The Prostate 20:29-41 (1992)
Effects of Extracellular Matrix Components and Dihydrotestosterone on the Structure and Function of Human Prostate Cancer Cells Brian C. Murphy, Kenneth J. Pienta, and Donald S. Coffey The Johns Hopkins Oncology Center, and The Brady Urological Research Institute, The Johns Hopkins School of Medicine, Baltimore, Maryland The extracellular matrix (ECM) has been shown to play a major role in cell structure and function. Several studies have demonstrated that the ECM can alter cell morphology and effect DNA synthesis and gene expression. The ECM also interacts with growth hormones which have been shown to be located in or near the ECM where they are believed to effect cell structure and function. In the nontransformed cell, these ECM and hormone-mediated effects appear to be tightly regulated and this is believed to be accomplished through cell receptor-tissue matrix interactions. We, therefore, undertook a study to determine the effects of a variety of ECM components and the androgenic hormone dihydrotestosterone (DHT) on the structure and function of the human prostate cancer cell line, LNCaP. The effects of individual matrix components in the presence and absence of 1 nM DHT on the static and dynamic morphology, growth rate, and PSA production of the LNCaP cell line were studied. We determined that the ECM and DHT interact in complex ways to effect cell structure and function. DHT produced alterations in cytoplasmic structure that increased cell size and decreased the nuclear aredcytoplasmic area ratio. Dynamic cell structure as measured by cell motility was very sensitive to the ECM components and the presence of DHT. PSA and growth could be regulated by substratum and DHT and there was an inverse relationship between PSA production and growth rate. These data exemplify the complex interactions which occur between prostate cancer cells, ECM components, and exogenous DHT that are reflected in cell structure and function.
Key words: basement membrane, LNCaP cells, androgen effects, PSA production
INTRODUCTION
What a cell touches has a major role in determining what a cell does. In classic studies, Moscona and Folkman controlled DNA synthesis in endothelial cells by manipulating cell shape by changing the substratum on which the cells were plated and found that DNA synthesis could be shut off by induced rounding up of the cell [ 11. Furthermore, Gospodarowicz et al. demonstrated that the regulation of corneal epithelial cell shape by different substrata directed the cells’ ability to respond to mitogenic hormones [2]. In addition to this role in the control of cell shape, the
Received for publication April 30, 1991; accepted June 26, 1991.
Dr. Kenneth J. Pienta’s current address is Wayne State University School of Medicine, Division of Hematology/Oncology, P.O. Box 02188, Detroit, MI 48202-0188. Address reprint requests there. 0 1992 Wiley-Liss, Inc.
30
Murphyet al.
substratum or extracellular matrix (ECM) has been shown in many studies to play an important part in the control of cell differentiation and function. Cunha and colleagues have demonstrated that the ECM can be responsible for functional differentiation in development when they showed that urogenital sinus mesenchyme induces adult urinary bladder epithelial cells to remodel to form prostatic epithelial cells [3]. Reddi and Anderson have observed that mature fibroblasts undergo further differentiation to form new chondroblasts and chondrocytes when they were exposed to contact with a demineralized bone collagen matrix and osteogenic factors [4]. These studies as well as others demonstrate that altering cell morphology alters DNA synthesis and gene expression [5-91. The ECM itself also interacts with hormones to effect cell structure and function. Several types of hormones have been shown to be located or sequestered in the ECM where they are believed to effect gene expression. Lactogenic hormones located in the basement membrane have been shown to alter cell shape and cause changes in mRNA stability and its association with the cytoskeleton [ 101. Bissell and colleagues demonstrated that when mammary epithelial cells were placed on a biomatrix with mammary basement membrane, their expression of casein and lactalbumin increased as much as 160-fold when compared to cells grown on tissue culture plastic [ 10,l I]. Reid and her colleagues demonstrated that normal rat hepatocytes required both a hormone-supplemented media and a specific biomatrix to synthesize and stabilize mRNA-neither alone was sufficient [ 121. Therefore, the extracellular matrix has been shown to alter the hormonal responsiveness, cell shape, and gene expression of cells in a specific manner. This coordination of cell function by the ECM may have added significance in the study of cell transformation, and recently, it has been shown that ECM components are altered in transformed rat prostate fibroblasts [ 131. It is now evident that not only are individual extracellular matrix components important but that the interaction of several of the components is involved in regulative hormonal responses and in cell signalling and coordinating normal cell function. It has been postulated that this coordination of cellular function is accomplished by the tissue matrix [14-161. The tissue matrix is the three-dimensional structural framework of the cell which consists of the interacting dynamic skeletons of the nuclear matrix, the cytoskeleton, and the extracellular matrix. This dynamic skeletal system physically connects the extracellular matrix to the nucleus and the DNA, is tissue specific, and may be altered in the cancer cell [15-171. Therefore, we undertook a study to determine if cell structure as dictated by the extracellular matrix components and hormones correlated with alterations in cell function of the human prostate cancer, LNCaP. Specifically, the effects of the individual matrix components collagen I, collagen IV, laminin, and fibronectin, and the complete basement membrane Matrigel, on the morphology, motility, growth rate, and PSA production of the LNCaP cell line were studied. We then investigated the effects of these ECM subcomponents on the structural and functional parameters in the presence of DHT,a steroid which has been demonstrated to effect prostate cell function. MATERIALS AND METHODS Materials BioCoat Matrix culture dishes (35 mm) precoated with fibronectin, type I collagen, and laminin were obtained from Collaborative Research (Bedford, MA). The
ECM and DHT Effects on LNCaP Cells
31
extracellular matrix products human fibronectin, murine laminin, rat tail type I collagen, murine type IV collagen, and Matrigel, as well as dispase were also obtained from Collaborative Research. Lab-Tek single-well chamber slides (2 cm X 4.2 cm) and double-well chamber slides (2 cm X 2 cm) were obtained from Nunc Inc. (Naperville, IL). Preparation of substrata for the motility, growth, and PSA assays was as follows: type IV collagen 15-20 pg/cm2 substratum was formed as per Collaborative Research specifications on single-well chamber slides; Matrigel 625 pg/cm2 substratum was formed according to Collaborative Research protocol also on single-well chamber slides; glass single-well chamber slides were used for the glass substratum; and the precoated 35 mm culture dishes of fibronectin, type I collagen, and laminin were used for those surfaces respectively. Preparation of substrata for the morphology assay was as follows: fibronectin 1-2 pg/cm2, laminin 1-2 pg/cm2, Matrigel 125 pg/cm2, type I collagen 80-100 pg/cm2, and type IV collagen 15-20 pg/cm2 substrata were formed on double-well chamber slides according to Collaborative Research specifications. Glass double-well chamber slides were used for the glass substratum. Dihydrotestosterone (DHT) was obtained from Sigma Chemical Co. (St. Louis, MO). Cell Culture The human metastatic prostate cancer LNCaP cell line at passage 15 was obtained from the American Type Culture Collection (ATCC-12301, Rockville, MD) 1181. LNCaP cells were grown and maintained in RPMI 1640 medium containing 10% fetal bovine serum (FBS). Cell lines for morphometric assay alone were as follows: 20 year-old normal nonimmortal skin fibroblast GM03440A (N.I.G.M.S. Human Genetic Mutant Cell Repository, Camden, NJ), fibrosarcoma Hs 913T (ATCC HTB 152), immortalized normal rat kidney epithelial cell NRK-52E (ATCC CRL 1571), H-ras-transformed rat kidney epithelial cell KNRK (ATCC CRL 1569), nonimmortal primary culture human hepatocyte, transformed human hepatocyte FAO. 1 (courtesy of Dr. L. Reid, Albert Einstein School of Medicine), nonimmortal Syrian hamster embryo cell at passage 11 (courtesy of Dr. J. C. Barrett, N.I.E.H.S.), chemically transformed SHE cell line BP6T (courtesy of Dr. J. C. Barrett), Dunning rat prostate adenocarcinoma R3327G, and Dunning rat prostate adenocarcinoma MAT-LyLu (both cell lines courtesy of Dr. J. Isaacs, Johns Hopkins University). Cell Growth Assay
LNCaP cells were seeded in triplicate at a concentration of 1 X lo4 cells on each of the precoated culture dishes or single-well chamber slides prepared with the appropriate substrata as described above. Cells were incubated at 37°C for 48 hr in 1 nM DHT or control medium and then trypsinized (Dispase was used for Matrigel chambers). Cell counts were performed in triplicate by using a Coulter cell counter and checked for accuracy with a hemocytometer. Cell Motility Assay LNCaP cells were filmed by using time-lapse videomicroscopy in 2 hr segments. After seeding with 2 x lo4 cells/35 mm precoated culture dish or single-well chamber slide, cells were incubated at 37°C for 24 hr in 1 nM DHT on control medium. Cells were viewed with a high-resolution black and white video camera
32
Murphy et al.
(DACE MTI, Series 66, Michigan City, IN) at X 400 magnification with an inverted Zeiss (IM35) microscope (Hoffman optics) with heated stage as previously described [19]. Cells were videotaped with a time-lapse video recorded (JVC Br9000) every 15 seconds. Translational motility of each cell was measured from the TV monitor over 2 hr by directly measuring the total path distance each cell travelled [ 191. A minimum of 25 cells were measured for each surface. Morphometric Analysis Cells were plated on substrata-prepared double-well chamber slides at a concentration of 2 x lo4 cells per well. Cells were incubated at 37°C for 24 hr in 1 nM DHT or control medium. After incubation, chambers were transferred to a 37°C heated microscope stage (Zeiss TRZ 2700, Thornwood, NY) for microscopy. Ten images of cell fields were captured by a Compac Deskpro 386/25 by using the Zeiss DynaCell Program. Fifty cells and nuclei of each specimen were digitized and the X-Y coordinates of the cell and nuclear boundaries were stored by using DynaCell software and a digitizer tablet (Summasketch model MM-1201). This system digitized standard contours with an accuracy of greater than 95% [20]. Cell and nuclear area in square microns and perimeter in microns are measured for each cell. Data computation and analysis algorithms for DynaCell were written in APL*Plus (a program language, STSC, Rockville, MD). Prostatic-Specific Antigen (PSA) Assay A Tandem-R PSA Kit was obtained from Hybritech, Inc. (San Diego, CA). LNCaP cells were seeded in triplicate at a concentration of 1.5 X lo5 on each of the precoated culture dishes or single-well chamber slides prepared as described above. Cells were incubated at 37°C for 48 hr in 1 nM DHT or control medium and then a 100 p1 sample of medium from each culture dish was removed and tested according to the Tandem-R PSA Kit protocol by using a gamma counter. Statistics
Statistical analysis was performed by using Statgraphics v.4.0 (Statistical Graphics Corp., New York, NY). Statistical significance was determined by using the Student’s t test. RESULTS Static Cell Structure Cell and nuclear area are a two-dimensional projection of cell morphology; i.e., the more spread out a cell is, the larger its cell and nuclear area will be. As a cell flattens out, it nucleus also flattens out. Conversely, as a cell rounds up to a more spherical shape, its cell and nuclear area decrease, presumably with no change in cell volume. These alterations in cell and nuclear shape can be monitored by calculating a nuclear areakytoplasmic area (N/C) ratio. Nonconfluent cells which are spread out on a surface have proportionally less of their area taken up by the nucleus and have a low N/C ratio. As cells round up or become less attached to the substratum, the N/C ratio increases as more of the relative two-dimensional area is occupied by the nucleus. Figure 1 demonstrates that nontransformed cells in vitro typically have N/C ratios of less than 0.075, whereas transformed cells, which are structurally altered and typically are less well attached to the substratum, have N/C ratios of 0.1 or greater.
ECM and DHT Effects on LNCaP Cells
FIBROBLAST
KIDNEY
LIVER
SYRIAN HAMSTER EMBRYO
33
PROSTATE
Fig. 1. The nuclear areakell area ratio of nontransformed and transformed cell lines. The nuclear areakytoplasmic area (N/C) ratio of nonconfluent cells in vitro can be used to monitor cell morphology (see text). Nontransformed cells demonstrate N/C ratios of less than 0.075 (dotted line). Transformed cells demonstrate N/C ratios of 0.1 or greater across all cell lines studied. Fibroblast: nontransformed = 20 year-old normal nonimmortal skin fibroblast, transformed = fibrosarcoma; kidney: nontransformed = immortalized normal rat kidney epithelial cell, transformed = H-ras-transformed rat kidney epithelial cell; liver: nontransformed = nonimmortal primary culture human hepatocyte, transformed = transformed human hepatocyte; Syrian hamster embryo: nontransformed = nonimmortal Syrian hamster embryo cell, transformed = chemically transformed SHE cell line; prostate: transformed = Dunning rat prostate adenocarcinoma MAT-LyLu.
The N/C ratio, therefore, can be used to monitor cell and nuclear morphology. Previously, it has been demonstrated that the nucleus of a capillary endothelial cell was connected to the cell periphery by the cytoskeleton and when they exposed these cells to trypsin-EDTA, the nuclei rounded up in parallel with the cytoplasm [16,21]. Therefore, in nontransformed cells alterations in nuclear morphology appear to be coordinated with alterations in cell morphology. Nonconfluent LNCaP cells plated on the standard tissue culture substratum glass have a cell area of 680 pm2 and a nuclear area of 172 pm2, resulting in a nuclear/cytoplasmic area (N/C) ratio of 0.25 (see Fig. 2A-C). Altering the substrata has significant effects on cell and nuclear size. When these cells are plated on Matrigel, collagen I, and fibronectin there is no significant change in cell size and very little observable alteration in gross celI morphology. However, nuclear size increases on Matrigel, does not change on collagen I, and decreases on fibronectin. Plating the cells on collagen IV or laminin results in a significant increase in cell size (P< 0.001), but nuclear area decreases on collagen IV and increases on laminin. This suggests that the interaction of the LNCaP nucleus with the ECM through the cytoskeleton is not a passive one where the nuclear morphology simply mirrors cell morphology. The addition of 1 nM DHT to LNCaP cells on each of these surfaces for 24 hr resulted in dramatic increases in cell size which paralleled a more spread out cellular appearance. Figure 2A illustrates the uniform increase in LNCaP cell size on all surfaces after treatment with 1 nM DHT. Statistically significant increases in cell size
34
Murphyetal.
SUBSTRATUM Fig. 2.
ECM and DHT Effects on LNCaP Cells
35
as compared to controls occurred on glass, collagen I, laminin, and fibronectin (P
Effects of Extracellular Matrix Components and Dihydrotestosterone on the Structure and Function of Human Prostate Cancer Cells Brian C. Murphy, Kenneth J. Pienta, and Donald S. Coffey The Johns Hopkins Oncology Center, and The Brady Urological Research Institute, The Johns Hopkins School of Medicine, Baltimore, Maryland The extracellular matrix (ECM) has been shown to play a major role in cell structure and function. Several studies have demonstrated that the ECM can alter cell morphology and effect DNA synthesis and gene expression. The ECM also interacts with growth hormones which have been shown to be located in or near the ECM where they are believed to effect cell structure and function. In the nontransformed cell, these ECM and hormone-mediated effects appear to be tightly regulated and this is believed to be accomplished through cell receptor-tissue matrix interactions. We, therefore, undertook a study to determine the effects of a variety of ECM components and the androgenic hormone dihydrotestosterone (DHT) on the structure and function of the human prostate cancer cell line, LNCaP. The effects of individual matrix components in the presence and absence of 1 nM DHT on the static and dynamic morphology, growth rate, and PSA production of the LNCaP cell line were studied. We determined that the ECM and DHT interact in complex ways to effect cell structure and function. DHT produced alterations in cytoplasmic structure that increased cell size and decreased the nuclear aredcytoplasmic area ratio. Dynamic cell structure as measured by cell motility was very sensitive to the ECM components and the presence of DHT. PSA and growth could be regulated by substratum and DHT and there was an inverse relationship between PSA production and growth rate. These data exemplify the complex interactions which occur between prostate cancer cells, ECM components, and exogenous DHT that are reflected in cell structure and function.
Key words: basement membrane, LNCaP cells, androgen effects, PSA production
INTRODUCTION
What a cell touches has a major role in determining what a cell does. In classic studies, Moscona and Folkman controlled DNA synthesis in endothelial cells by manipulating cell shape by changing the substratum on which the cells were plated and found that DNA synthesis could be shut off by induced rounding up of the cell [ 11. Furthermore, Gospodarowicz et al. demonstrated that the regulation of corneal epithelial cell shape by different substrata directed the cells’ ability to respond to mitogenic hormones [2]. In addition to this role in the control of cell shape, the
Received for publication April 30, 1991; accepted June 26, 1991.
Dr. Kenneth J. Pienta’s current address is Wayne State University School of Medicine, Division of Hematology/Oncology, P.O. Box 02188, Detroit, MI 48202-0188. Address reprint requests there. 0 1992 Wiley-Liss, Inc.
30
Murphyet al.
substratum or extracellular matrix (ECM) has been shown in many studies to play an important part in the control of cell differentiation and function. Cunha and colleagues have demonstrated that the ECM can be responsible for functional differentiation in development when they showed that urogenital sinus mesenchyme induces adult urinary bladder epithelial cells to remodel to form prostatic epithelial cells [3]. Reddi and Anderson have observed that mature fibroblasts undergo further differentiation to form new chondroblasts and chondrocytes when they were exposed to contact with a demineralized bone collagen matrix and osteogenic factors [4]. These studies as well as others demonstrate that altering cell morphology alters DNA synthesis and gene expression [5-91. The ECM itself also interacts with hormones to effect cell structure and function. Several types of hormones have been shown to be located or sequestered in the ECM where they are believed to effect gene expression. Lactogenic hormones located in the basement membrane have been shown to alter cell shape and cause changes in mRNA stability and its association with the cytoskeleton [ 101. Bissell and colleagues demonstrated that when mammary epithelial cells were placed on a biomatrix with mammary basement membrane, their expression of casein and lactalbumin increased as much as 160-fold when compared to cells grown on tissue culture plastic [ 10,l I]. Reid and her colleagues demonstrated that normal rat hepatocytes required both a hormone-supplemented media and a specific biomatrix to synthesize and stabilize mRNA-neither alone was sufficient [ 121. Therefore, the extracellular matrix has been shown to alter the hormonal responsiveness, cell shape, and gene expression of cells in a specific manner. This coordination of cell function by the ECM may have added significance in the study of cell transformation, and recently, it has been shown that ECM components are altered in transformed rat prostate fibroblasts [ 131. It is now evident that not only are individual extracellular matrix components important but that the interaction of several of the components is involved in regulative hormonal responses and in cell signalling and coordinating normal cell function. It has been postulated that this coordination of cellular function is accomplished by the tissue matrix [14-161. The tissue matrix is the three-dimensional structural framework of the cell which consists of the interacting dynamic skeletons of the nuclear matrix, the cytoskeleton, and the extracellular matrix. This dynamic skeletal system physically connects the extracellular matrix to the nucleus and the DNA, is tissue specific, and may be altered in the cancer cell [15-171. Therefore, we undertook a study to determine if cell structure as dictated by the extracellular matrix components and hormones correlated with alterations in cell function of the human prostate cancer, LNCaP. Specifically, the effects of the individual matrix components collagen I, collagen IV, laminin, and fibronectin, and the complete basement membrane Matrigel, on the morphology, motility, growth rate, and PSA production of the LNCaP cell line were studied. We then investigated the effects of these ECM subcomponents on the structural and functional parameters in the presence of DHT,a steroid which has been demonstrated to effect prostate cell function. MATERIALS AND METHODS Materials BioCoat Matrix culture dishes (35 mm) precoated with fibronectin, type I collagen, and laminin were obtained from Collaborative Research (Bedford, MA). The
ECM and DHT Effects on LNCaP Cells
31
extracellular matrix products human fibronectin, murine laminin, rat tail type I collagen, murine type IV collagen, and Matrigel, as well as dispase were also obtained from Collaborative Research. Lab-Tek single-well chamber slides (2 cm X 4.2 cm) and double-well chamber slides (2 cm X 2 cm) were obtained from Nunc Inc. (Naperville, IL). Preparation of substrata for the motility, growth, and PSA assays was as follows: type IV collagen 15-20 pg/cm2 substratum was formed as per Collaborative Research specifications on single-well chamber slides; Matrigel 625 pg/cm2 substratum was formed according to Collaborative Research protocol also on single-well chamber slides; glass single-well chamber slides were used for the glass substratum; and the precoated 35 mm culture dishes of fibronectin, type I collagen, and laminin were used for those surfaces respectively. Preparation of substrata for the morphology assay was as follows: fibronectin 1-2 pg/cm2, laminin 1-2 pg/cm2, Matrigel 125 pg/cm2, type I collagen 80-100 pg/cm2, and type IV collagen 15-20 pg/cm2 substrata were formed on double-well chamber slides according to Collaborative Research specifications. Glass double-well chamber slides were used for the glass substratum. Dihydrotestosterone (DHT) was obtained from Sigma Chemical Co. (St. Louis, MO). Cell Culture The human metastatic prostate cancer LNCaP cell line at passage 15 was obtained from the American Type Culture Collection (ATCC-12301, Rockville, MD) 1181. LNCaP cells were grown and maintained in RPMI 1640 medium containing 10% fetal bovine serum (FBS). Cell lines for morphometric assay alone were as follows: 20 year-old normal nonimmortal skin fibroblast GM03440A (N.I.G.M.S. Human Genetic Mutant Cell Repository, Camden, NJ), fibrosarcoma Hs 913T (ATCC HTB 152), immortalized normal rat kidney epithelial cell NRK-52E (ATCC CRL 1571), H-ras-transformed rat kidney epithelial cell KNRK (ATCC CRL 1569), nonimmortal primary culture human hepatocyte, transformed human hepatocyte FAO. 1 (courtesy of Dr. L. Reid, Albert Einstein School of Medicine), nonimmortal Syrian hamster embryo cell at passage 11 (courtesy of Dr. J. C. Barrett, N.I.E.H.S.), chemically transformed SHE cell line BP6T (courtesy of Dr. J. C. Barrett), Dunning rat prostate adenocarcinoma R3327G, and Dunning rat prostate adenocarcinoma MAT-LyLu (both cell lines courtesy of Dr. J. Isaacs, Johns Hopkins University). Cell Growth Assay
LNCaP cells were seeded in triplicate at a concentration of 1 X lo4 cells on each of the precoated culture dishes or single-well chamber slides prepared with the appropriate substrata as described above. Cells were incubated at 37°C for 48 hr in 1 nM DHT or control medium and then trypsinized (Dispase was used for Matrigel chambers). Cell counts were performed in triplicate by using a Coulter cell counter and checked for accuracy with a hemocytometer. Cell Motility Assay LNCaP cells were filmed by using time-lapse videomicroscopy in 2 hr segments. After seeding with 2 x lo4 cells/35 mm precoated culture dish or single-well chamber slide, cells were incubated at 37°C for 24 hr in 1 nM DHT on control medium. Cells were viewed with a high-resolution black and white video camera
32
Murphy et al.
(DACE MTI, Series 66, Michigan City, IN) at X 400 magnification with an inverted Zeiss (IM35) microscope (Hoffman optics) with heated stage as previously described [19]. Cells were videotaped with a time-lapse video recorded (JVC Br9000) every 15 seconds. Translational motility of each cell was measured from the TV monitor over 2 hr by directly measuring the total path distance each cell travelled [ 191. A minimum of 25 cells were measured for each surface. Morphometric Analysis Cells were plated on substrata-prepared double-well chamber slides at a concentration of 2 x lo4 cells per well. Cells were incubated at 37°C for 24 hr in 1 nM DHT or control medium. After incubation, chambers were transferred to a 37°C heated microscope stage (Zeiss TRZ 2700, Thornwood, NY) for microscopy. Ten images of cell fields were captured by a Compac Deskpro 386/25 by using the Zeiss DynaCell Program. Fifty cells and nuclei of each specimen were digitized and the X-Y coordinates of the cell and nuclear boundaries were stored by using DynaCell software and a digitizer tablet (Summasketch model MM-1201). This system digitized standard contours with an accuracy of greater than 95% [20]. Cell and nuclear area in square microns and perimeter in microns are measured for each cell. Data computation and analysis algorithms for DynaCell were written in APL*Plus (a program language, STSC, Rockville, MD). Prostatic-Specific Antigen (PSA) Assay A Tandem-R PSA Kit was obtained from Hybritech, Inc. (San Diego, CA). LNCaP cells were seeded in triplicate at a concentration of 1.5 X lo5 on each of the precoated culture dishes or single-well chamber slides prepared as described above. Cells were incubated at 37°C for 48 hr in 1 nM DHT or control medium and then a 100 p1 sample of medium from each culture dish was removed and tested according to the Tandem-R PSA Kit protocol by using a gamma counter. Statistics
Statistical analysis was performed by using Statgraphics v.4.0 (Statistical Graphics Corp., New York, NY). Statistical significance was determined by using the Student’s t test. RESULTS Static Cell Structure Cell and nuclear area are a two-dimensional projection of cell morphology; i.e., the more spread out a cell is, the larger its cell and nuclear area will be. As a cell flattens out, it nucleus also flattens out. Conversely, as a cell rounds up to a more spherical shape, its cell and nuclear area decrease, presumably with no change in cell volume. These alterations in cell and nuclear shape can be monitored by calculating a nuclear areakytoplasmic area (N/C) ratio. Nonconfluent cells which are spread out on a surface have proportionally less of their area taken up by the nucleus and have a low N/C ratio. As cells round up or become less attached to the substratum, the N/C ratio increases as more of the relative two-dimensional area is occupied by the nucleus. Figure 1 demonstrates that nontransformed cells in vitro typically have N/C ratios of less than 0.075, whereas transformed cells, which are structurally altered and typically are less well attached to the substratum, have N/C ratios of 0.1 or greater.
ECM and DHT Effects on LNCaP Cells
FIBROBLAST
KIDNEY
LIVER
SYRIAN HAMSTER EMBRYO
33
PROSTATE
Fig. 1. The nuclear areakell area ratio of nontransformed and transformed cell lines. The nuclear areakytoplasmic area (N/C) ratio of nonconfluent cells in vitro can be used to monitor cell morphology (see text). Nontransformed cells demonstrate N/C ratios of less than 0.075 (dotted line). Transformed cells demonstrate N/C ratios of 0.1 or greater across all cell lines studied. Fibroblast: nontransformed = 20 year-old normal nonimmortal skin fibroblast, transformed = fibrosarcoma; kidney: nontransformed = immortalized normal rat kidney epithelial cell, transformed = H-ras-transformed rat kidney epithelial cell; liver: nontransformed = nonimmortal primary culture human hepatocyte, transformed = transformed human hepatocyte; Syrian hamster embryo: nontransformed = nonimmortal Syrian hamster embryo cell, transformed = chemically transformed SHE cell line; prostate: transformed = Dunning rat prostate adenocarcinoma MAT-LyLu.
The N/C ratio, therefore, can be used to monitor cell and nuclear morphology. Previously, it has been demonstrated that the nucleus of a capillary endothelial cell was connected to the cell periphery by the cytoskeleton and when they exposed these cells to trypsin-EDTA, the nuclei rounded up in parallel with the cytoplasm [16,21]. Therefore, in nontransformed cells alterations in nuclear morphology appear to be coordinated with alterations in cell morphology. Nonconfluent LNCaP cells plated on the standard tissue culture substratum glass have a cell area of 680 pm2 and a nuclear area of 172 pm2, resulting in a nuclear/cytoplasmic area (N/C) ratio of 0.25 (see Fig. 2A-C). Altering the substrata has significant effects on cell and nuclear size. When these cells are plated on Matrigel, collagen I, and fibronectin there is no significant change in cell size and very little observable alteration in gross celI morphology. However, nuclear size increases on Matrigel, does not change on collagen I, and decreases on fibronectin. Plating the cells on collagen IV or laminin results in a significant increase in cell size (P< 0.001), but nuclear area decreases on collagen IV and increases on laminin. This suggests that the interaction of the LNCaP nucleus with the ECM through the cytoskeleton is not a passive one where the nuclear morphology simply mirrors cell morphology. The addition of 1 nM DHT to LNCaP cells on each of these surfaces for 24 hr resulted in dramatic increases in cell size which paralleled a more spread out cellular appearance. Figure 2A illustrates the uniform increase in LNCaP cell size on all surfaces after treatment with 1 nM DHT. Statistically significant increases in cell size
34
Murphyetal.
SUBSTRATUM Fig. 2.
ECM and DHT Effects on LNCaP Cells
35
as compared to controls occurred on glass, collagen I, laminin, and fibronectin (P
