differential expression of g1, cyclins during human placentogenesis

8
Placenta (1997), 18, 9-I 6 Differential Expression of G1 Cyclins During Human Placentogenesis J. A. DeLoiaacbrC, J. M. Burlingameb and J. S. Krasnowb Magee-Womens Research Institute, a Departments of Cell Biology and Physiology and b Obstetrics, Gynaecology and Reproductive Sciences, University of Pittsburgh, Pittsburgh, PA 15213, USA Paper accepted 6 August 1996 Cyclins are proteins that support the progression of cell-cycle stages in proliferating cells. The purpose of this study was to determine which of the cyclin genes is involved in the regulation of normal human trophoblast proliferation. The presence and cellular localization of four G, cyclins Dl, D2, D3 and E, were determined by immunohistochemistry. This analysis indicated that cyclins E and D3 are the predominant cyclins in villous trophoblast. D2 was present only within the villous core, in fetal macrophages. Positive immunoreactivity for cyclin Dl was strongest in second and third trimester placentae, in the cells lining the intravillous vessels with additional reactivity in extravillous cytotrophoblasts. Because cyclin E protein was present in a greater percentage of cells than those that are dividing, Western blot analysis was performed to validate the fidelity of the immunohistochemistry data. The results of the Western analysis revealed that two forms of cyclin E protein of the appropriate size were present. Data collected from this study suggest that within the trophoblast lineage, cyclins D3 and E are important cell cycle regulatory proteins, and further, that cyclin E may function in trophoblast terminal differentiation as well. Placenta (1997), 18, 9-16 0 1997 W. B. Saunders Company Ltd INTRODUCTION The human placenta is unique in many aspects of its growth and differentiation, including the requirement to invade another organ, the uterus, for survival. Although placentally derived trophoblast cells are highly invasive and mitotically active, in normal development, their proliferation is controlled and invasion is limited. A fundamental question in placental biology is what influences the decision of trophoblast cells to divide or to differentiate and syncytialize. Recent studies in a variety of animal systems have produced a considerable body of knowledge regarding the control of cell proliferation in eukaryotic systems. A key group of proteins directly involved in regulating cell division are the cyclins, so named, because the first cyclin proteins discovered were found to oscillate in relation to cell cycle stage (Evans et al., 1983). Cyclins function by binding to, and activating, pre-existing cyclin dependent kinases (cdks), which in turn phosphorylate proteins necessary for chromosome condensation, cytoskelelal reorganization and nuclear envelop breakdown (O’Farrell et al., 1989; Nurse 1990). The regulation of the eukaryotic cell cycle occurs through the sequential formation, activation, and subsequent inactivation of cyclin/kinase complexes. Widely conserved in evolution, cyclins can be divided into two groups based on the timing of their appearance in the cell cycle, the mitotic cyclins; A and B, and the G, cyclins; cyclins C, the D type cyclins and cyclin E (Lew, Dulic and Reed, 1991). Although much is known about the orderly appear- c To whom correspondence should be addressed. 0143%4004/97/010009+08 $12.00/O antes, disappearances and associations of cyclins in yeast, control of cell division in mammals is considerably more complex. However, the defining checkpoint in cell-cycle pro- gression in all eukaryotic cells is at the G, stage (Dou et al., 1993). Entry into and progression through G, is dependant upon extracellular signals such as growth factors, cytokines and nutrient availability (Pardee, 1989). Once a cell passes this restriction point, termed R in mammalian cells and START in yeast cells, it is then committed to a round of DNA synthesis (S) and cell division. In mammalian cells, multiple cdks, as well as cyclins exist. As a cell enters into a round of division, at least one of the D-type cyclins appears and associateswith cdk4, cdk5, cdk6 or cdk2, depending on the cell type (Xiong, Zhang and Beach, 1992). For example, in macrophages, cdk4 is the predominate cdk, while in the brain, cdk5 alone is expressed (Tsai et al., 1993, 1994). In the latter part of the G, phase of the cell cycle, cyclin E accumulates and forms a complex with the protein kinase cdk2, suggesting that cyclin E is an important regulatory molecule for the Gi-S transition (Koff et al., 1992; Dulic, Lees and Reed, 1992). Unlike the individual D-type cyclins, which are not always present in dividing cells, cyclin E appears to be essential for the Gi-S transition. In addition to cyclin E accumulation late in G,, cyclin C mRNA also peaks, although the significance of this cyclin is unclear. Once a cell enters S phase, cyclin E is no longer required, and cdk2 is found in association with cyclin A, which is necessary for the onset of DNA replication (Girard et al., 1991). The purpose of this study was to investigate the early intracellular events that regulate cellular proliferation within 0 1997 W. B. Saunders Company Ltd

Upload: ja-deloia

Post on 15-Sep-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Differential expression of g1, cyclins during human placentogenesis

Placenta (1997), 18, 9-I 6

Differential Expression of G1 Cyclins During Human Placentogenesis

J. A. DeLoiaacbrC, J. M. Burlingameb and J. S. Krasnowb

Magee-Womens Research Institute, a Departments of Cell Biology and Physiology and b Obstetrics, Gynaecology and Reproductive Sciences, University of Pittsburgh, Pittsburgh, PA 15213, USA

Paper accepted 6 August 1996

Cyclins are proteins that support the progression of cell-cycle stages in proliferating cells. The purpose of this study was to determine which of the cyclin genes is involved in the regulation of normal human trophoblast proliferation. The presence and cellular localization of four G, cyclins Dl, D2, D3 and E, were determined by immunohistochemistry. This analysis indicated that cyclins E and D3 are the predominant cyclins in villous trophoblast. D2 was present only within the villous core, in fetal macrophages. Positive immunoreactivity for cyclin Dl was strongest in second and third trimester placentae, in the cells lining the intravillous vessels with additional reactivity in extravillous cytotrophoblasts. Because cyclin E protein was present in a greater percentage of cells than those that are dividing, Western blot analysis was performed to validate the fidelity of the immunohistochemistry data. The results of the Western analysis revealed that two forms of cyclin E protein of the appropriate size were present. Data collected from this study suggest that within the trophoblast lineage, cyclins D3 and E are important cell cycle regulatory proteins, and further, that cyclin E may function in trophoblast terminal differentiation as well.

Placenta (1997), 18, 9-16 0 1997 W. B. Saunders Company Ltd

INTRODUCTION

The human placenta is unique in many aspects of its growth and differentiation, including the requirement to invade another organ, the uterus, for survival. Although placentally derived trophoblast cells are highly invasive and mitotically active, in normal development, their proliferation is controlled and invasion is limited. A fundamental question in placental biology is what influences the decision of trophoblast cells to divide or to differentiate and syncytialize. Recent studies in a variety of animal systems have produced a considerable body of knowledge regarding the control of cell proliferation in eukaryotic systems. A key group of proteins directly involved in regulating cell division are the cyclins, so named, because the first cyclin proteins discovered were found to oscillate in relation to cell cycle stage (Evans et al., 1983). Cyclins function by binding to, and activating, pre-existing cyclin dependent kinases (cdks), which in turn phosphorylate proteins necessary for chromosome condensation, cytoskelelal reorganization and nuclear envelop breakdown (O’Farrell et al., 1989; Nurse 1990). The regulation of the eukaryotic cell cycle occurs through the sequential formation, activation, and subsequent inactivation of cyclin/kinase complexes.

Widely conserved in evolution, cyclins can be divided into two groups based on the timing of their appearance in the cell cycle, the mitotic cyclins; A and B, and the G, cyclins; cyclins C, the D type cyclins and cyclin E (Lew, Dulic and Reed, 1991). Although much is known about the orderly appear-

c To whom correspondence should be addressed.

0143%4004/97/010009+08 $12.00/O

antes, disappearances and associations of cyclins in yeast, control of cell division in mammals is considerably more complex. However, the defining checkpoint in cell-cycle pro- gression in all eukaryotic cells is at the G, stage (Dou et al., 1993). Entry into and progression through G, is dependant upon extracellular signals such as growth factors, cytokines and nutrient availability (Pardee, 1989). Once a cell passes this restriction point, termed R in mammalian cells and START in yeast cells, it is then committed to a round of DNA synthesis (S) and cell division.

In mammalian cells, multiple cdks, as well as cyclins exist. As a cell enters into a round of division, at least one of the D-type cyclins appears and associates with cdk4, cdk5, cdk6 or cdk2, depending on the cell type (Xiong, Zhang and Beach, 1992). For example, in macrophages, cdk4 is the predominate cdk, while in the brain, cdk5 alone is expressed (Tsai et al., 1993, 1994). In the latter part of the G, phase of the cell cycle, cyclin E accumulates and forms a complex with the protein kinase cdk2, suggesting that cyclin E is an important regulatory molecule for the Gi-S transition (Koff et al., 1992; Dulic, Lees and Reed, 1992). Unlike the individual D-type cyclins, which are not always present in dividing cells, cyclin E appears to be essential for the Gi-S transition. In addition to cyclin E accumulation late in G,, cyclin C mRNA also peaks, although the significance of this cyclin is unclear. Once a cell enters S phase, cyclin E is no longer required, and cdk2 is found in association with cyclin A, which is necessary for the onset of DNA replication (Girard et al., 1991).

The purpose of this study was to investigate the early intracellular events that regulate cellular proliferation within

0 1997 W. B. Saunders Company Ltd

Page 2: Differential expression of g1, cyclins during human placentogenesis

10

the human trophoblast lineages. To address this question, we ascertained the presence and cellular localization of four cyclin proteins expressed in the early and late G, phase of the cell cycle; cyclins Dl, D2, D3 and E. Results of this study indicate that cyclins D3 and E are present within the trophoblast cells of the developing villi as well as in term villi, while cyclins Dl and D2 are present in non-trophoblast lineages of the placental villi.

MATERIALS AND METHODS

Sample preparation

Placental tissue was collected at the time of routine dilatation, evacuation and curettage, for pregnancy terminations between 6 and 16 weeks gestation. Weeks 17-20 were collected at the time of dilatation and evacuation, and third trimester placentae were obtained at the time of caesarean section. Only normal placentae were used in this study. For protein isolation, samples were snap frozen in liquid nitrogen. Tissues for paraffin embedding were fixed overnight at room temperature in 10 per cent buffered formalin, dehydrated through graded alcohols, cleared in xylenes and embedded in Paraplast. A total of 22 samples was analysed in this study; 11 first trimester, eight second trimester and three term placentae.

lmmunohistochemistry

Paraffin-embedded tissue was sectioned at 4-6 pm thickness, placed on SuperFrost/Plus glass slides (Fisher Scientific, Pittsburgh, PA, USA) and baked for 5-10 min at 60°C. Sections were deparaffinzed with xylenes and rehydrated through a graded series of alcohols. After peroxide and protein blocking steps, the primary antibody was allowed to bind for 60 min at room temperature in a humidified chamber. Primary antibody binding was detected with the Super Sensitive Mulitlink Immunostaining Kit (Biogenex, San Ramon, CA, USA). Briefly, a biotinylated multilink secondary antibody, compatible with both mouse and rabbit primary antibodies, was incubated for 20 min at room temperature, followed by addition of the label; a horseradish peroxidase-tagged strepa- vidin molecule. The substrate for the horseradish peroxidase, DAB (3,3’-diaminobenzidine), was metabolized to a deep

Placenta (1997), Vol. 18

brown product, indicative of positive primary antibody binding. Mayer’s haematoxylin was used as a (blue) counterstain. An avidin-biotin block (Vector Laboratories, Burlingame, CA, USA) was used for cyclins E and Dl and CD-68 immunostaining.

The primary antibodies used in this study were affinity- purified rabbit polyclonal antibodies to human cyclins Dl, D2, D3 and E, (Santa Cruz Biotechnology, Santa Cruz, CA, USA; catalogue numbers SC-92, ~~-181, SC-182 and ~~-198, respectively), mouse monoclonal IgG, antibody to human cytokeratin-18, (Biogenex; clone DClO), and mouse monoclonal antibody to human CD68, (Dako Corporation, Carpinteria, CA, USA; clone KPl). The cyclin D2, D3 and E primary antibodies were diluted 1 : 100 with KPBS, 0.4 per cent Triton-X, to a final working concentration of 1 pg/ml. Cyclin Dl was diluted 1 : 200 for a working concentration of 0.5 pg/ml. Negative controls for these experiments were non- immune serum from the same source as the primary antibody. Additionally, the cyclin E antibody was immunoadsorbed by reacting it with a control peptide (~~-198 P) at a lo-fold (by weight) excess of peptide overnight at 4°C. The CD68 anti- body required an antigen retrieval step to unmask the antigen, which consisted of boiling the sections in 10 mM citrate buffer (pH 6.0) for 5 min before the protein block.

Western blot analysis

Fifteen micrograms of protein from human placental samples was electrophoresed through a 10 per cent sodium dodecyl sulphate-polyacrylamide gel with a 5 per cent stacking gel. Protein was electrophoretically transferred to a nitrocellulose membrane using the Bio-Rad Trans-Blot Electorphoretic Transfer Cell (Bio-Rad Laboratories, Hercules, CA, USA). The membrane was then blocked with a 1 per cent casein solution and incubated with a 1 : 500 dilution of 100 pg/ml rabbit polyclonal anti-cyclin E solution, and washed thor- oughly. A dilution of 1 : 750 of a 150 U/ml secondary horse- radish peroxidase detection antibody solution was then incubated with the filter and after a series of washes, the bound secondary antibody was detected using a chemiluminescence assay (Boehringer Mannheim, Indianapolis, IN, USA). Posi- tive control samples consisted of 50 ng of a 10 amino acid cyclin E peptide and a 1 : 100 dilution of total rabbit IgG.

Figure. 1. Comparison of immunostaining for cyclins Dl, D3 and E in placental villi through development. Cyclins Dl (a-c), D3 (d-f) and E (g-i) were identified in formalin fixed, paraffin embedded sections by indirect staining using diaminobenzidine (DAB) and Mayer’s haematoxylin counterstain. First trimester placenta showed specific immunoreactivity for cyclin Dl lining the newly formed blood vessels within the villi (a). Nucleated erythrocytes, and trophoblast cells are non-staining. Second trimester (b) and term placenta (c) show increased Dl immunoreactivity concomitant with increased numbers of blood vessels. D3 cyclin immunoreactivity appears to be limited to cytotrophoblasts of the first trimester villi (d) and second trimester villi (e). The variation in D3 staining intensity may be related to cell-cycle stage. By term (f), D3 immunostaining is confined to scattered solitary trophoblast cells. Unlike the other cyclins, cyclin E immunoreactivity occurs in different trophoblast cell types as development proceeds. In the first trimester placenta (g), there is abundant DAB staining in the syncytiotrophoblast layer, as well as in trophoblast columns (top, center). Few villous cytotrophoblasts appear positive for cyclin E immunostaining. In the second trimester placenta (h), this pattern of staining is no longer present; many cytotrophoblasts are now positive, while few syncytiotrophoblasts are immunoreactive for cyclin E. In the term placenta (i), both trophoblast cell types are positive for cyclin E staining. There are also some unidentified cell types which are immunoreactive, both within the villous core as well as within some blood vessels, (lower left hand corner). In (j), the same term placenta is shown following incubation with non-immune rabbit serum. All photomicrographs are x 200 magnification, with the exception of (c), which is a x 400 magnification.

Page 3: Differential expression of g1, cyclins during human placentogenesis

DeLoia, Burlingame and Krasnow: Cyclin Expression in the Placenta

Page 4: Differential expression of g1, cyclins during human placentogenesis

12 Placenta (1997), Vol. 18

Total smooth muscle protein was used as a non-proliferative negative control.

RESULTS

Cyclins Dl and D2 are expressed primarily in non-trophoblast lineages

Immunoreactivity for cyclins Dl and D2 was assessed on normal placentae from gestational week 7 through term. At 7 weeks gestation, there was already Dl immunoreactivity, which increased with increasing gestational age, throughout pregnancy. Northern blot analysis for detection of cyclin Dl message paralleled this pattern of expression (data not shown). At 7 weeks, only sparse staining was detectable, within the villous core [Figure l(a)]. Those cells which were positive were thin and elongated. None of the villous trophoblast popu- lations was positive for Dl staining. By the second trimester, cyclin Dl immunoreactivity was abundant around the intra- villous blood vessels, in what appeared to be the endothelial cells lining the vessels [Figure l(b)]. Note that the blood cells within the vessels were non-reactive. By the third trimester, the periluminal staining was remarkable [Figure l(c)], increas- ing concurrently with the increasing number of intravillous blood vessels. In addition to the periluminal staining, extravil- lous cells also stained positively for cyclin Dl antibody. These immunoreactive cells had the appearance of extravillous tro- phoblasts; therefore, we stained serial sections for cyclin Dl and cytokeratin 18, a marker of trophoblast cells. The pattern of immunoreactivity to both antibodies paralleled one another. Therefore, we suggest that these cyclin Dl reactive cells are most likely extravillous trophoblast cells (Figure 2).

D2 cyclin immunoreactivity was apparent in placentae from 7-19 weeks gestation, but essentially absent in term placentae. Positive staining cells were restricted to intravillous, non- trophoblast cells, in a pattern unique from that of Dl im- munoreactivity. The cellular morphology; large, pleomorphic cells, with granulated cytoplasm and a vacuolated appearance, and the location within the villous mesenchyme, suggested that D2 antibody stained placental macrophages, or Hofbauer cells (Figure 3). To address this possibility, we compared the pattern of immunoreactivity of cyclin D2 and CD68, a marker for tissue macrophages, in serial sections 5 urn thick. Although individual cells were not double-labelled with the two primary antibodies, the patterns of expression of D2 and CD68 mirrored one another, thereby supporting our original identi- fication of these cells as macrophages. In contrast to the first and second trimester placentae, in the term placenta, the patterns of immunostaining for D2 and CD68 immuno- reactivity no longer parallel one another. By the end of pregnancy D2 immunoreactive cells were found rarely and when present, were only within fetal blood vessels of the term placental villi. In contrast, CD68 positive cells were located within the villous mesenchyme as well as within placental blood vessels.

Figure. 2. Cyclin Dl immunostaining in extravillous trophoblasts. Serial sections of an l&week placenta were stained for cyclin Dl (a) and cytokeratin-18 (b). The similar pattern of cellular staining demonstrates the trophoblast identity of these Dl positive cells outside the placental villi. A non-immune serum control is shown in panel (c). Magnification is X ZOO.

Cyclin D3 in the developing placenta

The pattern of cyclin D3 immunoreactivity in the placenta is consistent with a functional role in the promotion of villous trophoblast proliferation. Within the first trimester placenta, cyclin D3 protein is found in the nuclei of villous cytotropho- blast cells [Figure l(d)], and occasional cells within cyto- trophoblast columns (data not shown), both of which are proliferating cell populations. The second trimester placenta is

Page 5: Differential expression of g1, cyclins during human placentogenesis

DeLoia, Burlingame and Krasnow: Cyclin Expression in the Placenta 13

Figure 3. Expression of cyclin D2 in fetal macrophages. Strong immuno- reactivity to cyclin D2 antibody was apparent in pleomorphic cells within the villous core of a second trimester placenta (a). In (b) is a serial section incubated with an antibody to CD68, a macrophage marker. The immuno- reactive cells had a morphology similar to those that were cyclin D2 positive. Magnification is X 200.

also strongly reactive to cyclin D3 antibody within the villous cytotrophoblasts [Figure l(e)]. Neither the syncytiotro- phoblast nuclei nor cytoplasm showed appreciable anti-D3 reactivity. No other cell types in the villi reacted with the cyclin D3 antibody. This pattern of cellular staining was maintained in the term placenta as well, albeit at a decreased rate of reactivity [Figure l(f)]. Thus the expression of cyclin D3 protein parallels the mitotic index found in trophoblasts throughout placental development (Tedde and Tedde-Piras, 1978). Also notable was a variation in the intensity of D3 immunostaining. This variation may be related to the stage of the cell cycle; lighter staining cells may be just entering or leaving G,. These data support the notion that cyclin D3 plays a pivotal role in trophoblast cell proliferation.

Cyclin E localization and developmental stage

The most intriguing expression pattern in the placentae examined was that of cyclin E. In the first trimester placentae cyclin E protein was found primarily in the nuclei of syn- cytiotrophoblast cells and the cells within trophoblastic

Marker

CycE

SmM

Blank

8-4

16-1

18-3

19-6

21-4

Term

Term

Figure 4. Western blot analysis of cyclin E expression. Fifteen micrograms of protein from placental cell lysates were electrophoresed through a polyacryl- amide stacking gel, transferred to a nitrocellulose membrane and hybridized with a cyclin E antibody. The identity of the samples analysed are as follows: lane 1, protein size marker; lane 2 is an immunoglobulin positive control for the chemiluminescence detection assay; lane 3 is a 10 amino acid cyclin E positive control, which ran much lower than what is represented on the gel; lane 4 is smooth muscle, a negative control, lane 5 is blank, and the remaining lanes are samples of placenta collected at the times indicated. The highest levels of cyclin E are found between 17 and 36 weeks gestation. The observation of two protein bands is consistent with the presence of two forms of cyclin E, 52 and 50 kilodaltons.

columns [Figure l(g)]. Although there was some immuno- reactivity in occasional villous cytotrophoblast cells, the strongest reactivity was in the syncytiotrophoblast cells. How- ever, with increasing gestational age, immunoreactivity shifted from the syncytiotrophoblast to the cytotrophoblast cells [Fig- ure l(h)]. This pattern then continued through the remainder of the second trimester. In the term placenta, many cells, both cytotrophoblasts and syncytiotrophoblasts, were immuno- reactive for cyclin E [Figure l(i)]. Also obvious were posi- tively stained cells within the villous core; the identity of which

Page 6: Differential expression of g1, cyclins during human placentogenesis

14

remains uncertain. In contrast to cyclin D3 immunostaining, the intensity of cyclin E staining is fairly uniform throughout gestation. To verify the specificity of the immunocytochemis- try results, we immunoadsorbed the cyclin E antibody with control peptide before placing it on the slide. The staining pattern that we obtained from the immunoadsorbed antibody was indistinguishable from that seen with the non-immune negative control serum [Figure l(j)].

The presence of such a high number of cyclin E immuno- reactive cells throughout gestation prompted a biochemical analysis of the protein(s) which was reacting with the cyclin E polyclonal antibody. Specifically, we wanted to determine the number of protein species which were being identified by this antibody. Thus, protein samples were isolated from placentae throughout gestation and subjected to a Western blot analysis using the same anti-cyclin E antibody that was used for the immunocytochemistry. The results from these experiments demonstrated two major species of proteins which were immunoreactive with the Santa Cruz antibody (Figure 4), one at approximately 50 kDa and the other at approximately 52 kDa. These sizes correspond to the EL (~52) and ES (~50) forms of the protein and most likely represent true cyclin E immunoreactive protein in the placental samples (Ohtsubo et al., 1995). No other protein species was detectable by Western blot analysis.

Placenta (1997), Vol. 18

expression for cyclin Dl; in the endothelial cells lining the villous blood vessels. The presence of cyclin Dl protein in the earliest detectable blood vessels suggests participation in the process of angiogenesis. However, it should be noted that mice completely lacking a functional cyclin Dl gene are not only viable but fertile as well. The phenotypes associated with a cyclin Dl null mutation in mice included jaw malformations, retinopathy, a failure to lactate, and growth retardation (Fantl et al., 1995). The cause of the growth retardation (between 10 and 40 per cent) was not investigated, though it could be explained by impaired placental angiogenesis.

Fetal macrophages, or Hofbauer cells, are easily identified in placental villi from about 18 days p.c. to 4 or 5 months of gestation. Because of stromal compression, their numbers ap- pear to decrease with increasing gestation; however, there are still numerous CD68+ cells present. In this manuscript we demonstrate cyclin D2 immunoreactivity is restricted to Hofbauer cells in the first and second trimester placentae. The observation that cyclin D2 antibody does not react with the Hofbauer cells present in term placentae implies that whatever functional role cyclin D2 is playing in the macrophage lineage may no longer be required at this point in placental develop- ment. In other words, cyclin D2 function may be restricted to both a lineage, as well as to a developmental stage. The notion of functional heterogeneity in the fetal macrophage cells had been proposed by others as well, based on detailed morphological observations (Castellucci et al., 1987).

Of the D-type cyclins only cyclin D3 was present in villous cytotrophoblast cells throughout gestation. The pattern of cyclin D3 staining correlates well with that of a cell cycle regulatory molecule in this lineage, suggesting that cyclin D3 immunoreactivity may serve as a useful marker for cytotro- phoblast division. Proliferating cell nuclear antigen (PCNA), another cyclin-like molecule, which is an auxiliary factor for DNA polymerase 8, has also been used to ascertain the distribution of cytotrophoblast cells engaged in the cell cycle in both human (Wolf and Michalopoulos, 1992) and nonhuman I&rates (Blankenship and King, 1994). In both of these studies, PCNA protein was easily detected in cytotrophoblast nuclei, albeit at numbers greater than those reported for i’H]-thymidine labelling, and greater than labelling for Ki-67 (Blankenship and King, 1994), another S-phase marker of proliferation. This discrepancy in cell labelling is easily explained by the fairly lengthy half-life for the PCNA protein, which is approximately 20 h (Bravo and Macdonald-Bravo, 1987). As a consequence of residual PCNA, cells which have already exited the cell cycle will still be labelled. Hence, D3 cyclin presence may reflect more accurately the number of cycling cells.

Unlike the D-type cyclins, which show lineage specific expression, cyclin E appears to be essential for progression of late G, and the transition to the S phase. Human cyclin E is present in at least two, and perhaps three forms, which result from alternative splicing of the messenger RNA (Sewing et al., 1994; Ohtsubo et al., 1995). The significance of the multiple forms is not clear. Ohtsubo et al., ( 1995) claim that both of the

DISCUSSION

The mechanisms regulating cell cycle progression and cess- ation are quite complex and not fully understood in mammalian cells. A large body of knowledge exists regarding the numerous growth factors, cytokines and receptors present at the maternal- fetal junction. Virtually all of these external mitogenic factors act at the G, phase of the cell cycle. We undertook this study to address the question of cell-cycle control in villous trophoblast populations. Immunocytochemical analysis of four cyclins present in the G, phase of the cell cycle demonstrated that tiic D-type and E cyclins are differentially and sometimes retiun dantly expressed. Although this study was intended to investi- gate cyclin expression in trophoblast cells throughout normal development, there were obvious restrictions to obtaining early third trimester samples. Nonetheless, data generated from these studies has illuminated two G, cyclin proteins as potential key regulators of trophoblast proliferation; cyclins D3 and E. In addition, the presence of cgclin Dl and D2 proteins in non- trophoblast cell lineages suggests that they may play a role regulating proliferation within these populations.

The association of a D cyclin with its catalytic partner, cdk4 or cdkb, is an early event in cell cycle initiation (Matsushime et al., 1992; Xiong, Zhang and Beach, 1992; Bates et al., 1994). Cyclins Dl, D2, and D3 have overlapping, yet unique patterns of expression both in transformed tumor cell lines (Inaba et al., 1992) and in differentiating hematopoietic cell lines, (Ando, Ajchenbaum-Cymbalista and Griffin, 1993; Kato and Sherr, 1993). In this report we demonstrate a unique pattern of

Page 7: Differential expression of g1, cyclins during human placentogenesis

DeLoia, Burlingame and Krasnow: Cyclin Expression in the Placenta

forms appear to be functional; when over-expressed in vitro, both forms will decrease the time necessary to complete G, phase (Ohtsubo and Roberts, 1993) and both forms associate with the cdk2 kinase. In those mammalian cells studied, the larger form of cyclin E was most abundant, but the faster migrating form was also present in vivo. In contrast, Sewing et al., (1994) reported that the smaller, faster migrating, form of cyclin E does not bind to and activate cdk2. The discrepancy may be due to differences in the cell lines examined. In addition, there has been a report of a third form of cyclin E which results from phosphorylation of the 52-kDa peptide and consequent activation of the cyclin E-Cdk2 complex (D&c, Lees and Reed, 1992). In this study, we have shown that within the placenta there are two prominent protein products, and that the slower migrating form is most abundant through- out gestation. Investigation into the identity and functional activity of the observed placental cyclin E forms is being pursued, especially in the first trimester immunoreactive syncytiotrophoblast.

The question as to why there is so much cyclin E nuclear staining in these first trimester syncytiotrophoblast cells, a non-proliferative cell type, remains. Although cyclin E does not contain a ‘destruction motif, cyclin E mRNA levels do vary approximately eightfold during the cell cycle (Lew, Dulic and Reed, 1991). Also, the cyclin E protein contains a PEST sequence, which targets proteins for rapid degradation (Rogers, Wells and Rechsteiner, 1986). Therefore, the pres- ence of residual cyclin E would not be anticipated, unless it was sequestered in another protein-protein complex. Recently, a new class of cell cycle regulatory proteins, designated mitotic inhibitors, has been described (El-Deiry et al., 1993; Gyuris et al., 1993; Harper et al., 1993). There are two families of

ACKNOWLEDGEMENTS

15

inhibitors; the INK4 family, which includes ~15, ~16, ~18, and ~19; and the ClP/KlP family, whose members include ~21, ~27, and ~57. INK4 members function by competing directly with cyclins for cdk binding, while ClP/KlP inhibitors bind directly to the cyclin or cyclin-cdk complexes. All three of the CIP family members are able to bind and inactivate the cyclin E-cdk2 complex. Therefore, in the presence of a bound inhibitor, cyclin E-cdk2 complexes could still be in abundance, but have no enzymatic activity.

There is precedence for such a mechanism in both senescent and differentiated cells. Diploid human fibroblasts have a finite number of cell divisions, after which they remain alive but are no longer capable of responding to mitogenic stimulation (Goldstein, 1990). When Dulic et al., (1993) ascertained expression of cyclin genes in these senescent cells, they found a lo- to IS-fold increase in levels of cyclin Dl and cyclin E. However, even though cyclin E protein was associated with the appropriate cyclin-dependent kinase, the cyclinE-cdk2 com- plex was not active. In a similar study by Lucibello et al., (1993) the authors reported that although cyclin E did rise dramatically with senescence, there was a concurrent decrease in cdk2 levels, which could be explained by inhibitor binding to the cyclin protein and not to the cyclin-cdk complex. Burger et al., (1994) demonstrated further, that persistent cyclin presence can also occur in terminally differentiated cells; in this case, myeloid cells, and suggested that differentiation and cyclin presence were not incompatible. Syncytiotropho- blast cells are a highly differentiated cell population, incapable of any further division, but functionally quite active. It is probable that the cyclin protein present in these cells has been inactivated upon terminal differentiation, potentially through binding to a cell-cycle inhibitor.

This work was supported in part by IRG-58-33 from the American Cancer Society and the Magee-Womens Health Foundation

REFERENCES

Ando, K., Ajchenbaum-Cymbalista, F. & Griffin, J. D. (1993). Regulation of Gl/S transition by cyclins Dl and D3 in hematopoietic cells. Proceedings of the National Academy of Sciences of the USA, 90, 9571-9575.

Bates, S., Bonetta, L., MacAllan, D., Parry, D., Holder, A., Dickson, C. & Peters, G. (1994). CDK5 (PLSTIRE) and CDK4 (PSK-J3) are a distinct subset of the cyclin-dependent kinases that associate with cyclin Dl. Oncogene, 9, 71-79.

Blankenship, T. N. & King, B. F. (1994). Developmkntal expression of Ki-67 antigen and proliferating cell nuclear antigen in macaque placentas. Developmeml Dynamics, 201, 326333.

Bravo, R. & Macdonald-Bravo, H. (1987). Existence of two populations of cyclim’proliferating cell nuclear antigen during the cell cycle: Association with DN.4 replication sites. 3oumal of Cell Biology, 105, 1549-1554.

Burger, C., Wick, M. & Muller, R. (1994). Lineage-specific regulation of cell cycle gene expression in differentiating myeloid cells. 3oumal of Cell Science, 107, 2047-2054.

Castellucci, M., Celona, A., Bartels, H., Steininger, B., Benedetto, V. & Kaufman, P. (1987). Mitosis of the Hofbauer cell: possible implications for a fetal macrophage. Placenta, 8, 65-76.

Dou, Q-P., Levin, A. H., Zhao, S. & Pardee, A. B. (1993). Cyclin E and Cyclin A as candidates for the restriction point protein. Cancer Research, 53, 1493-1497.

Dulic, V., Drullinger, L. F., Lees, E., Reed, S. 1. & Stein, G. H. (1993). Altered regulation of G, cydins in senescent human diploid fibroblasts: accumulation of inactive cycling E-Cdk2 and cyclin Dl-Cdk2 complexes. Proceedings of the National Academy of Sciences of the USA, 90, 11034- 11038.

Dulic, V., Lees, E. & Reed, S. I. (1992). Association of human cyclin E with a periodic Gl-S phase protein kinase. Science, 257, 1958-1961.

El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W. & Vogelstein, B. (1993). WAFl, a potential mediator of p53 tumor suppression. Cell, 75, 817-825.

Evans, T., Rosenthal, E. T., Youngblom, J., Distel, D. &Hunt, T. (1983). Cyclin: a protein specified by maternal mRNA in sea urchin eggs that is destroyed at each cleavage division. Cell, 33, 389-396.

Fantl, V, Stamp, G., Andrew, A., Rosewell, I. & Dickson, C. (1995). Mice lacking cyclin Dl are small and show defects in eye and mammary gland development. Genes and Development, 9, 2364-2372.

Girard, F., Strausfeld, U., Fernandez, A. & Lamb, N. J. C. (1991). Cyclin A is required for the onset of DNA replication in mammalian fibroblasts. Cell, 67, 1169-1179.

Goldstein, S. (1990) Replicative senescence: the human fibroblast comes of age. Science, 249, 1129-l 133.

Gyuris, J., Golemis, E., Chertkov, H. & Brent, R. (1993). Cdil, a human Gl and S phase protein phosphatase that associates with cdk2. Cell, 75, 791-803.

Page 8: Differential expression of g1, cyclins during human placentogenesis

16

Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K. & Elledge, S. J. (1993). The p21 Cdk-interacting protein Cipl is a potent inhibitor of Gl cyclin-dependent kinases. Cell, 75, 805-816.

Inaba, T., Matsushime, H., Valentine, M., Roussel, M. F., Sherr, C. J. & Look, A. T. (1992). Genomic organization, chromosomal localization and independent expression of human cyclin D genes. Genomics, 13, 565-574.

Kato, J.-Y. & Sherr, C. J. (1993). Inhibition of granulocyte differentiation by Gl cyclins D2 and D3 but not Dl. Proceedings of the National Academy of Sciences of the USA, 90, 11513-11517.

Koff, A., Giordano, A., Desai, D., Yamashita, K., Harper, J. W., Elledge, S., Nishimoto, T., Morgan, D. O., Franza, B. R. & Roberts, J. M. (1992). Formation and activation of a cyclin E-cdk2 complex during the GI phase of the human cell cycle. Science, 257, 1689-1694.

Lew, D. J., Dulic, V. & Reed, S. I. (1991). Isolation of three novel human cyclins by rescue of Gl cyclin (Cln) function in yeast. Cell, 66, 1197-1206.

Lucibello, F. C., Sewing, A., Brusselbach, S., Burger, C. & Muller, R. (1993). Deregulation of cyclins Dl and E and suppression of cdk2 and cdk4 in senescent human fibroblasts. 3ournal Cell Science, 105, 123-133.

Matsushime, H., Ewen, M. E., Strom, D. K., Kato, J. Y., Hanks, S. K., Roussel, M. F. & Sherr, C. J. (1992). Identification and properties of an atypical catalytic subunit (~34 PSK-J3/cdk4) for mammalian-D type Gl cyclins. Cell, 71, 323-334.

Nurse, P. (1990). Universal control mechanism regulating onset of M-phase. Nature, 344, 503-508.

O’Farrell, P. H., Edgar, B. A., Lakich, E. & Lehner, C. F. (1989). Directing cell division during development. Science, 246, 635-640.

Placenta (1997), Vol. 18

Ohtsubo, M. & Roberts, J. IM. (1993). Cyclin-dependent regulation of G, in mammalian fibroblasts. Sczence, 259, 1908-1912.

Ohtsubo, M., Theodoras, A. M., Schumacher, J., Roberts, J. M. & Pagano, M. (1995). Human cyclin E, a nuclear protein essential for the G,-to-S phase transition. Molecular end Cellular Biology, 15, 2612-2624.

Pardee, A. B. (1989). G, events and regulation of cell proliferation. Science, 246, 603-608.

Rogers, S., Wells, R., & Rechsteiner, M. (1986). Amino acid sequences common to rapidly degraded proteins: the PEST hypothesis. Science, 234, 3644368.

Sewing, A., Ronicke, V., Burger, C., Funk, M. & Mullet-, R. (1994). Alternative splicing of human cyclin E. 3ournal of Cell Scieme, 107, 581-588.

Tedde, G. & Tedde-Piras, A. (1978). Mitotic index of the Langhans’ cells in the normal human placenta from the early stages of pregnancy to the term. Acta Anatomy, 104, 114-l 19.

Tsai, L.-H., Takahashi, T., Cavinessn, V. S. & Harlow, E. (1993). Activity and expression pattern of cyclin-dependent kinase 5 in the embryonic mouse nervous system. Development, 119, 1029-1040.

Tsai, L. H., Delalle, I., Caviness, V. S., Char, T. & Ha&w, E. (1994). p35 is a neural-specific regulatory subunit of cyclin-dependent kinase 5. Nature, 371, 419-426.

Wolf, H. K. & Michalopoulos, G. K. (1992). Proliferating cell nuclear antigen in human placenta and trophoblastic disease. Pediatrzc Pathology, 12, 147-154.

Xiong, Y., Zhang, H. & Beach, D. (1992). D type cyclins associate with multiple protein kinases and the DNA replication and repair factor PCNA. Cell, 71, 505-514.