| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Models |
From The University of Texas M. D. Anderson Cancer
Center,*
Science Park-Research Division, Smithville, Texas;
and the Department of Cell and Molecular Biology,*
Centro de
Investigaciones Enérgeticas, Mediambientales y
Tecnológicas,
Madrid, Spain
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cyclin D1 and cyclin D2 seem to contribute to the neoplastic phenotypes in human and mouse tumors. Indeed, the cyclin D1 gene was originally cloned as an oncogene termed PRAD1, that was activated by chromosomal translocations present in parathyroid adenoma.13 Cyclin D2 overexpression or amplification was also described in several tumors. In fact, cyclin D2 accumulation in the cytoplasm of gastric carcinoma cells seems to play a role in cancer progression.14 Also, the overexpression of cyclin D2 in carcinoma in situ identified it as a candidate gene in male germ-cell malignancies.15,16 Cyclin D2 is also overexpressed in chronic B-cell malignancies.17 Although, fewer reports suggest that cyclin D3 plays a role in tumorigenesis, cyclin D3 overexpression was associated with increased expression of p27Kip1 in a subset of aggressive B-cell lymphomas.18 In addition, coordinated elevation of cyclin D3 and cyclin D1 was observed in the breast cell line MCF-7.19 These data suggest that normal progression through the G1 phase requires distinct sets of D-type cyclins in different tissues. For example, G1/S transition in hematopoietic cells and inhibition of granulocyte differentiation are regulated by cyclin D2 and D3 whereas cyclin D1 seems to be dispensable in these cell types.20,21 In contrast cyclin D3, but not D1 or D2, was up-regulated on induction of HL-60 leukemia cells to differentiation and has been shown to accumulate at high levels in a wide range of quiescent cell types.22
The function of D-type cyclins has also been studied in D-type cyclin-deficient mice. Cyclin D1 knockout mice presented symptoms of neurological impairment as well as deficient development of retina and mammary glands.23,24 Cyclin D2-deficient mice, showed alterations in gonadal cell proliferation.25 At present, generation of cyclin D3-deficient mice has not been reported. Taken together, these data suggest tissue-specific functions of D-type cyclins.
The murine skin model is a valuable system for studying epidermal proliferation, precancerous changes, and tumor progression in vivo.26 The use of this model allowed us to detect expression of the three D-type cyclins in normal, hyperplastic and neoplastic epidermis in vivo.27,28 Cyclin D1 and cyclin D2 are expressed at the mid-G1 phase and form complexes with CDK4/6, a process that is dependent on the relative abundance of these cyclins. Cyclin D3 also forms complexes with CDK4/6; however, complex formation requires an additional regulatory event other than the simple relative abundance of these proteins.29 Another interesting difference is that during premalignant tumor progression in chemically-induced mouse carcinogenesis, cyclin D1 and cyclin D2 are overexpressed and form complexes with CDK4/6, whereas cyclin D3 only forms complexes with CDK4/6 in hyperproliferative epidermis (hyperplastic skin).27
We have previously reported the generation of a transgenic mouse that expresses human cyclin D1 in squamous epithelial tissues, resulting in moderate epidermal and severe thymic hyperplasia.30,31 To complete the study of the role of D-type cyclins in squamous epithelial tissues, we generated transgenic mice that expressed either cyclin D2 or cyclin D3, driven by the regulatory sequence of bovine keratin 5 (K5D2 and K5D3). We determined that overexpression of either cyclin D2 or cyclin D3 results in hyperproliferative epidermal hyperplasia. However, there was a clear difference in the thymic phenotypes. Thymic hyperplasia in the cyclin D2 transgenic mice regressed spontaneously in older mice, in contrast to the hyperplasia in the cyclin D1 transgenic mice which was progressive and fatal.30 At the other end of the spectrum are the cyclin D3 mice, which did not develop thymic hyperplasia. Thus, whereas overexpression of cyclin D1, D2, and D3 produces a similar epidermal phenotype, each D-type cyclin transgene induces a unique thymic phenotype.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An EcoRV/XbaI fragment containing the mouse cyclin D2 or cyclin D3 cDNA was excised from the plasmid pBluescript II and introduced into the polylinker of the vector pBK5 which contained the 5.2-kb bovine keratin 5 (K5) regulatory sequences, ß-globin intron 2 and the 3' polyadenylylation sequences. These constructs were designated as pK5D2 and pK5D3. The transgenes were excised from the plasmid vector by digestion with BssHII, separated by low-melting-point agarose electrophoresis and purified using a Geneclean II Kit (BIO101, Vista, CA). These transgenes were microinjected into the C57BL/6xSJL hybrid strains, which took place in the National Institute of Child Health and Human Development National Transgenic Mouse Development Facility (NTMDF) at the University of Alabama, Birmingham. Transgenic mice were crossed for two generations with the SSIN strain to generate 75% SSIN background mice.
Transgenes DNA-Specific Polymerase Chain Reaction (PCR)
Genomic DNA was extracted from mouse tail clips and used for PCR detection of the transgenes. We used an upstream primer (5'TTCAGGGTGTTGTTTAGAATGG3') and a downstream primer (5'CAATAAGAATATTTCCACGCCA3') specific for the ß-globin intron 2 sequence. With this process, we screened the entire transgenic mouse lines. The DNA amplification renders a 450-bp PCR product. PCR was performed by denaturation at 95°C for 1 minute, followed by 32 cycles of amplification as follows: denaturation at 95°C for 30 seconds, annealing at 55°C for 40 seconds, and extension at 72°C for 45 seconds, with a final extension at 72°C for 10 minutes.
Cyclin D2 and Cyclin D3 Immunohistochemical Staining
Immunohistochemical staining of formalin-fixed paraffin-embedded tissues was performed with polyclonal mouse cyclin D2 (M20) or cyclin D3 (C16) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Epithelial cell proliferation was measured by intraperitoneal injection of BrdU (60 µg/g; Sigma Chemical Co., St. Louis, MO) 30 minutes before the mice were killed. BrdU incorporation was detected by immunohistochemical staining of paraffin-embedded sections with mouse anti-BrdU monoclonal antibody (Becton-Dickinson Immunocytometry System; Becton-Dickinson, San Jose, CA). The reaction was visualized with a biotin-conjugated anti-mouse antibody (Vector Laboratories, Inc., Burlingame, CA) and avidin-biotin-peroxidase kit (Vectastain Elite, Vector Laboratories, Inc.) with diaminobenzidine as chromogen.
Western Blotting Analysis, Immunoprecipitation, and Kinase Assay
Mouse dorsal skins were treated with a depilatory agent for 1 minute and then washed. After mice were sacrificed, the epidermal tissue was scraped off with a razor blade, placed into homogenization buffer (50 mmol/L HEPES, pH 7.5, 150 mmol/L NaCl, 2.5 mmol/L EGTA, 1 mmol/L ethylenediaminetetraacetic acid, 0.1% Tween-20, 1 mmol/L dithiothreitol, 0.1 mmol/L phenyl methyl sulfonyl fluoride, 10 mmol/L ß-glycerophosphate, 0.2 mmol/L sodium vanadate, and 2 mmol/L NaF) and homogenization was achieved with a manual homogenizer. The epidermal homogenate was centrifuged at 11,000 x g to collect the supernatant which was used directly for Western blotting analysis or stored at -70°C. Thymic proteins were extracted by using the same buffer and conditions as stated above. The protein concentration in each skin or thymus lysate was measured with the Bio-Rad protein assay system (Bio-Rad Laboratories, Richmond, CA). Protein lysates (25 µg from each sample) were electrophoresed through 12% acrylamide gels and electrophoretically transferred onto nitrocellulose membranes. After being blocked with 5% nonfat powdered milk in Dulbecco’s phosphate-buffered saline (Sigma Chemical Co.), the membranes were incubated with 1 µg/ml of specific antibodies. The following antibodies were used: polyclonal antibodies against cyclin D2 (M-20), cyclin D3 (C-16), pRb (C15), p107 (C18), and p130 (C20) (Santa Cruz Biotechnology, Inc.). pRb (G3-245) (Pharmingen, San Diego, CA.) was used for Western blot analysis of thymus proteins. Horseradish peroxidase-conjugated secondary antibody (Amersham Corp., Arlington Heights, IL), followed by enhanced chemiluminescence (ECL detection kit; Amersham Corp.) were used for immunoblotting detection. Bio-image analysis was used to quantitate the expression levels of those proteins.
To study cyclin D/CDK complex formations and kinase activities, we used
polyclonal anti-CDK4 and anti-CDK2 antibodies conjugated with protein
A-Sepharose beads (Life Technologies Inc., Grand Island, NY) to
immunoprecipitate fresh protein lysates for 1 hour at 4°C with
constant rotation. After washing three times with extraction buffer,
Western blot analysis was performed as described above with polyclonal
antibody described previously. To study the kinase activities of CDK4
and CDK2, protein lysates were obtained as described above, but the
homogenate was frozen on powdered dry-ice, thawed in ice water,
incubated on ice for 15 minutes and centrifuged at 10,000 x
g for 10 minutes at 4°C. The supernatant was collected and
used for a kinase assay. Eight hundred micrograms of protein lysate
were immunoprecipitated with antibodies against CDK4 or CDK2. Thirty
µl of precoated antibody beads (Life Technologies Inc.) was incubated
with the lysate for 1 hour at 4°C. The beads were washed twice with
homogenization buffer and twice with kinase buffer (50 mmol/L HEPES, pH
7.5, and 10 mmol/L MgCl2). Then, 30 µl of
kinase buffer, 0.5 µg of pRb substrate (Santa Cruz Biotechnology,
Inc.), 5 µCi [
-32P]-ATP (6,000 Ci/mmol),
2.5 mmol/L EGTA, 1 mmol/L dithiothreitol, 20 µmol/L ATP, 10 mmol/L
ß-glycerophosphate, 0.2 mmol/L sodium vanadate, and 2 mmol/L NaF) was
added to the bead pellet and incubated for 30 minutes at 30°C. Sodium
dodecyl sample buffer was added, and each sample was boiled for 5
minutes and electrophoresed through a 10% acrylamide gel.
Flow Cytometry
Thymocytes were obtained by pressing thymic tissue through a nylon mesh. Thymocytes were stained with anti-CD4-coupled phycoerythrin and anti-CD8-coupled fluorescein isothiocyanate and analyzed by two-color immunofluorescence with a Coulter Elite Flow cytometer as previously described.30
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The construct used to generate transgenic mice is depicted in
Figure 1A
. The expression of D-type
cyclins was targeted to stratified epithelia by the 5'-regulatory
fragment of the bovine K5 gene. The K5-cyclin D2 and K5-cyclin D3
vectors were made by subcloning the mouse cyclin D2 or cyclin D3 cDNA
into a vector containing a 5.2-kb fragment of the bovine K5 promoter,
the rabbit ß-globin intron 2, and the SV40 polyadenylation signal. As
reported, this fragment drives expression of a reporter gene in basal
cells of squamous stratified epithelia, where K5 is normally
expressed.32,33
All of the transgenic mice were generated
in the C57BL/6xSJL genetic background. Three mice with cyclin
D3-positive and five mice with cyclin D2-positive integration were
identified by PCR analysis (Figure 1B)
. Based on those results, the
integration-positive mice were selected as founders and crossed with
SSIN inbred mice. A second screening to verify transgene expression was
performed by Western blot analysis of epidermal preparations with
cyclin D2 or cyclin D3 antibodies as described.29
These
results permitted the selection of two mice of each D-type cyclin
transgene as founders of high expression lines (2101 and 2102 of K5D2,
2201 and 2203 of K5D3 mice).
|
To quantify the level of cyclin D2 and cyclin D3 protein
expression, we isolated the epidermis of transgenic and normal siblings
for immunoblot analysis. The cyclin D3 protein is expressed at high
levels in both K5D3 lines compared to their normal siblings (fourfold
in 2201 and 5.5-fold in 2203 transgenic lines) (Figure 2)
. Increased levels of cyclin D2 protein
were observed in both K5D2 lines, although the level of expression was
barely twofold higher than the level observed in their normal siblings
(Figure 2)
. It is worth mentioning that the protein level of cyclin D3
decreases in K5D2 mice and the cyclin D2 protein level decreases in
K5D3 animals. The effect was most notable in K5D3 animals (lines 2201
and 2203) where the level of cyclin D2 expression was half compared to
the wild-type animals (Figure 2)
.
|
. Both pRb
and p107 protein expression were clearly increased in 2102-K5D2 and
both K5D3 transgenic lines. The transgenic line 2203-K5D3 showed
a strong induction of p107 as well as changes in mobility consistent
with phosphorylation. Similar phosphorylation was also found in the
transgenic line 2102-K5D2. No apparent change in phosphorylation of pRb
was observed. Increased p130 protein was detected in both K5D2 and K5D3
transgenic lines, whereas the 2203-K5D3 line again showed a stronger
induction of this protein (Figure 2)
. These data are consistent with
the stronger expression of cyclin D22102 or cyclin
D32203 in these transgenic animals and with our
previously reported data of elevated levels of p107 and pRb protein in
mouse epidermal tissue after proliferative induction by
TPA.29 Development of Epidermal Hyperproliferation and Hyperplasia in Transgenic Mice
The newborn cyclin D2 and cyclin D3 transgenic mice did not
demonstrate any obvious developmental abnormalities and there was no
difference in size compared to wild-type littermates. Cytokeratin 5 is
normally expressed in the basal cell layer of the epidermis. Consistent
with this, immunohistochemical staining showed high expression of
cyclin D2 or cyclin D3 in the epidermis of the respective transgenic
mice (Figure 3C
and Figure 4C
). Because the presence of cyclin D2
and D3 expressing cells is difficult to detect in wild-type epidermis
the high level of expression detected in the K5D2 and K5D3 lines
confirms the results obtained by Western blot and PCR analysis (Figures 1 and 2)
.
|
|
. However, no gross phenotype was
evident in the hair follicles, which showed normal distribution and
morphology. The increased number of nucleated cells in K5D2 and K5D3
mice is similar to that of previously described K5D1-7111 transgenic
mice30
(Figure 5B)
. However, direct comparison with the
earlier report was not possible because of the different genetic
backgrounds. For this reason, these parameters where re-evaluated after
crossing the K5D1 transgene into the same genetic background as the
K5D2 and K5D3 mice. Expression of each cyclin D transgene in the same
genetic background results in a significant increase in the number of
nucleated cells of the interfollicular epidermis compared to normal
siblings (2101, P = 0.001; 2102, P <
0.008; 2201, P < 0.0003; 2203, P <
0.02; 7111, P < 0.01, Mann-Whitney) (Figures 3, 4, and 5)
. Consistent with the epidermal hyperplasia, hyperproliferation of
the interfollicular epithelia was observed by BrdU incorporation as a
marker of cells in S phase (Figures 3E, 4E, and 5A)
. Compared to
nontransgenic littermates, there was a 1.7- and 2.2-fold increase
respectively in the number of BrdU-positive cells in K5D2 transgenic
lines 2101 and 2102. The K5D3 transgenic lines, 2201 and 2203, also
showed a 2.9- and 2.5-fold increase in BrdU-positive cells and the
K5D1-7111 line showed a mild increase of 2.5-fold (2101,
P = 0.02; 2102, P = 0.008; 2201,
P = 0.001; 2203, P = 0.05; 7111,
P = 0.0001, Mann-Whitney) (Figure 5A)
. To complete the
description of interfollicular epidermis, we determined the thickness
of the epidermis in transgenic and wild-type animals (Figure 5C)
. Each
of the transgenic lines with the exception of K5D3-2201 demonstrated an
increase in epidermal thickness. A direct comparison with K5D1-7111
showed that, with the exception of the K5D3-2201 line, the increase of
any of the three D-type cyclins produced a similar phenotype in mouse
epidermis. Table 1
summarizes the changes
in the three principal parameters (BrdU label Index, number of
nucleated cells and epidermal thickness) that define the epidermal
phenotype of cyclin D1, D2, and D3 transgenic lines. To study whether
D-type cyclin expression affects the normal pattern of epidermal
differentiation, immunohistochemical staining was performed to detect
K1, which is expressed only in terminally differentiated cells, and K5,
which is expressed in basal epithelial cells. The expression pattern of
these keratins was comparable to nontransgenic controls (data not
shown). However, the transgenic epithelia have a more densely packed
basal and suprabasal cell compartment (basal cell hyperplasia)
reflecting an expanded proliferative compartment.
|
|
Three of the five founders of K5D2 transgenic animals died at 5
months of age. Autopsy revealed a severe hyperplasia of the thymus. Two
of these founders died before transgenic lines could be established.
The remaining founder, 2102, was crossed with SSIN inbred mice and a
transgenic line was established. To study the time frame of development
of hyperplasia, K5D2-2102 mice were sacrificed at intervals from 2 to
30 weeks of age. The maximum thymus weight of the normal siblings
reached 0.11g at 8 weeks as previously reported.30
The
maximum thymus weight in the cyclin D2 transgenic mice was 0.24 g
at 12 weeks of age which represents an increase of 4.5-fold compared to
wild-type animals (Figure 6, A and B)
.
After this time point, the thymic weight declines and at 20 weeks both
the transgenic and wild-type thymi are similar in size (Figure 6A)
. To
determine cyclin D2 expression levels, thymic extracts were analyzed by
Western blot. Consistent with the thymic hyperplastic phenotype, the
K5D2-2102 thymi expressed five times more cyclin D2 than the wild-type
animals at 12 and 24 weeks of age (Figure 6C)
. On the other hand, the
K5D2-2101 line did not develop thymic hyperplasia and the level of
protein expression was only twofold greater than the wild type.
Notably, cyclin D2 protein expression did not decrease at 24 weeks of
age in the 2102 line when the thymic involution had occurred (Figure 6C
, lines 2 and 3).
|
shows a similar level of association
between CDK4 and cyclin D2 at 12 and 20 weeks of age in transgenic
mice. Although CDK4/cyclin D2 levels were barely detectable in the
wild-type thymus, these complexes were apparent when the Western blot
membranes were overexposed (data not shown). The
p27Kip1 inhibitor complexed with CDK4 in both
transgenic and wild-type mice, although there was a slight decrease in
the level of CDK4-associated p27Kip1 in
20-week-old wild-type thymi (Figure 7A)
. Interestingly,
p21Cip1 was clearly associated with CDK4 in
transgenic, but not in wild-type mice (Figure 7A)
. As expected cyclin
D2 did not co-precipitate with CDK2, whereas high levels of
p27Kip1 were associated with CDK2 in transgenic
and wild-type thymi. Furthermore, there was a decrease in the level of
p27Kip1 associated with CDK2 in 20-week wild-type
thymi (Figure 7A)
. These results suggest that neither
p27Kip1 nor p21Cip1 are
involved in the regression of thymi hyperplasia by blocking CDK4 or
CDK2 activities at 20 weeks. p16Ink4a and
p15Ink4b inhibitors were not detected in CDK4
complexes (data not shown). We also analyzed the kinase activity of
CDK4 and CDK2 complexes in vitro using pRb as a substrate.
Figure 7B
shows an increase in CDK4 kinase activity at 20 weeks in
transgenic and wild-type thymi. A similar pattern was observed for CDK2
activity in wild-type thymi. However, the kinase activity in cyclin D2
transgenic thymi was greater at 12 than at 20 weeks. The elevated
kinase activity in young thymi could be related to the observed
hyperplasia. However, we cannot rule out the possibility that the
different cell populations that compose the thymus could mask the CDK
activities of the epithelial component (which overexpress cyclin D2).
In addition, pRb protein levels and phosphorylation states were
determined in thymus extracts of transgenic and wild-type mice. Figure 7C
shows that the pRb level is elevated in transgenic mice compared
with normal siblings. Also, a slower migrating band consistent with
hyperphosphorylation is observed in transgenic mice at 20 weeks.
Overexposure of the membrane showed that pRb protein levels also
increased in wild-type mice at 20 weeks of age, although the protein
level was still inferior compared to transgenic mice (Figure 7C)
.
Altogether, these results indicate that CDK4 and CDK2 kinase activities
are not reduced during the thymic size regression in transgenic mice
and neither p27Kip2 nor
p21Cip1 seemed to be involved in this event.
|
. Consistent with these data none of the three
K5D3 founders developed thymus hyperplasia as was established for
cyclin D130
and cyclin D2. To determine whether expression of the cyclin D2 transgene in thymic epithelium altered T cell development, we analyzed the distribution of thymocyte subsets in normal and in transgenic mice at 12 weeks of age, when the thymus was hyperplastic. Flow cytometry determination showed that each major thymocyte subset, defined by CD4 and CD8 expression, is present in 2102 transgenic mice at 12 weeks of age (thymic hyperplasia) (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
94% between human and
mouse D-type cyclins) suggest that each family member has important and
unique roles in cell-cycle regulation that constrain evolutionary
divergence. Indeed, individual D-type cyclins have a greater similarity
between mouse and human species (94% identity) than among themselves
in the same species (cyclin D1/cyclin D2, 64%; cyclin D1/cyclin D3,
49%; cyclin D2/cyclin D3, 64%).35-37
The three D-type
cyclins are differentially expressed in normal mouse
epidermis,.27,29
Cyclin D1 protein expression and
cyclinD1-CDK4/6 complex formation are strongly regulated during the
G1 phase after
12-0-tetradecanoylphorbol-13-acetate (TPA) stimulation of
keratinocytes.29
On the other hand, cyclin D2 and cyclin
D3 protein expression seem to be constitutive and only minor changes in
their protein levels were observed after TPA stimulation. Cyclin D1 and
cyclin D2 overexpression or amplification have been described in
several human and mouse tumors and were considered as candidates for
oncogenes.27,38-40
To study the role of each D-type cyclin in normal proliferation and
tumorigenesis in mouse skin, we have developed two new transgenic mice
that express cyclin D2 or cyclin D3 in squamous epithelial tissues.
Previously, we described the generation of a transgenic mouse that
expressed human cyclin D1 under the regulatory sequence of keratin 5
(K5D1) directing expression to basal cells in squamous epithelial
tissues.30
The principal phenotypic characteristics of the
cyclin D1 transgenic animals were mild epidermal hyperplasia and severe
thymic hyperplasia in three integration events of different genetic
backgrounds.30
The cyclin D2 and cyclin D3 transgenic mice
express the respective transgene under the same keratin 5 promoter. The
K5D2 and K5D3 showed a similar epidermal phenotype including an
increased number of nucleated cells, epidermal thickness, and elevated
labeling index. BrdU incorporation showed that in S phase, cells were
present not only in basal but also in the suprabasal layers. The
epidermal cells are more densely packed than in normal siblings,
however, no differences in the cell size were detected. Comparison with
K5D1 transgenic mice on the same genetic background reveals a similar
phenotype. The more substantial increase in skin thickness, that was
previously shown in the K5D1 animals, is attributed to the different
background in the earlier study.30
The cyclin D2 and
cyclin D3 transgenic lines showed detectable cyclin D2 or D3 protein in
basal cells consistent with the K5D2 and K5D3 epidermal phenotype.
However, whereas cyclin D2 was observed in 40% of basal cells in the
interfollicular epidermis, cyclin D3 was observed in almost all of the
basal cells (Figures 3C and 4C)
. Furthermore, there was only a twofold
increase in cyclin D2 in Western blot whereas, cyclin D3 expression was
elevated fourfold to fivefold. Differences in the stability and/or
degradation between these two cyclins may be responsible for those
results.
The pattern of expression of keratin 1 (expressed in terminally differentiated cells) and keratin 5 (expressed in basal epithelial cells) indicated that epidermal differentiation was not affected and hyperproliferation was compensated by terminal differentiation. Thus, transgenic animals have a more densely packed basal cell compartment and suprabasal proliferative cells resulting in mild acanthosis (increase in the thickness of the nucleated layers of the epithelia) without hyperkeratosis. No evidence of spontaneous skin tumor development was found until 10 months of age in either K5D2 or K5D3 animals. We find it noteworthy that overexpression of cyclin D1 and cyclin D2, but not cyclin D3, was detected in premalignant lesions after 7,12-dimethylbenz[a]antracene (DMBA)/TPA applications (two-stage carcinogenesis protocol).27 In this sense, cyclin D1 expression is necessary but not sufficient for development of skin tumors41,42 because D1 knockout mice have a reduced number of papillomas whereas K5D1 mice did not have an increased number of skin tumors. We also performed carcinogenesis experiments using the two-stage protocol to test whether overexpression of cyclin D2 or cyclin D3 increased the susceptibility to chemical carcinogenesis. Overexpression of cyclin D2 did not seem to increase sensitivity to spontaneous or carcinogen-induced skin tumors (Rodriguez-Puebla M, Conti CJ, unpublished results). However, as in the case of cyclin D1, cyclin D2 expression may be necessary to promote skin tumor development.
Another unexpected result is the down-regulation of cyclin D2 in K5-D3
mice and down-regulation of cyclin D3 in K5-D2 mice (Figure 2)
. These
results are not specific for these two D-type cyclins because K5-D1
mice also show a reduced protein level of cyclin D3.42
We
can hypothesize that overexpression of one of the D-type cyclins is
compensated by reducing the level of expression of another member of
the family. It is possible that the transcription factors involved in
the regulation of D-type cyclin genes are responsible for these
regulatory loops. Also noteworthy is the finding that various
transcription factor binding sites were found in mouse cyclin D3 gene,
that includes GATA, NF-
B, ATF, E2F, Aprf, TCE, GAGA, TRE/Ap1, Sp1,
and Ap2.43
Because Sp1 and E2F binding sites were found in
the cyclin D1 gene,44
these transcription factors could
participate in this regulatory circuit.
Expression levels of the pRb family of proteins, pRb and p107, were clearly elevated in K5D3 and in K5D2–2102 transgenic lines. The K5D3–2203 line showed a stronger induction of p107 and changes in mobility consistent with phosphorylation. This is consistent with the greater level of cyclin D3 expression in this transgenic line. Similarly, the K5D3-2203 transgenic mice showed stronger induction of p130 other cyclin D2 and D3 lines showed a mild increase of this protein. Consistent with these results, we previously found an increase in p107 and pRb protein levels in mouse epidermal tissue after proliferative induction by TPA.29 pRb and p107 gene regulation has been suggested to be mediated by binding of E2F-1 to E2F sites.45 We therefore believe that induction of p107 and pRb in transgenic animals may be mediated by E2F.
One of the more relevant differences between the three transgenic mice overexpressing D-type cyclins was the development of thymic hyperplasia in cyclin D130 and cyclin D2, but not in cyclin D3 transgenic mice. The fact that cyclin D2 expression was only double in the K5D2-2101 line whereas it increased by fivefold in the K5D2-2102 line may explain why the K5D2-2101 line did not develop thymic hyperplasia. Mechanistic studies of CDK4 and CDK2 complex formations in the thymus of K5D2-2102 did not show relevant differences at 12 and 20 weeks of age. We did in vitro immunoprecipitation with thymic lysates of the transgenic and wild-type mice. To determine whether cyclin D2 overexpression resulted in changes in the composition of CDK/cyclin/CKI complexes, we analyzed CDK4 and CDK2 complex formations and kinase activities. We determined that neither p21Cip1 nor p27Kip1 are involved in the thymic regression at 20 weeks of age in transgenic mice. The kinase activities of CDK4 and CDK2 did not change during the thymic size regression. However, elevated CDK2 kinase activity at 12 weeks of age may be responsible for the hyperplastic phenotype in cyclin D2 transgenic thymus. Several reports have described that kinase activation of CDK2 occurs when increased levels of CDK4/cyclin complexes bind to the CKIs and release CDK2. A similar mechanism could be involved in the development of the hyperplastic thymus, where an increase of cyclin D2 protein levels bind some inhibitor in a binary or ternary complex with CDK4, although CKIs other than p21Cip1 and p27Kip1 would be responsible for this event. Interestingly, pRb protein levels were also increased in transgenic mouse thymus compared with normal sibling mice, as was shown in epidermal cells. In addition, the level of pRb phosphorylation increased at 20 weeks in the thymus of K5D2 animals. These data clearly show that in the older thymus (20 weeks) the pRb protein seems to be relatively inactive rather than the active form. However, we cannot rule out the possibility that the presence of cell populations in the thymus other than epithelial cells mask subtle differences in complex formations and also influence the pRb state observed at 12 and 20 weeks of age.
Interestingly the K5D3-2203 transgenic line expressed the transgene at a 3.5-fold higher level than nontransgenic, but did not develop thymic hyperplasia. This is consistent with the fact that none of the K5D3 founders developed thymic hyperplasia whereas three K5D130 and three K5D2 founders developed this phenotype. An interesting possibility is that cyclin D3 has a distinct role in thymic epithelial cells compared to the other D-type cyclins. In this regard, cyclin D3 has been reported to be induced in differentiation in other systems. For example, cyclin D3 is strongly up-regulated on induction of HL-60 leukemia cells to differentiate and it accumulates to high levels in a wide range of quiescent cells in mouse and human tissues.22 In addition, myoblast induction of cyclin D3 expression is closely coupled with withdrawal from the cell cycle and differentiation.46 In contrast, cyclin D1 and cyclin D2 have been established to play a role in proliferation and this function may be responsible for the hyperplastic thymus phenotype.
Each major subset of T cell, defined by CD4 and CD8 expression, is present in K5D2-2102 transgenic line at 12 weeks of age (thymic hyperplasia). These results show that despite the altered architecture, the thymic environment is able to generate mature T cells. Furthermore, these results confirm the histological diagnosis of hyperplasia and rule out the possibility of thymic lymphoma because, in this case, a single predominant phenotype should be present whereas thymus from K5D2-2102 transgenic line contain each of the major T cell subsets.
A relevant difference between thymic hyperplasia of
K5D1-710830
and K5D2-2102 transgenic mice was that, in the
latter, the increasing size of the thymus stops at 12 weeks of age and
at 20 weeks, the size was similar to normal siblings (Figure 6A)
. The
possibility that the transgene expression was turned-off after 12 weeks
was ruled-out because the expression levels determined by Western blot
analysis at 24 weeks of age was similar to 12-week-old mice (Figure 6C)
. Again the results suggest that a different mechanism of action
exists between cyclin D1 and cyclin D2. Although both transgenic mice
overexpressing these cyclins show thymic hyperplasia, their
histological characteristics are different and in one case the
hyperplasia reverts with age whereas, in the other, it progresses and
causes the death of the animal.
In summary, the mammalian D-type cyclins seem to have redundant functions in some systems, but clear differences have also been reported. Cyclin D1 and cyclin D2 are considered as proto-oncogenes based on genetic aberrations in human and animal malignancies.47 In contrast, cyclin D3 has not been firmly implicated in oncogenesis. Despite considerable similarities, unique phenotypes were reported for cyclin D1 and cyclin D2 knockout mice,23-25 whereas effects of cyclin D3 knockouts are unknown. Our results, using two new transgenic models, suggest that in some tissues (epidermis) D-type cyclins may have similar functions, whereas in others (thymus) the biological roles of the individual D-type cyclins are not fully redundant.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by Department of Human Services Grants CA 42157 and CA 57596; an M. D. Anderson Cancer Center institutional grant (CA 16672) for the animal facility; and a center grant (P30-E507784-01) for the histology service
Accepted for publication June 13, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  X. Mao, A. K. Stewart, R. Hurren, A. Datti, X. Zhu, Y. Zhu, C. Shi, K. Lee, R. Tiedemann, Y. Eberhard, et al. A chemical biology screen identifies glucocorticoids that regulate c-maf expression by increasing its proteasomal degradation through up-regulation of ubiquitin Blood, December 1, 2007; 110(12): 4047 - 4054. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-M. Chien, S. Rabin, E. Macias, P. L. Miliani de Marval, K. Garrison, J. Orthel, M. Rodriguez-Puebla, and M. L. Fero Genetic mosaics reveal both cell-autonomous and cell-nonautonomous function of murine p27Kip1. PNAS, March 14, 2006; 103(11): 4122 - 4127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vespa, S. J.A. D'Souza, and L. Dagnino A Novel Role for Integrin-linked Kinase in Epithelial Sheet Morphogenesis Mol. Biol. Cell, September 1, 2005; 16(9): 4084 - 4095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B.S. Pasumarthi, H. Nakajima, H. O. Nakajima, M. H. Soonpaa, and L. J. Field Targeted Expression of Cyclin D2 Results in Cardiomyocyte DNA Synthesis and Infarct Regression in Transgenic Mice Circ. Res., January 7, 2005; 96(1): 110 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Richard-Parpaillon, R. A. Cosgrove, C. Devine, A. E. Vernon, and A. Philpott G1/S phase cyclin-dependent kinase overexpression perturbs early development and delays tissue-specific differentiation in Xenopus Development, June 1, 2004; 131(11): 2577 - 2586. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Scheijen, M. Bronk, T. van der Meer, D. De Jong, and R. Bernards High Incidence of Thymic Epithelial Tumors in E2F2 Transgenic Mice J. Biol. Chem., March 12, 2004; 279(11): 10476 - 10483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Holzschu, L. A. Lapierre, and M. D. Lairmore Comparative Pathogenesis of Epsilonretroviruses J. Virol., December 1, 2003; 77(23): 12385 - 12391. [Full Text] [PDF]
|
|
|
|
 |
|
 X. Xu, S. Lyle, Y. Liu, B. Solky, and G. Cotsarelis Differential Expression of Cyclin D1 in the Human Hair Follicle Am. J. Pathol., September 1, 2003; 163(3): 969 - 978. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Benavides, M. F. Starost, M. Flores, I. B. Gimenez-Conti, J.-L. Guenet, and C. J. Conti Impaired Hair Follicle Morphogenesis and Cycling with Abnormal Epidermal Differentiation in nackt Mice, a Cathepsin L-Deficient Mutation Am. J. Pathol., August 1, 2002; 161(2): 693 - 703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Reichelt and T. M. Magin Hyperproliferation, induction of c-Myc and 14-3-3{sigma}, but no cell fragility in keratin-10-null mice J. Cell Sci., January 7, 2002; 115(13): 2639 - 2650. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||
