metaphases in cultures from NPC-affected Individuals, (1.33 ± 1.02 SCE/cell for diploids versus 12.42 ± 8.34 SCE/cell for hyperdiploids t -- 3.47, p ~ 0.01) (Table 3).
DISCUSSION In the majority of NPC cases, there is no Clear genetic pattern, as is seen in the autosomal dominant heritable cancer syndromes. However, on the basis of pedigree (family and clinical) data, there are families with histologically confirmed NPC in
Table 3 Numerical chromosome distribution and frequency of sister chromatid exchanges (SCE) in metaphase preparations derived from lines established from skin biopsies from two NPC-affected members and two spouse controls Chromosome number/ metaphase
Donor source of cell line
Number of SCE metaphases analyzeda
Mean
Range
SCE/cell
Number of SCE
Mean ± SD
Range
298 32 18
12.42 ± 8.34 1.33 ± 1.02 c 1.50 ± 1.45
1-32 0-4 0-4
30 20
1.25 ± 1.49 2.22 ± 1.15
NPC CLINICALLY AFFECTED
(I]I:3 a n d III:lO)
H D T
25 25 10
72 46 90
62-105 45-46 b 89-92
LOW-RISK SPOUSES (III:6 a n d III:11)
D T
25 10
46 91
46 90-92
0-4 0-5
~rypes of metaphases: H, hyperdiploid; D, diploid; T, tetraploid. b4S count due to a presumed broken metaphase. CFrequency of SCE/cell in hyperdiploid metapheses versus frequency of SCE/cell in diploid metapheses from dermal flbroblast cultures derived from NPC-affected family members, p ~ 0.01.
Hyperdiploidy in Nasopharyngeal Cancer
113
consecutive generations or in familial clusters (several members in different generations with NPC without evidence of vertical transmission of its clinical occurrence, i.e., segregation of genetic predisposition for NPC). Such families have been identified in the southern Chinese population [12], and other families, although rare, do exist in other geographical and racial groups. The family described (Fig. 1) in this article fell into the latter category. NPC-affected family member I:1 (by family history without a tissue diagnosis) and his wife (I:2) emigrated from eastern Norway to the United States around 1890. This family met the criteria for a hereditary NPC tumor on the basis of seven clinically affected members in three consecutive generations and vertical transmission from a clinically affected member to an offspring (I:1-->II:1,3,6; II:l--~III:3; II:6-~,III:7). In this family, NPA, a recognized hereditary benign tumor [13], also met the same criteria for a hereditary basis, as there were four affected members in four generations (II:8,III:13, IV:8, V:3) and vertical transmission (II:8->III:13). However, NPC and NPA did not occur in any single family member, and IV:8 with NPA was the offspring of III:7 with NPC. NPA had an earlier age of onset and a broader age range (mean 45 years, range 20-74 years) than NPC (mean 64 years, range 60-66 years). On the basis of pedigree data (Fig. 1) the hereditary association between NPC and NPA could not be clarified. An association between IVH and several hereditary single tumors (breast, melanoma, ovary, and pancreas) has been demonstrated in human dermal fibroblast monolayer cultures derived from families with these heritable single tumors [3, 4]. With the recognition that this numerical alteration in chromosome complement occurred rarely in such cultures from the general population without a family history of such single tumors (specificity was high, approximately 99%-100%) [3, 4], IVH appeared to be a potential marker for genetic predisposition for such hereditary tumors. On the basis of the present family study (Tables 1 and 2), NPC has been added to this list. However, IVH was not gene-specific (the in vitro expression of cancer proneness for a single hereditary tumor), but appeared to be the in vitro expression of cancer proneness for a group of hereditary tumors. So far, it has not been established whether IVH is the in vitro expression of a gene/genes conveying cancer proneness per se, or one determining tissue specificity of cancer expression. Through this in vitro family study, an association between IVH and either NPC or NPA has been demonstrated (Fig. 1; Tables 1 and 2). Sufficient numbers have not been studied from families with NPC, NPA, or both to calculate sensitivity. The biologic significance of IVH has not been established, nor has any relevancy to cancer proneness. Since the mutational theory of cancer was proposed by Boveri in 1914 [14], the possible relationship of cancer to chromosomal rearrangements resulting from chromosome instability has been entertained. Cellular hyperdiploidy has been considered to convey genetic instability [15]. Chromosomal instability was reported [2, 9] for numerical and structural aberrations in cultured human dermal fibroblasts from patients with such genetic colon cancer syndromes as familial polyposis colt and related syndromes. On these bases, it appeared relevant to assay for such chromosomal instability in the present family. Chromosomal instability in numerically normal (diploid) and altered (tetraploid and hyperdiploid) fibroblasts have been evaluated by a parameter of chromosome stability--the frequency of SCE/cell. This assay has been used to detect chromosomal instability in such genetic cancer disorders as Bloom's syndrome [11] and del(13)-retinoblastoma [16]. Two previous in vitro studies [17, 18] failed to agree on whether or not there was a significant difference in the mean SCE frequencies occurring in dermal fibroblast cultures from individuals with and without heritable colon cancer. Irrespective of donor source, the diploid and tetraploid fihroblasts exhibited little evidence of chromosome instability based on the mean frequency of SCE/ceI1, al-
114
B . S . Danes
though there was a statistically significant difference in the m e a n frequency of SCE/ cell b e t w e e n h y p e r d i p l o i d a n d d i p l o i d cells in cultures derived from NPC-affected family m e m b e r s (p ~ 0.01]. There a p p e a r e d to be a s u h p o p u l a t i o n in such fibroblast cultures that was h y p e r d i p l o i d , w i t h an increased m e a n frequency of SCE/cell. As IVH has been s h o w n to be associated w i t h hereditary single tumors [3, 4], it is suggested that such tumors are associated w i t h an increased frequency of hyperd i p l o i d cells w i t h a genomic instability detectable by the SCE assay (Table 3). As all fibroblasts w i t h the p r e s u m e d germinal m u t a t i o n for cancer p r o n e n e s s d i d not show increased SCE frequency/cell (range 0-32, Table 3), this instability might have been c o n v e y e d s e c o n d a r i l y b y a n u m e r i c a l alteration in c h r o m o s o m e complement. However, as the m e a n frequency of SCE/cell was not increased in tetraploid cells, irrespective of source, mere n u m e r i c a l alteration in the c h r o m o s o m e c o m p l e m e n t from d i p l o i d y c o u l d not be responsible for this instability. The h y p e r d i p l o i d y state a p p e a r e d to convey genomic instability detectable as an increase in the m e a n frequency of SCE/ceI1. Supported by Grant 0053 from the Smokeless Tobacco Research Council.
REFERENCES
1. Danes BS (1981): Occurrence of in vitro tetraploidy in the heritable colon cancer syndromes. Cancer 48:1596-1601. 2. Delhanty JDA, Davis MB, Wood J (1983): Chromosome instability in lymphocytes, fibroblasts, and colon epithelial-like cells from patients with familial polyposis coli. Cancer Genet Cytogenet 8:27-50. 3. Danes BS (1984): In vitro expressions of cancer genes for heritable human tumors: Numerical alterations in chromosome complement. Dis Markers 2:371-380. 4. Danes BS, Lynch HT (1982): A familial aggregation of pancreatic cancer. JAMA 247:27982802. 5. Batsakis JG (1979): Tumors of the Head and Neck. Clinical and Pathological Considerations (ed. 2). Williams & Wilkins, Baltimore. 6. Danes BS (1976): The Gardner syndrome: Increased tetraploidy in cultured skin fibroblast. J Med Genet 13:52-56. 7. Barile MF, Malizia WF, Riggs DB (1962): Incidence and detection of pleuro-pneumonialike organisms in cell culture by fluorescent antibody and culture procedures. J Bacterial 84:130-136. 8. Kato H, Sandberg AA (1977): The effect of sara on sister chromatid exchanges in vitro. Exp Cell Res 109:445-448. 9. Gardner EJ, Rogers SW, Woodward S (1982): Numerical and structural chromosome aberrations in cultured lymphocytes and cutaneous fibroblasts of patients with multiple adenomas of the colorectum. Cancer 49:1413-1419. 10. Chen TR (1981): Frequencies of sister chromatid exchanges in heteroploid cell lines of human melanoma origin. J Natl Cancer Inst 66:273-277. 11. Chaganti RSK, Schonberg S, German J (1974): A manyfold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc Natl Acad Sci USA 71:4508-4512. 12. Ho JHC (1975): Epidemiology of nasopharyngeal carcinoma. J R Coll Surg Edinburg 20:223-235. 13. Ryan TJ, Wells RS (1971): Hereditary benign telangiectasis. Trans St John's Dermatol Soc 57:148-156. 14. Boveri T (1914}: Zur Frage der Entstehung maligner Tumoren. Gustav Fischer Verlag, Jena. 15. Nowell PC (1976): The clonal evolution of tumor cell populations. Science 194:23-28. 16. Turleau C, Cabanis M-O, de Grouchy J (1980}: Augmentation des echanges de chromatides
Hyperdiploidy in Nasopharyngeal Cancer
115
dans les fibroblastes d'un enfant atteint de del (13] -retinoblastome. Ann Genet 23:169170. 17. Gardner EJ, Rogers SW, Woodward SR, Burt RW, Neff LK (1981): Diagnosis of familial polyposis coli, Gardner syndrome and hereditary discrete colorectal polyps. In: The Large Bowel Cancer Program: Its Achievments and Future Directions of Investigation. Program and Book of Abstracts of the 1981 Workshop on Large Bowel Cancer. University of Texas System Cancer Center, Houston, pp 67-71. 18. Helm S (1985): Sister chromatid exchange with and without in vitro mutagen induction in cultured skin fibroblasts from patients with adenomatosis of the colon and rectum. Clin Genet 27:51-58.