differentiation pathways in primary invasive breast carcinoma as suggested by intermediate filament...

JOURNAL OF PATHOLOGY, VOL. 179: 386-391 (1996)

DIFFERENTIATION PATHWAYS IN PRIMARY INVASIVE BREAST CARCINOMA AS SUGGESTED BY

INTERMEDIATE FILAMENT AND BIOPATHOLOGICAL MARKER EXPRESSION

DONATELLA SANTINI*, CLAUD10 CECCARELLI?, MARIO TAFFURELLII, STEFAN0 PILERI” AND DOMENICO MARRAN01

*I! Servizio rli Anatoniia ed Istologia Patologica and ?Laboratorio di Irnrnunocitopatologia Oncologica, Istituto di Anatornia ed Istologiu Patobgicu, Universita di Bologna, Italy; 2 Clinira Chirurgira I, Universitu di Bolognu, Ifaly

SUMMARY

The expression of intermediate filament proteins (IFPs) in 65 primary breast carcinomas was analysed by a panel of specific antibodies. Results were integrated with the oestrogen and progesterone receptor (ER and PGR) status, Ki-67 marking, and epidermal growth factor receptor (EGFr) expression. Invasive breast carcinomas could he divided into three main groups: group 1 revealed positivity only for ‘simple epithelial’ cytokeratins (CKs 7, 8, 18, and 19); group 2 also stained with the antibodies K8.12 and 34m12; while group 3 showed co-expression of CKs 14 and 17, vimentin, and a-smooth muscle actin. Group 3 consistently comprised tumours with the highest Ki-67 levels, EGFr positivity, and ER-PGR negative status. On the other hand, groups I and 2 usually exhibited a positive hormonal status, lower proliferative activity, and EGFr negativity. The results of this study indicate that the determination of IFPs can significantly contribute to the identification of groups of patients with different biopathological settings and possibly different clinical behaviour.

KEY woKDs-intermediate filaments; breast cancer; epidermal growth factor receptor; immunohistochemistry

INTRODUCTION

The expression and distribution of intermediate filament proteins ( I FPs) reflect the histogenesis, differentiation, and functional status of normal and pathological tissue.‘ Experimental studies suggest that the cell-specific IFP pattern may vary depending on different biological conditions, such as growth regulation, cell-cycle status, and hormonal environment.h l 3 Pre- vious studies have demonstrated a specific pattern of IFP expression i n the luminal (cytokeratins 7, 8, 18, and 19) and myoepithelial (cytokeratins 5, 14, and 17; vimentin and w-smooth muscle actin) cells of normal human mammary gland.14,15 Although several studies on IFP expression and distribution in pathological breast tissue have been published,Ih 24 the complex relationships between the cytoskeletal proteir, composition and the main biopathological prognostic markers of breast cancer have not been explored in detail. Only vimentin expression has been found to be correlated with oestrogen receptor (ER)-negative, epidermal growth factor receptor (EGFr)-positive breast cancers; with high Ki-67 score, and with higher histological

The aim of the present study was to assess the IFP patterns in breast cancer by means of a large panel of monoclonal antibodies, comparing these patterns with tumour morphology and with the main biological prognostic indicators.

28

Addressee for correspondence: Donatella Santini, MD, I1 Servizio di Anatoniia Patologica, Policlinico S. Orsola, Via Massarenti 9,40138 Bologna, Italy.

CCC 0022-34 1 7/96/080386-06 ~ “ 8 1996 by John Wiley & Sons, Ltd.

MATERIALS AND METHODS

Patients and diagnosis

Sixty-five primary invasive breast carcinomas were included in this study. The tumours were staged following the UICC-TNM system and histologically classified according to the World Health Organization. Ductal not otherwise specified (NOS) invasive carcinomas were histologically graded (G) using Elston and Ellis’s cri- teria.29 All the neoplasms were also typed by nuclear grading (NG).

Immunohistochemical method

Sections cut from metacarnoy-fixed, paraffin- embedded tissue blocks were allowed to dry at 37°C overnight and then processed for immunohistology according to a modified immunoperoxidase ABC method, improving CK single isoform determination by appropriate cocktails of different monoclonal antibodies. 3o

To evaluate each IFP-immunopositive determination, samples were scored using a six-level ordinal scale as follows: O=negative; I=less than 5 per cent of tumour cell positive; 2 = 5-20 per cent of positive cells; 3 = 2 1 -50 per cent of positive cells; 4 ~ 5 1 - 8 0 per cent of positive cells; 5=more than 80 per cent of positive cells. EGFr expression was semi-quantitatively evaluated as follows: O=negative; 1 =less than 5 per cent of turnour cells positive; 2=more than 5 per cent of tumour cells positive.

Received 4 August 1995 Accepted 19 February 1996

lFPS AND BIOLOGICAL MARKERS IN BREAST CANCER

Table I-List of the monoclonal antibodies used

387

Antibody clone

6B10 215B8

RCKlO5

35PH11 a4.1.17

OV-TL12/30

DE-KIO

1 C7 2D7

LL002

E3

CY 90 KS-B 17.2

BAI 7 Ks19.1 LP2K

Ks13.1

K8.12

34PE12

v 9

1 A4

31G7

Specificity Source Dilution

CK 4 CK 4

CK 7 CK 7

CK 8 CK 8

CK 10

CK 13 CK 13

CK 14

CK 17

CK 18 CK 18

CK 19 CK 19 CK 19

CKs l3, 14, 16

CKs 13, 15 16 Cks 1, 5, 10, 14

Vimentin

a-SMA

EGFr

Sigma lmmunochemicals (U.S.A.) Boehringer Mannheim (Germany)

Sanbio BV (The Netherlands) BioGenex Laboratories (U.S.A.)

Dakopatts a/s (Denmark) Dr M. Osborn (Germany)

Euro-Diagnostics BV (The Netherlands)

Euro-Diagnostics BV (The Netherlands) Euro-Diagnostics BV (The Netherlands)

BioGenex Laboratories (U.S.A.)

Progen (Germany)

Sigma Immunochemicals (U.S.A.) Sigma lmmunochemicals (U.S.A.)

Dakopatts a/s (Denmark) Progen (Germany) Amersham (U.K.)

Progen (Germany)

Miles Scientific (U.S.A.)

Dakopatts a/s (Denmark)

Dakopatts a/s (Denmark)

Sigma Immunochemicals (U.S.A.)

Triton Diagnostics (U.S.A.)

1:15 000 1:500

1:450 1:lOO

1:300 1 :4000

I :300

1 :600 1:600

1 :250

1 :400

1 :50 000 1 :8000

1 :2300 1:1300 1 :50

1:1200

1:lOOO

1:300

1 :700

1:7000

1:lOO

- 13=mainly CK 13. - 13, @=mainly CKs 13 and 16

Determination of steroid hovmone receptors and growth fraction

The oestrogen receptor (ER) and progesterone receptor (PGR) status was determined on cryostat sections using immunocytochemical assay (ICA) Abbott kit (Abbot, North Chicago, IL, U.S.A.). Growth fraction was determined on two acetone-fixed, chloroform- treated frozen sections, immunostained using an anti- Ki-67 monoclonal antibody (MAb) (Dakopatts ah, Glostrup, Denmark), followed by an ABC technique. The nuclear immunostaining of ER/PGR-ICA and Ki-67 was quantified by image cytometry with the ‘Cytometrica’ software (C & V, Bologna, Italy), as previously rep~r ted .~’ ER and PGR status was considered positive if the percentage of immunopositive cells exceeded 5 per cent of the evaluated neoplastic population.

statistical analyses

Statistical methods chosen for the evaluation of the results depended on the nature of the variables studied. Monotonicity between two naturally ordered variables was estimated by the gamma coefficient. A linear trend between a naturally ordered variable and a continuous

variable was explored by means of the Kruskal-Wallis test. The association among the IFP immunoprofile groups and histological grade, TNM, G, and NG was evaluated using the chi square test. Only statistically significant results are reported in the text.

RESULTS

Histopathology Forty-five out of 65 primary invasive breast tumours

were diagnosed as ductal NOS carcinoma, ten as lobular, eight as mixed ductal-lobular, one as medullary, and one as metaplastic carcinoma. Ductal NOS carcinomas were graded as follows: two were G I, 24 G 11, and 18 G 111. NG gave rise to the following results: 11 cases were NG I, 22 NG 11, and 32 NG 111. Axillary lymph-node metastases (N+) were present in 36 cases.

Intermediate filament protein expression and distribution

‘Simple epithelial’ cytokeratins (CKs) 8, 18, and 19 were found in all cases. CK 7 was detected in 57165 cases

D. SANTINI ET A L . 388

Table II-ERIPGR, Ki-67, EGFr, and IFP immunoprofile

ER and PGR Ki-67 EGFr Vim a-SMA CK 14 CK 17 C K 7 CKs 8, 18, 19 K8.12 34pE12

Group I Positive cases 6/8+and+ 8/8 0/8 0/8 0/8 0/8 0/8 6/8 8/8 018 0/8

218 + and - Mean values 43.8% and 11.3%~ 13.30% O* O* O* O* O* 1.4* 5* O* O*

Group 2 Positive cases 28/38 +and + 38/38 4/38 1/38 1/38 0138 0/38 33/38 38/38 33/38 29/38

6/38 +and - 2.2* 2.5* Mean values 57.9% and 29.9% 19.8% 0.2* 0.1* 0.1* O* O* 3.2* 5*

Group 3 Positive cases 0/19+and+ 19/19 18/19 18/19 11/19 14/19 15/19 18/19 19/19 18/19 18/19 Mean values 0% and 0% 47.5% 1.8* 3.1* 1.8* 1.7* 1.5* 3.4* 4.7* 2.8* 3.2*

ER/PGR+ =more than 5 per cent of positive neoplastic cells. Ki-67=percentage of positive neoplastic cells. *=Mean of scoring values.

displaying a quantitative variability. CKs 14 and 17 were detected in 14 and 15 cases, respectively. A significant relationship was found between CK 14 and 17 expression by means of the gamma correlation test ( y value= 0.904620, P<O.OOOl). The intensity of the staining with anti-CK 14 often exceeded that obtained with the MAb against CK 17. The antibody K8.12 (CKs 13, 15, and 16) was positive in 51/65 cases. Immunoreac- tivity for the MAb 34pE12 (CKs 1, 5 , 10, and 14) was found in 46 samples. Since the antibody Ks 13.1 (mainly CK 13) as well as those against CK 4, 10, and 13 were negative, they were neither reported nor further discussed.

Vimentin was expressed in 18 cases. Intense and diffuse a-smooth muscle (a-SM) actin positivity was detected in those cases (12/65) showing CK 14 and/or CK 17 staining. A11 cases but one, which presented vimentin and/or a-SM actin positivity, were characterized by a complex CK composition, co-expressing also ‘myoepithelial’ CKs. Conversely, the samples exhibiting a simpler CK pattern consistently lacked both vimentin and a-SM actin immunolocalization.

By combining our data, the 65 cases could be divided into at least three main groups (Table 11): group 1 (eight cases), which showed only ‘simple epithelial’ CK expression (CKs 7,8,18, and 19) (Figs 1A and 1 B); group 2 (38 cases), which besides primary ‘simple-epithelial’-type CKs, reacted with the antibodies K8.12 and 34pE12 (Figs 2A and 2B); and group 3 (19 cases), which in addition expressed the ‘myoepithelia1’-type CKs 14 and/or 17, as well as vimentin and/or a-SM actin (Figs 3A, 3B, 3C, and 3D).

Histologically, all group 3 cases but one (diagnosed as metaplastic carcinoma) were invasive ductal carcinomas. No significant correlation was observed between the three groups and histological type, grade, or stage. On the contrary, a statistically significant correlation was found, by means of the chi-square test, between the three groups and NG (chi square value 23.914; P<O.OOOl) .

Correlation of hormone receptor status, Ki-67, EGFr, and IFP immunoprofiles

ER/PGR, Ki-67, and EGFr expression is reported in Table 11. Group 3 cases were characterized by an inverse relationship with ER/PGR immunostaining, which turned out to be consistently negative. Conversely, in group 1, 75 per cent of the cases presented a high ER/PGR content, the remainder being ER-positive/ PGR-negative. Finally, group 2 presented a more heterogeneous hormonal status: 65-7 per cent of the cases were ER-/PGR-positive, 23.6 per cent ER-positive/ PGR-negative, and 10.5 per cent ER-/PGR-negative.

The Ki-67 scores showed a similar distribution: they were low and moderate in groups 1 and 2, respectively, the highest values being observed in group 3. No significant association was found between Ki-67 values and single CK expression, including the specific immunoreactivity for antibody K8.12, as shown by the Kruskal- Wallis test ( H valuez3.698, P=0.594). On the other hand, by using the same statistical analysis, a strong correlation was seen between Ki-67 marking and the three groups mentioned above (Kruskal-Wallis test: H value= 30.45 1, P<O.OOOl) .

EGFr positivity, in the form of membrane-bound staining, was observed in 21 cases. Interestingly, all group 3 cases were EGFr-positive, while the other groups were essentially negative.

DISCUSSION

The present study shows a high degree of complexity in intermediate filament protein (IFP) expression in invasive human breast carcinoma which appears specifi- cally related to different biopathological patterns. Despite the complexity of IFP expression in each indi- vidual tumour, three major immunoprofiles can be identified: group 1, characterized by the expression of

IFPS AND BIOLOGICAL MARKERS IN BREAST CANCER 389

Fig. l-Invasive breast carcinoma of group 1 showing CK 19 (A) and CK 7 (B) expression. (Immunoperoxidase method)

Fig. 2-Invasive breast carcinoma of group 2 revealing positivity for antibodies K8.12 (A) and 34pE12 (B). (Immunoperoxidase method)

‘simple epithelial’-type CKs (7, 8, 18, and 19); group 2, showing in addition to these CKs a heterogeneous combination of other isoforms, as revealed by the antibodies K8.12 and 34pE12; and finally group 3, displaying the most complex immunophenotype, due to the co-expression of all four ‘simple epithelial’ CKs, ‘basal cell-myoepithelia1’-type CKs, vimentin, and a-SM actin, along with heterogeneous immunostaining with the MAbs K8.12 and 34pE12.

The presence of immunopositivity with the antibody K8.12 (mainly specific to CKs 13 and 16) together with the absence of CK 13 seems to confirm the presence of CK 16 in some breast carcinoma^.^^,^^ This CK, with its CK 6 counterpart, was indicated as ‘hyperproliferation- related’.33-36 We were unable to find a significant association between Ki-67 marking and immunostaining with the K8.12 antibody. This observation is in keeping with the report of Wetzels et of a low con- cordance between Ki-67 immunoreactivity and the hyperproliferation-related CK 16.

When compared with previous reports, 17-22,24,37,38 the CK expression patterns of the present study show a higher incidence of immunopositive cell populations with each specific antibody. However, our findings parallel those of Gould et u Z . ~ ~ and Wetzels et U Z . , ~ ~ with

special reference to ‘myoepithelia1’-type CK expression and the results with the K8.12 antibody. The broader spectrum of immunoreactivity in our study may be explained by the use of metacarnoy fixation, together with the simultaneous application of different isoform- specific monoclonal antibodies.

The analysis of CK composition has led to the identification of two main groups of invasive breast carcinoma, with and without ‘myoepithelial’ differentiation. 17-21,39 Our results expand the spectrum of our knowledge in this field showing that, irrespectively of the CK pattern, the presence of ‘rnyoepithelia1’-type CKs 14 and 17 paralleled a-SM actin and vimentin immunoreactivity. This raises the intriguing question of whether the different steps in IFP organization correspond to a differentiation-related phenomenon. When referred to the three main identified immunoprofiles, CK hetero- geneity and complexity seem to correlate with the acqui- sition of vimentin expression, which is usually indicative of dedifferentiation rather than maturation. In keeping with this hypothesis, there are some reports on breast cancer cell line experiments which demonstrate that vimentin expression marks the progression from hormone dependence to hormone independence, and that vimentin mRNA levels are g r o w t h - r e g ~ l a t e d . ~ , ~ ~ ~ ~ These

390 D. SANTlNl ET A L .

Fig. 3--Invasive breast carcinoma of group 3 characterized by co-expression of CK 14 (A) and CK 17 (B) and showing in addition ditruse immunostaining for vimentin (C) and a-smooth muscle actin (D). (Immunoperoxidase method)

in vitro observations support the concept that vimentin expression is preferentially detected in ER-negative, EGFr-positive human breast carcinomas25 and is positively correlated with high histological grade and growth fraction.26-28

In our series, we found a strong relationship among group 3 cases, tumour G, NG, EGFr expression, ER ind PGR negativity, and high Ki-67 score. This specific immunoprofile corresponded to a group of NOS ductal carcinomas which could not be distinguished morpho- logically from the other neoplasms. Thus, the analysis of IFP expression seems to be more powerful than conven- tional light microscopy in assessing the differentiation status of mammary carcinoma.

Our data, along with the results of previous studies,” 28 strengthen the concept that biopathological factors commonly applied to breast cancer characterization, such as hormonal status and proliferative activity, should not be regarded as absolute parameters, but should be considered in the wider scenario of cytoskeletal protein organization. In particular, the increase in the complexity of IFP expression, which correlates inversely with ER/PGR content and culminates in our group 3 cases (Table II), suggests that in breast cancer, cytoskeletal organization might also be driven by hormonally-mediated stimuli, possibly under the control of growth factors.?’

ACKNOWLEDGEMENTS

This study was supported by grants from MURST 60% funds. The skilful technical help of Ms Milena Pariali and Ms Michela Gamberini was greatly appreci- ated. Many thanks also to Mr Alessandro Busi for photographic work.

REFERENCES 1. Franke WW, Schmid E, Schiller DL, rt a/. DitTerentiation-related patterns

of expression of proteins of intermediate-sized filaments in tissues and cultured cells. Cold Spring Horb Synip Qucint Biol 1982; 46: 431453.

2. Osborn M , Weber K. Tumor diagnosis by intermediate filament typing: a novel tool for surgical pathology. Lub Invest 1983; 4 8 372-388.

3. Fuchs E, Tyner AL, Giudice GJ, rr a[. The human keratin genes and their differential expression. Curr Top Dev B i d 1987; 22: 5-34.

4. Nagle RB. Intermediate filaments: a review of the basic biology. Artt J Surg Patho/ 1988; 12(Suppl 1): &16.

5. Alberts K. Fuchs E. The molecular biology of intermediate filaments. In/ Rev Cytol 1992; 134 243-219.

6. Connel ND, Rheinwald JG. Regulation of the cytoskeleton in mesothelial cells: reversible loss of keratin and increase in vimentin during rapid growth in culture. Cell 1983; 34 245-253.

7. Schmid E, Schiller DL, Grund C. Stadler J, Franke WW. Tissue type- specific expression of intermediate filament proteins in a cultured epithelial cell line from bovine mammary gland. J C ~ l l B id 1983; 96: 37-50.

8. Ben-Ze’ev A. Differential control of cytokeratins and vimentin synthesis by celkel l contact and cell spreading in cultured epithelial cells. J Cell R i d 1984; 9 9 1424-1433.

9. Ben-Ze’ev A. Cellkell interaction and cell configuration related control of cytokeratins and vimentin expression in epithelial cells and in fibroblasts. In: Wang E, er al. eds. Intermediate Filaments. Ann N Y Amd .W 1985; 445: 597.

IFPS AND BIOLOGICAL MARKERS IN BREAST CANCER 391

10. Rheinwald JG, O’Connell TM, Connell ND, et al. Expression of specific keratin subsets and vimentin in normal human epithelial cells: a function of cell type and conditions of a growth during serial culture. Cancer CeLs 1986; I: 217.

I I . Agnor C, Walker-Jones D, Valverins E, ef a/. Vimentin expression charac- terizes hormone-independent breast cancer cell lines and mammary epithelial cells transformed by two oncogenes. Proc A m Assoc Cuncer Res 1987; 28: I I (Abstract 41).

12. Fuchs E. Epidermal differentiation. Curr Op Cell Biol 1990; 2 1028-1035. 13. Moll R, Franke WW, Schiller DL, Geiger 8, Krepler R. The catalog of

human cytokeratins: patterns of expression in normal epithelia, tumors, and cultured cells. Cell 1982: 31: 11-24,

14. Taylor-Papadimitriou J. Lane EB. Keratin expression in the mammary gland. In: Neville MC, Daniels C, eds. The Mammary Gland: Development, Regulation, and Function. New York: Plenum Press, 1987; 181-215.

15. Bocker W, Bier B, Freytag G, et al. An immunohistochemical study of the breast using antibodies to basal and luininal keratins, alpha-smooth muscle actin, vimentin, collagen IV and laminin. Part 1: normal breast and benign proliferative lesions. Virchows Arch A [Pathol Anar] 1992; 421: 31 5-322.

16. Altmannsberger M, Osborn M, Holscher A, Schauer A, Weber K. The distribution of keratin type intermediate filaments in human breast cancer. Virrhows Arch B [Cell Pathol] 1981; 31: 217-284.

17. Altmannsberger M, Dirk T, Droese M, Weber K, Osborn M. Keratin polypeptide distribution in benign and malignant breast tumors: subdivision of ductal carcinomas using monoclonal antibodies. Virchows Arch B [ Cell Puthol] 1986; 51: 265-275.

18. Nagle RB, Bocker W, Davis JR, et al. Characterization of breast carcinomas by two monoclonal antibodies distinguishing myoepithelial from luminal epithelial cells. J Hislochem Cytocheni 1986; 34: 869-881.

19. Dairkee SH, Puett L, Hackett AJ. Expression of basal and luminal epithelium-specific keratins in normal, benign, and malignant breast tissue. J Nut1 Cancer Inst 1988; 8 0 691-695.

20. Guelstein VI, Tchypysheva TA. Ermilova VD, Litvinova LV, Troyanovsky SM, Bannikov GA. Monoclonal antibody mapping of keratins 8 and 17 and of vimentin in normal human mammary gland, benign tumors, dysplasias and breast cancer. Int J Cuncer 1988; 4 2 147-153.

21. Jarasch E-D, Nagle RB, Kaufman M, Maurer C, Bocker WJ. Differential diagnosis of benign epithelial proliferations and carcinomas of the breast using antibodies to cytokeratins. Hum Pathol 1988; 19: 276289.

22. Pellegrino MB, Asch BB, Connolly JL, Asch HL. DiRerential expression of keratins 13 and 16 in normal epithelium, benign lesions, and ductal carcinomas of the human breast determined by the monoclonal antibody Ks8.12. Cancer Res 1988; 48: 5831-5836.

23. Gould VE, Koukoulis GK, Jansson DS, Nagle RB, Frdnke WW, Moll R. Coexpression patterns of vimentin and glial filament protein with cytokeratins in the normal, hyperplastic, and neoplastic breast. A m J Pathol 1990; 137: 1143-1 155.

24. Heatley M, Maxwell P, Whiteside C, Toner P. Cytokeratin intermediate filament expression in benign and malignant breast disease. J Clin Pafhol 1995; 4 8 2632.

25. Cattoretti G, Andreola S, Clemente C, D’Amato L, Rilke F. Vimentin and p53 expression on epidermal growth factor receptor-positive. oestrogen receptor-negative breast carcinomas. Br J Cancer 1988: 57: 353-357.

26. Raymond WA, Leong AS-Y. Coexpression of cytokeratin and vimentin intermediate filament proteins in benign and malignant breast epithelium. J Palliol 1989; 157: 299-306.

27. Domagala W, Lasota J , Bartkowiak J, Weber K, Osborn M. Vimentin is preferentially expressed in human breast carcinomas with low oestrogen receptor and high Ki-67 growth fraction. A m J Pathol 1990; 136: 219-227.

28. Domagala W, Wozniak L, Lasota J, Weber K, Osborn M. Vimentin is preferentially expressed in high-grade ductal and medullary, but not in lobular breast carcinomas. A m J Pathol 1990: 137: 1059-1064.

29. Elston CW, Ellis 10. Pathological prognostic factors in breast cancer. 1. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histoputhology 1991; 1 9 403410.

30. Santini D, Ceccarelli C, Mazzoleni G, Pasquinelli G, Jasonni VM, Martinelli GN. Demonstration of cytokeratin intermediate filaments in oocytes of the developing and adult human ovary. Hirtachemistry 1993; 9 9 311-319.

31. Ceccarelli C, Santini D, Chieco P, Taffurelli M, Marrano D, Mancini AM. Multiple expression patterns of biopathological markers in primary invasive breast carcinoma: a useful tool for elucidating its biological behaviour. Ann Oncol 1995; 6 275-282.

32. Wetzels RHW, Kuijpers HJH, Lane EB, er al. Basal cell-specific and hyperproliferation-related keratins in human breast cancer. A m J Pathol 1991; 138 751-763.

33. Weiss RA, Eichner RA, Sun T-T. Monoclonal antibody analysis of keratin expression in epidermal diseases: a 48- and 56-kilodalton keratin as molecular markers for hyperproliferative kerdtinocytes. J Cell B i d 1984; 98: 1397-1406.

34. Quinlan RA, Schiller DL, Hatzfeld M, et al. Patterns of expression and organization of cytokeratin intermediate filaments. Ann N Y Acad Sci 1985; 455 282-306.

35. Tyner AL, Fuchs E. Evidence for post-transcriptional regulation of the keratins expressed during hyperproliferation and malignant transformation in human epidermis. J Cell Biol 1986; 103: 1945-1955.

36. Galvin S, Loomis C, Manabe M, Dhouailly D, Sun T-T. The major pathways of keratinocyte differentiation as defined by keratin expression: an overview. Adv Dermatol 1989; 4: 277-300.

37. Gusterson BA, Warburton MJ, Mitchell D, Ellison M, Neville AM, Rudland PS. Distribution of myoepithelial cells and basement membrane proteins in the normal breast and in benign and malignant breast diseases. Cuncer Re5 1982; 4 2 41634770.

38. Tsubura A, Okada H, Senzaki H, Hatano T, Morii S. Keratin expression in the normal breast and in breast carcinoma. Histopatholr/gy 1991; 1 8 71 6-724.

39. Wetzels RHW, Holland R, van Haelst UJGM, Lane EB, Leigh IM, Ramaekers FCS. Detection of basement membrane components and basal cell keratin 14 in noninvasive and invasive carcinomas of the breast. A m J Puthol 1989; 134 571-579.

40. Ferrari S, Bettini R, Kaczmarek L, ef a/. Coding sequence and growth regulation of the human vimentin gene. Mol Cell Bid 1986; 6: 3 6 1 6 3620.

differentiation pathways in primary invasive breast carcinoma as suggested by intermediate filament...

Documents