3 Results

↓27

3.1  Brk is required for intestinal homeostasis

3.1.1  Brk protein expression is restricted to differentiated cells

Expression of the gene encoding the non-receptor tyrosine kinase Brk is restricted to epithelial cells of the skin, prostate, and gastrointestinal tract in the mouse (Derry, et al., 2003;Llor, et al., 1999;Vasioukhin, et al., 1995). Brk mRNA expression is developmentally regulated. It is detected late in gestation in the mouse as regenerating epithelia begin to mature. Furthermore, mRNA expression is restricted to non-proliferating epithelial cells, suggesting a biological function for Brk in the process of differentiation in these epithelial tissues (Vasioukhin, et al., 1995).

To confirm the developmentally regulated expression pattern of Brk based on mRNA studies, Brk protein expression throughout the gastrointestinal tract of wild-type mice was examined by immunohistochemistry (Fig. 4). Jejunum and colon of adult mice were extracted, fixed in 4% paraformaldehyde, embedded and 5 µm paraffin sections were incubated with anti-Brk antibodies or anti-IgG antibodies as controls. Reactions were detected with the Vector ABC Kit and visualized with DAB. Nuclei were counterstained with hematoxylin. Brk protein was detected in both jejunum and colon, where its expression was restricted to epithelial cells that exit the cell cycle and undergo terminal differentiation (Fig. 4). This highly specific expression pattern supported the hypothesis that the Brk tyrosine kinase plays a role in promoting differentiation in normal regenerating epithelial linings.

↓28

To investigate a putative function for the non-receptor tyrosine kinase Brk in the differentiation process of the gastrointestinal tract, mice deficient for Brk were generated (Wenjun Bie, 2005). No overt phenotypic changes were observed in Brk deficient mice. They were viable, fertile and did not develop spontaneous tumors (Wenjun Bie, 2005). These findings suggested that a high degree of redundancy exists between different members of the Brk family.

Fig. 4: Expression of Brk protein in the gastrointestinal tract. Immunohistochemistry for Brk in adult wild-type jejunum and colon using anti-Brk antibodies or anti-rabbit IgG antibodies as controls and DAB as a substrate (brown). Nuclei were counterstained with hematoxylin (blue). Size bar represents 50 µm. Brk expression is restricted to differentiated epithelial cells.

Loss of Brk mRNA expression in mutant mice was confirmed by RNase protection assay (Fig. 5A). Furthermore, expression of Brk protein was compared in neonatal and adult ileum and colon by Western blotting (Fig. 5B). Tissue lysates were separated by SDS-page and immunoblotting for Brk and β-actin as a loading control were performed. Highest levels of Brk protein were detected in neonatal colon, whereas adult mice showed peak levels of Brk protein in ileum (Fig. 5B). These data suggested that Brk protein expression is regulated along the length of the intestine. Differential regulation during development has also been shown for Brk mRNA expression (Llor, et al., 1999). No Brk protein expression was observed in the intestine of knockout mice, confirming the loss-of-function mutation in the Brk gene (Fig. 5B).

↓29

Fig. 5: Brk mRNA and protein expression in the gastrointestinal tract. (A) RNase protection analysis for Brk mRNA synthesis in homozygous mutant and wild-type mice. The loading control was performed with a probe for cyclophilin. (B) Western blot analysis for Brk synthesis in neonatal (14 days) and adult (8 weeks) wild-type (WT) and mutant (Brk -/-) ileum and colon. The loading control was performed with anti-β-actin antibodies.

3.1.2 Increased epithelial cell turnover in Brk knockout mice

To investigate the influence of Brk on the homeostatic balance of the gastrointestinal epithelium, wild-type and knockout mice were examined for their proliferative state. Age-matched mice were pulse labeled with BrdU 2 hours prior to sacrificing, and their gastrointestinal tract was dissected, fixed and embedded. Paraffin sections of the small intestine and colon were subjected to immunohistochemistry with antibodies directed against PCNA and BrdU (Fig. 6A, B). Analysis of PCNA- and BrdU-positive intestinal epithelial cells revealed an increased number of proliferating cells in Brk knockout mice compared to wild-type controls (Fig. 6A, B). Importantly, the proliferating cells were present in the middle and upper regions of the crypt in Brk knockout mice. These areas of the crypt are remote from the stem cell area and are normally fully differentiated and non-proliferating. The expanded proliferative zone and increased number of proliferating cells in Brk deficient mice suggested a deregulated balance between proliferation and differentiation in the intestinal epithelium in the absence of Brk signaling.

Fig. 6: Defects in intestinal epithelial homeostasis in the absence of Brk signaling. Photomicrographs of immunohistochemical staining for PCNA (A) and BrdU (B) (brown) from sections of jejunum and colon of wild-type (WT) and knockout (Brk -/-) mice. Mice were injected with 50 µg BrdU/ g bodyweight and sacrificed 2 hours later. Sections were counterstained with hematoxylin (blue). Size bars represent 50 µm. The amount of proliferation indicated by PCNA and BrdU positive cells is increased in the mutant epithelium.

↓30

In addition, paraffin sections of the distal ileum and jejunum of age-matched wild-type and knockout mice were stained with hematoxylin and eosin and observed by light microscopy (Fig. 7A, C). Cross visual comparison between wild-type and knockout mice showed an increased height of intestinal villi in knockout mice. To further examine and quantify this difference, villus height and crypt depths were measured for at least 3 animals per group and tissue with at least 30 scores each (Fig. 7B, D). The resulting data showed a statistically significant difference (P-value ≤ 0.05) in villus height of the distal ileum and distal jejunum of wild-type and knockout mice. The villi in Brk deficient mice were significantly longer than in their wild-type counterparts. On the other hand, no statistically significant difference was measured for the crypt depths of the analyzed tissues (P-value = 0.11) (data not shown). To ensure that the differences in villus height are not due to differences in general body size, total animal weight and length of the entire small intestine were measured. No significant difference in either parameter was observed in wild-type and knockout mice (data not shown). In conclusion, these data suggested that the observed increase in proliferation in the intestinal epithelium of Brk knockout mice contributes to the phenotypic appearance of longer villi.

Fig. 7: Loss of Brk expression affects the crypt-villus morphology. (A, C) Representative sections of the distal ileum and distal jejunum from wild-type (WT) and knockout (Brk -/-) mice stained with hematoxylin/eosin. Size bars represent 100 µm. (B, D) Histograms of average villus length in wild-type and knockout distal ileum (B) and jejunum (D). Error bars represent ± SD. P-values were determined using the Student’s test. The average villus length is drastically increased in the mutant mucosa.

3.1.3  Increased accumulation of nuclear β-catenin in the absence of Brk

The expansion of the progenitor zone in Brk deficient mice suggested that Wnt signaling might be affected in the intestines of these mice. Wnt signaling has been shown to define the intestinal epithelial progenitor cell phenotype (van de Wetering, et al., 2002). Activation of canonical Wnt signaling results in accumulation of nuclear β-catenin, which in complex with TCF-4 controls proliferation versus differentiation in intestinal epithelial cells (Batlle, et al., 2002;Pinto, et al., 2003). To determine by which molecular mechanism loss of Brk could affect proliferation in the adult intestine, the expression of β-catenin and its target gene c-myc was examined in the small and large intestine of wild-type and knockout mice. Immunostaining with anti-β-catenin antibodies revealed membrane-localized β-catenin along the crypt-villus axis as well as nuclear β-catenin in cells occupying basal positions of the crypt in sections of small and large intestine (Fig. 8A, B). Interestingly, comparing numerous fields, an increased number of cells positive for nuclear β-catenin were observed in Brk knockout mice compared to their wild-type counterparts (Fig. 8A-C).

↓31

To address, whether this Brk-dependent up-regulation of nuclear β-catenin affected the expression of β-catenin target genes, c-Myc expression in the gastrointestinal tract was analyzed. It has been previously shown that the formation of nuclear β-catenin/Tcf4 complexes results in the direct up-regulation of c-myc, which in turn represses p21 CIP1/WAF1 by direct promoter binding, thereby allowing cells to proliferate (He, et al., 1998;van de Wetering, et al., 2002). Distal jejunum and colon sections of wild-type and Brk knockout mice were immunostained with anti-c-Myc antibodies or anti-IgG as controls (Fig. 9A, B). In both wild-type and knockout animals, c-Myc protein expression was detected in the nucleus of crypt cells and c-Myc expression was absent from differentiated villus cells. However, knockout mice exhibited an increased c-Myc expression when compared to wild-type controls.

Taken together, these results demonstrated that in the absence of Brk signaling β-catenin and its target gene c-myc are induced. Loss of Brk expression activated the Wnt pathway, which might contribute to and/or result in the increased proliferation observed in Brk deficient mice. These in vivo findings supported a proposed role for Brk in maintaining intestinal epithelial cells in a non-proliferative, differentiated state. Mutations in the APC gene resulting in increased accumulation of nuclear β-catenin have been shown to promote the development of spontaneous intestinal tumors, as shown in the case of APC MIN mice (Moser, et al., 1992;Su, et al., 1992). Thus, loss of Brk and the resulting activation of nuclear β-catenin might sensitize these knockout mice to intestinal tumorigenesis.

Fig. 8: Increased nuclear β-catenin expression in mutant intestinal epithelium. (A, B) Immunohistochemical analysis of wild-type (WT) and knockout (Brk -/-) distal jejunum and colon sections with antibodies against β-catenin and IgG as controls. Staining of β-catenin shows accumulation in the nucleus of cells at the crypt base (arrowheads), but is absent from the nucleus of cells along the crypt-villus axis, where it is only membrane-localized. Size bars represent 50 µm. Nuclear β-catenin is detected in more crypt cells of mutant intestinal epithelium compared to wild-type mice.

↓32

Fig. 9: Expression of nuclear β-catenin and its target gene c-myc in mutant intestinal epithelium. Representative sections of the distal jejunum (A) and colon (B) from wild-type (WT) and knockout (Brk -/-) adult animals were stained with anti-β-catenin and anti-c-myc antibodies. Immunostaining with anti-IgG antibodies served as controls. Size bars represent 50 µm. Brk mutant mice exhibit increased nuclear β-catenin accumulation and upregulated expression of the β-catenin target gene c-myc in the intestinal crypts.

3.1.4  Regulation of β-catenin activity by Brk may be mediated by Akt

Multiple signaling pathways can impinge on β-catenin, affecting its stability and/or subcellular localization. It has been reported that Wnt signaling is required for the nuclear activity of β-catenin, but is not sufficient to fully activate it (He, et al., 2004). The survival kinase Akt has been shown to further facilitate the stabilization and nuclear accumulation of β-catenin, mainly acting through Dishevelled and GSK-3β but also through direct phosphorylation of β-catenin (Fukumoto, et al., 2001;Persad, et al., 2001;Tian, et al., 2004). A recent study showed that Brk binds to and inhibits Akt kinase signaling (Zhang, et al., 2004).

To study, whether Akt is involved in the regulation of β-catenin by Brk, tissue lysates from distal ileum of adult wild-type and knockout mice were analyzed for Akt kinase activity (Fig. 10). Endogenous Akt was immunoprecipitated using immobilized Akt antibodies and in vitro kinase assays were performed by incubation with ATP and purified recombinant GSK-3 protein. Samples were subjected to Western blotting using phospo-specific GSK-3 antibodies. The analysis showed that Akt activity was consistently increased in mice deficient for Brk compared to wild-type mice from either an outbred (2.8) or inbred (2.8 G12) background (Fig. 10). Interestingly, higher levels of basal Akt kinase activity were observed in inbred mice. These data suggested that Brk signaling might inhibit β-catenin activity through the Akt pathway. Brk signaling may therefore have a role in balancing the role of Wnt-β-catenin in promoting stem cell self-renewal (Reya, et al., 2003). Loss of Brk may facilitate the activation of β-catenin by the survival kinase Akt.

↓33

Fig. 10: Increased Akt kinase activity in the intestinal epithelium of Brk mutant mice. Endogenous Akt was immunoprecipitated from total cell lysates of the distal ileum of inbred (2.8 G12) and outbred (2.8) wild-type (WT) and knockout (KO) mice. In vitro kinase assays were performed with purified recombinant GSK-3 and ATP. Reactions were stopped at 0 (control) and 30 minutes and subjected to Western blotting with phospho-GSK-3β and total GSK-3 β antibodies. Equal amounts of total GSK-3β protein were present in the reactions. Loss of Brk results in increased Akt kinase activity in the intestinal epithelium.

3.2 Brk signaling protects from intestinal inflammation

3.2.1  Chronic inflammation in Brk deficient mice

The regulation of proliferation of the stem cells in intestinal crypts is critical for the maintenance of the mucosal structure and the capacity for effective absorption and defensive barrier functions (Booth and Potten, 2001). Disruption of the balance between proliferation, differentiation and apoptosis leads to loss of normal absorptive functions and an increased risk of tumorigenesis.

Upon histological examination of Brk deficient mice, evidence for increased inflammation in multiple epithelial linings was observed. To investigate the intestinal immune homeostasis in Brk deficient mice, the entire length of the small and large intestine of Brk mutant mice was histologically examined by staining with hematoxylin and eosin, and monitored by light microscopy. Interestingly, an increased amount in lymphoid tissue was observed in knockout mice compared to wild-type controls (Fig. 11A). These organized mucosal lymphoid follicles, named Peyer’s patches, were found at high frequency not only in the lower small intestine but also in the colon of Brk deficient mice (Fig. 11A-C). Peyer’s patches are lymphoid follicles containing antibody producing B-cells to defend against invading bacteria, parasitic microbes, viruses and other foreign particles (Chen, et al., 2004). They normally appear in the mucus secreting lining of the small intestine as was observed in wild-type animals (data not shown). Peyer’s patches not only play a role in the intestinal host defense but also in regulating immune homeostasis of the intestinal epithelium (Chen, et al., 2004). It has been reported that this modulation of epithelial physiology by Peyer’s patch lymphocytes involves both cell-cell contact and cytokine signaling.

↓34

The observed increase in lymphoid tissue in Brk knockout mice suggested an increase in inflammation and activation of the immune system. Inflammation results in production of pro-inflammatory cytokines and chemokines, and aberrant cytokine signaling is a hallmark of chronic inflammation (Philip, et al., 2004). Altered cytokine production appears to be critical for inducing pathologically increased rates of epithelial cell turn over in inflammation (Podolsky, 2002). Transcriptional upregulation of pro-inflammatory cytokines has been shown not only to be a result of a global response of intestinal epithelial cells to increased invasion with microbes and bacteria but can also be caused by imbalance in intestinal epithelial homeostasis (Pedron, et al., 2003).

To study cytokine production in Brk deficient mice, age-matched wild-type and knockout mice were sacrificed and the distal small intestine was used to prepare total RNA. RNase protection assays with a multi-probe for inflammatory cytokines were performed (Fig. 11B). Analysis of three wild-type and three knockout outbred (2.8) mice showed a significant increase in interleukin-18 (IL-18) and interleukin-6 (IL-6) mRNA levels in Brk knockout mice compared to wild-type controls (Fig. 11B). IL-6 has been shown to be a pro-inflammatory cytokine, which plays a dual role in protecting against mucosal injury and promoting intestinal tumorigenesis (Heinrich, et al., 2003). Increased IL-6 production has been shown to correlate with colon tumor formation and growth (Becker, et al., 2004). IL-18 is produced by various cells including intestinal epithelial cells. It has been shown to induce interferon-γ (IFN-γ), and to play a crucial role in proliferation and maintenance of the intestinal epithelium (Okazawa, et al., 2004). Increased levels of IL-18 have been found in intestinal mucosal biopsies from IBD patients (Monteleone, et al., 1999;Pizarro, et al., 1999). Thus, up-regulation of both IL-6 and IL-18 expression in Brk deficient mice supported the hypothesis of increased inflammation in knockout mice as indicated by the large amount of Peyer’s patches.

Disruption of the mucosal barrier function and the concomitant immune hyperactivation by the microflora was shown to represent the key event that leads to the progressive transformation of epithelial cells and thus to chronic inflammation. To address this, the expression of proteins involved in maintaining barrier function was examined in the intestinal epithelium of wild-type and Brk deficient mice. To this extent, immunohistochemistry for the tight junction marker ZO-1 and the adherens type junction marker β-catenin were performed (Fig. 11C). Intestinal paraffin sections of wild-type and knockout mice were incubated with FITC-conjugated anti-ZO-1 antibodies and visualized by fluorescence microscopy. Comparing wild-type and Brk knockout mice for ZO-1 protein expression showed no significant difference in the appearance of tight junctions (Fig. 11C). Furthermore, immunohistochemistry for β-catenin was performed (Fig. 11C). Both wild-type and knockout animals showed a similar pattern and staining for membrane-bound β-catenin (Fig. 11C). Taken together, these data suggested no visible difference in epithelial barrier protein expression in Brk deficient mice and wild-type controls.

↓35

Fig. 11: Chronic inflammation in the intestinal epithelium of adult outbred Brk deficient mice (2.8). (A) Gross microscopical examination of intestinal sections from outbred 2.8 knockout mice. Sections from small intestine and colon were stained with hematoxylin/eosin. Increased amount of lymphoid tissue, Peyer’s patches, was observed in mutant epithelium. (B) RNase protection assays were performed with total RNA from distal ileum of adult (8 weeks) wild-type (WT) and knockout (Brk-/-) oubred (2.8) animals and 32P-labeled antisense multiprobes specific for cytokines. Rpl32 and GAPDH expression were examined as controls. Brk mutant mice exhibit increased expression of IL-18 and IL-6 in the intestinal epithelium. (C) Immunohistochemical analysis of the apical epithelial barrier. Immunostaining with antibodies specific for ZO-1 and β-catenin on colonic sections from wild-type (WT) and mutant (Brk -/-) animals. No differences in the expression and position of these tight (ZO-1) and adherens (β-catenin) junction markers between wild-type and mutant mice were observed. Size bars represent 50 µm.

3.2.2 Brk protects the intestinal epithelium from cytokine-mediated injury

It has been reported that the balance of proliferation and differentiation along the crypt-villus axis is an important factor in protecting the intestinal epithelium from injury (Booth and Potten, 2001). Furthermore, it is well known that immune responses in the intestine remain in a state of controlled inflammation and that their deregulation leads to the development of inflammatory bowel diseases (Kanai and Watanabe, 2004).

To test the effect of loss of Brk on intestinal epithelial integrity, a model of intestinal injury and inflammation was chosen. This model utilized the oral administration of dextran sodium sulfate (DSS). DSS is a synthetic, sulfated polysaccharide that causes acute and chronic colitis (Seril, et al., 2003). The severity of induced injury greatly depends on the mouse strain analyzed, suggesting the influence of host genetic determinants on the inflammation severity and tumor risk. In general, genetically engineered mice on a C57BL background are more resistant compared with equivalent mutant mice on other genetic backgrounds such as 129/Sv.

↓36

Inbred Brk mutant and wild-type mice were subjected to 3% DSS in drinking water ad libitum for 5 days (5 days) followed by a recovery period with water for 3 days (8 days) (Fig. 12A). Both groups were closely monitored for weight loss and rectal bleeding, and at 5 or 8 days DSS treated mice were sacrificed and the gastrointestinal tract excised. The entire colon was fixed and embedded, and paraffin sections were stained with hematoxylin and eosin and evaluated by light microscopy. As seen from the comparison of representative photomicrographs taken of wild-type and Brk mutant mice at the indicated time-points, Brk mutant colons show severe and extensive denudation of the surface epithelium (erosions) and mucodepletion of glands compared to wild-type control mice (Fig. 12B,C). Separation of the crypt base from the muscularis mucosa and subsequent loss of crypts was observed in mutant mice at 5 days DSS, whereas a similar phenotype was not observed in wild-type mice before 3 days post DSS. Furthermore, Brk mutant mice exhibited an extended amount of inflammatory infiltrate in their submucosa and lamina propria and a frequent appearance of focal erosions when compared to wild-type control mice (Fig. 12B,C). Especially at 3 days post DSS treatment an extensive inflammatory infiltration of the mucosa could be observed in Brk deficient mice (Fig. 12C). No difference in histological appearance was observed in untreated wild-type and Brk knockout colons (Fig. 12B).

In addition to the observed histological differences in wild-type and Brk deficient mice post DSS treatment, the expression of inflammatory cytokines was measured by performing RNase protection assays. Total RNA of untreated and DSS treated (8 days) colons of inbred wild-type and Brk knockout mice was prepared, and analyzed by multiprobe RNase protection assay. In contrast to our previous findings (Fig. 11B), no differences in cytokine mRNA levels were observed in untreated mice (Fig. 12D). This might be due to the genetic background of the analyzed animals. In contrast to the oubred mice (2.8) utilized for previous studies, inbred mice with a C57BL background (2.8G12) were used in the DSS study. Whereas genetically engineered mice on a 129/Sv strain appear to be especially susceptible to IBD-like disease, mice on a C57Bl background have been shown to be more resistant to inflammation (Rogers and Fox, 2004).

Significantly elevated levels of pro-inflammatory cytokines were observed post DSS treatment in both wild-type and knockout mice (Fig. 12D). However, an overall greater induction of cytokine mRNA was observed in Brk mutant mice when compared to wild-type control mice with DSS treatment (Fig. 12D). This upregulation of pro-inflammatory cytokine mRNAs in inbred Brk knockout mice suggested increased severity of acute inflammation in response to DSS treatment. Furthermore, loss of IL-18 expression, indicating the overall loss of epithelium, was observed in mutant mice.

↓37

Taken together, the mutant epithelium appeared more susceptible to DSS-induced histological damage than control epithelium accompanied by the enhanced production of many inflammatory mediators (Fig. 12A-D). This suggested that Brk signaling acts to ensure epithelial cell integrity and loss of Brk signaling results in increased epithelial susceptibility to injury.

Fig. 12: Colonic epithelial damage in inbred Brk deficient mice following DSS administration. (A) Study design. Mice were subjected to 5 days of 3% DSS (5d) in drinking water followed by 3 days water (8d). Arrows indicate days 0, 5 and 8 of DSS treatment. (B, C) Representative photomicrographs of colons from inbred (2.8 G12) wild-type (WT) and knockout (Brk -/-) mice stained with hematoxylin/eosin at days 0, 5 and 8 of DSS treatment. Brk knockout mice exhibit increased epithelial injury with severe crypt loss after treatment. By day 8 after treatment, the inflammatory cell infiltration has become more extensive in mutant animals (indicated by arrow). Size bars represent 100 µm. (D) RNase protection assays were performed with total RNA from distal colon of inbred (2.8 G12) wild-type (WT) and knockout (Brk-/-) animals and 32P-labeled antisense multiprobes specific for cytokines. Rpl32 and GAPDH expression was examined as controls. Brk mutant mice exhibit increased expression of pro-inflammatory cytokines, such as IL-6, but a decreased expression of the epithelial specific cytokines IL-18, in the colon after DSS treatment.

↓38

To elucidate the mechanisms behind the increased susceptibility of Brk deficient mice to DSS treatment, Brk protein expression in the intestine of wild-type mice during DSS induced colitis was analyzed. Ileum and colon mucosal lysates were prepared from the respective mice for Western blot analysis with anti-Brk-antibodies (Fig. 13A, B). Increased levels of Brk protein were observed in ileum and colon of wild-type mice treated with DSS compared to untreated mice (Fig. 13A, B). Consistent induction of Brk protein was observed in the ileum of treated wild-type mice at 5 days DSS and 3 days post DSS (8 days DSS) (Fig. 13A). However in colon, induction of Brk protein levels could only be detected in 5 days DSS mice whereas very low levels were present at 3 days post DSS treatment (Fig. 13B). This apparent loss of Brk protein is most likely due to the general loss of epithelial cells during the inflammatory process at the 8 day time-point. No signal was detected in Brk mutant mice, confirming their genotype.

To provide evidence that Brk is induced in epithelial cells post DSS treatment, Brk immunohistochemistry was performed. Colon sections of untreated, 5 days DSS and 3 days post DSS treated wild-type mice were immunostained with anti-Brk-antibodies and screened by light microscopy. Brk protein expression is restricted to non-proliferating cells in the top third of the intestinal epithelium in untreated wild-type mice (Fig. 13C). However, upon DSS treatment nuclear Brk is expressed throughout the crypt epithelium. Brk protein is excluded from inflammatory cells infiltrating the mucosa and the lamina propria (Fig. 13C, D). The observed strong nuclear localization of Brk suggested a possible role in the regulation of transcription in response to injury. Brk has been shown to phosphorylate and inhibit its nuclear substrate Sam68, which has been implied in the regulation of splicing (Derry, et al., 2000;Matter, et al., 2002). The physiological importance of this nuclear localization in response to injury needs to be further investigated.

↓39

In conclusion, Brk appeared to have a protective role in intestinal epithelial cells exposed to cytokines during inflammation. It may exert its protective function by promoting intestinal homeostasis, as loss of Brk expression resulted in increased proliferation. Thus, Brk signaling may be critical for the protection of gut injury and associated mortality.

Fig. 13: Induction of nuclear Brk tyrosine kinase expression upon DSS treatment in wild-type mice. (A, B) Western blot analyses were performed with total cell lysate from distal ileum (A) and distal colon (B) of wild-type (WT) and knockout (Brk-/-) animals using antibodies against Brk. Expression of β-actin was examined as a control for protein loading. Brk protein expression increased with treatment. (C, D) Immunohistochemical analysis of colons from wild-type (WT) animals at days 0, 5 and 8 of DSS treatment with anti-Brk antibodies or anti-IgG antibodies as a control. Size bars represent 50 µm. Induction of nuclear Brk protein expression throughout the colonic epithelium of wild-type mice after DSS treatment.

3.2.3  Increased susceptibility of Brk deficient mice to DSS inhibits tumor development in the AOM/DSS tumorigenesis model

Increased proliferation, pro-survival signaling and chronic inflammation have all been correlated with the promotion of cancer (Philip, et al., 2004;Sancho, et al., 2004). The observation of these characteristics in Brk deficient mice led to the hypothesis that Brk may have tumor suppressor-functions in wild-type mice (Macleod, 2000;Sherr, 2004). Brk expression is restricted to non-proliferating cells in the intestinal epithelium. Loss of this tyrosine kinase in mice resulted in increased epithelial cell turnover and chronic inflammation, predisposing these mice to tumor development.

↓40

To get insights into Brk’s putative tumor suppressor functions the AOM mouse model of colon carcinogenesis was utilized. AOM is a pro-carcinogen, which upon metabolic activation causes the formation of O6-methylguanine (Pegg, 1984). It induces tumors in the distal colon of rodents and is commonly used to elicit colorectal cancer in experimental animals (Boivin, et al., 2003). In pilot studies, wild-type and knockout mice were subjected to weekly AOM injections (10 mg/kg) for the duration of 6 weeks and sacrificed 10 weeks after the last injection (Heap and Bie, unpublished). At that time, knockout mice exhibited a considerably higher incidence of aberrant crypt foci formation (ACFs) within the distal colon compared to wild-type controls. Aberrant crypt foci are the smallest identifiable lesions proposed to lead to colorectal carcinoma (Cheng and Lai, 2003). Furthermore, at 24 weeks after the last AOM injection, knockout mice exhibited colonic tumors, which were absent in wild-type mice (Heap and Bie, unpublished). These data supported a tumor suppressor function for Brk. However, due to the small number of animals used in these pilot experiments, further studies needed to be performed to confirm these preliminary results.

To this extent, wild-type and Brk deficient mice were subjected to a novel inflammation-related carcinogenesis model. In this accelerated tumorigenesis model, mice are subjected to a combination treatment of AOM and DSS, which represents an extremely efficient way to generate adenocarcinomas of the intestine, and greatly enhances tumor development. The inflammation caused by DSS treatment following AOM injection is presumed to represent the key event that creates a microenvironment, that is permissive for the progressive transformation of colon epithelial cells, and greatly enhances the incidence of AOM-induced tumors (Okayasu, et al., 1996;Tanaka, et al., 2003). Mice subjected to this treatment develop tumors within 20 weeks of study begin. Age-matched wild-type and knockout mice were given a single intraperitoneal administration of AOM (10 mg/kg body weight), and a 1-week oral exposure to DSS (2%) (Fig. 14A). Twenty weeks post treatment mice were sacrificed and their entire colons were excised, opened longitudinally and fixed in 70% ethanol. Mice were closely monitored for weight loss during the entire study. Surprisingly, wild-type animals showed high mortality and morbidity upon treatment with a loss of 50% of treated mice (Fig. 14B). In contrast, Brk knockout mice showed 80% survival and lower morbidity upon treatment compared to wild-type controls (Fig. 14B). Furthermore, after week 12 of the study, an increased amount of wild-type mice with anal prolapse due to tumor development in the distal colon were observed.

To further analyze these surprising results, the colons of wild-type and knockout mice that survived untill the end of the study were macroscopically examined by light microscopy. Nodular and polyploid colonic tumors were observed in the middle and distal colon of wild-type mice (Fig. 14C,D). In sharp contrast, the number and size of tumors were substantially reduced in Brk mutant mice when compared to wild-type mice (Fig. 14C,D). Thus, despite the appearance of chronic inflammation in Brk deficient mice and in contrast to our pilot study, the colitis-associated tumor incidence in the DSS/AOM model was greatly reduced in Brk deficient mice. However, it was previously observed that the Brk mutant epithelium was more susceptible to DSS-induced histological damage than control epithelium (Fig. 13). Brk signaling appeared to be critical for the protection against DSS-induced inflammation, and treatment of Brk knockout mice with DSS resulted in an immediate and major upregulation of pro-inflammatory cytokines including IL-6. This is considered a hallmark of acute inflammation, which has been shown to counteract cancer development. This observation suggested that treatment of knockout mice with 2% DSS after the initial AOM injection resulted in acute inflammation in the colonic crypt and henceforth in major loss of crypt epithelium in Brk deficient mice. This loss of crypt epithelial cells, which were transformed by AOM, may be the cause for the lower tumor incidence in Brk deficient mice.

↓41

Fig. 14: Inflammation-related mouse colon carcinogenesis model. (A) Study design. Arrows indicate initial AOM injection at 0 weeks and study end at week 20 respectively. The black box indicates the 7 day treatment with 2% DSS. (B) Increased mortality in wild-type mice upon treatment. Survival of wild-type (WT) and mutant (Brk -/-) mice was monitored over time of treatment. (C, D) Macroscopic view of colons of wild-type (WT) and mutant (Brk -/-) mice after AOM/DSS treatment. Size bars represent 1 cm. Due to their increased susceptibility to inflammation, Brk deficient mice appear less susceptible to AOM/DSS induced tumor formation in the colon.

In wild-type mice on the other hand, tumor formation is promoted by the induction of inflammation by 2% DSS after the initial low dose of AOM, henceforth resulting in the high tumor incidence reported. To prove this hypothesis and support a tumor suppressor function for Brk and hence validate the previous pilot study, further experiments are needed. Studies with a lower dose of DSS (0.5%) combined with AOM and AOM only carcinogenesis models are underway.

3.3 Brk sensitizes cells to apoptosis in vivo and in vitro

3.3.1  Brk is required for DNA-damage induced intestinal apoptosis

Apoptosis is a cellular response that regulates important processes such as tissue homoeostasis, defense against certain pathogens and elimination of unwanted cells.The imbalance in proliferation and differentiation in the gastrointestinal tract of Brk deficient mice suggested that Brk is involved in the maintenance of tissue homeostasis. Brk knockout mice exhibited an expanded progenitor zone and the physiological appearance of longer villi. Since apoptosis is a mechanism involved in counterbalancing the crypt cell division (Potten, 1992), one could hypothesize that Brk deficient mice may be impaired in their apoptotic response. The small intestinal epithelium provides a well-characterized, readily accessible and quantifiable model to address questions regarding the importance of certain gene products on epithelial cell fate in vivo following genotoxic damage (Clarke, et al., 1994;Merritt, et al., 1997;Merritt, et al., 1995;Merritt, et al., 1994;Potten, 1992). Following genotoxic damage, cell death is induced in the crypt compartment (Potten, 1990).

↓42

To investigate a potential role of the Brk tyrosine kinase as a physiological regulator of apoptosis in mouse intestinal epithelia, Brk deficient mice were analyzed in their spontaneous or damage-induced apoptotic response. Wild-type and Brk deficient mice were either untreated or subjected to whole-body γ-irradiation (8 Gγ), which has been shown to induce apoptosis in epithelial cells of the small intestinal crypts (Potten, 1990). Apoptotic cells were identified by caspase 3 activation by immunostaining with anti-cleaved caspase 3-antibodies, and quantified microscopically on a cell positional basis. Previously, an increase in epithelial cell proliferation marked by increased BrdU labeling in the intestine of Brk deficient mice was noted (Fig. 6). This extended zone of proliferation correlated with the histological appearance of longer villi, suggesting that the increased proliferation in Brk deficient mice is not counterbalanced by an increased cell loss. In keeping with this observed intestinal phenotype of Brk knockout mice, no statistically significant differences in spontaneous apoptosis within the crypt epithelia of untreated wild-type and Brk knockout mice were observed (data not shown). Apoptotic small intestinal crypt cells were infrequent in both non-irradiated wild-type and Brk null mice.

However, treatment with γ-irradiation induced profound apoptosis in proliferating areas of small intestinal crypts in wild-type but not Brk deficient mice observed at 6 and 72 hours after γ-irradiation (Fig. 15A). With the dose of 8 Gγ, the 6 hour time-point represents the peak of early apoptosis in response to γ-irradiation in the small intestine (Potten, 1997, Potten, 1998 #3408). The 72 hour time-point on the other hand represents the time of crypt regeneration and repopulation, when apoptotic crypt cells are migrating towards the apex of the villus where they will be shed. Brk deficient mice exhibited significantly less apoptosis than their wild-type counterparts with a 2-fold decrease in number of apoptotic cells compared to wild-type mice at both time-points (Fig. 15B). Wild-type mice on the other hand exhibited 4 to 6 apoptotic events per crypt/villus unit consistent with the saturation level of small intestinal apoptosis reported in the literature (Fig. 15B) (Potten, 2004). By Student’s t-test the observed differences were highly significant (P ≤ 0.05). In addition, loss of Brk resulted in a reduced percentage of crypt-villus units that showed significant epithelial apoptosis (4 or more apoptotic cells per unit) (Fig. 15C). The observed reduction of apoptotic cells in Brk deficient mice at both 6 and 72 hours post irradiation suggested that the differences at the early time-point are not due to a delay in the apoptotic response in knockout mice. To support these findings, tissue lysates from distal ileum of wild-type and Brk deficient mice at 6 hours post irradiation were prepared for Western blotting. Less cleavage of caspase 3 detected by Western Blotting was observed in Brk deficient mice compared to wild-type mice (Figure 15D).

These data suggested that the non-receptor tyrosine kinase Brk is involved in the intestinal DNA-damage induced p53-dependent apoptosis in vivo. Therefore, Brk might be an important determinant of damage-induced apoptosis in intestinal epithelia.

↓43

Fig. 15: Resistance to DNA-damage induced apoptosis in the absence of Brk. (A) Radiation –induced apoptosis of cells in the small intestine. Sections of distal jejunum from untreated and γ-irradiated (8 Gγ) wild-type (WT) and knockout (Brk -/-) mice at 6 and 72 hours post irradiation were prepared, analyzed by staining for cleaved caspase 3 and detected with Avidin-FITC (apoptotic cells stain green). Size bar represents 100 µm. (B, C) Histogram (B) and frequency histogram (C) of apoptotic cells per crypt-villus unit. Approximately 100 crypt-villus units per section were scored. Values shown are mean ± S.D. from two sections from at least three different mice per group. (D) Western blot analysis of irradiated ileum. Total cell lysates from distal ileum of wild-type (WT) and knockout (Brk -/-) mice 6 hours after irradiation (8 Gγ) were subjected to Western blotting with anti-cleaved caspase 3 antibodies. Immunoblotting with anti-Brk and anti-β-actin antibodies served as controls for the genotypes and protein loading respectively. Brk deficient mice show increased resistance to radiation-induced apoptosis in the small intestine.

3.3.2 Enhanced pro-survival Akt and MAPK signaling in Brk knockout mice

As previously observed, untreated mice deficient for Brk exhibited enhanced Akt/ Protein Kinase B activation when compared with wild-type control mice (Fig. 10). The Akt serine/threonine kinase has been shown to act as a survival factor that stimulates progression of the cell cycle and prevents cells from undergoing apoptosis (Datta, et al., 1997;Weng, et al., 2001;Weng, et al., 2001)}. To investigate whether the reported enhanced Akt activation in knockout mice might contribute to the apoptotic resistance of these mice to irradiation, lysates from distal ileum of wild-type and Brk mutant mice were analyzed by Western Blotting and Akt in vitro kinase assays.

Tissue homogenates of age-matched wild-type and knockout mice at 6 hours post irradiation were separated by SDS-PAGE, and Western Blotting with anti-phospho-Ser473-Akt and anti-total Akt antibodies was performed (Fig. 16). Phosphorylation of Akt at both serine 473 and threonine 308 is required for complete activation of this kinase. During activation, Akt undergoes a conformational change exposing these sites to phosphorylation. However, phosphorylation of Thr308 is more transient than Ser473. Total tissue lysates of wild-type and Brk deficient mice contained similar amounts of total Akt protein whereas increased amounts of Akt activation by Ser473 phosphorylation were detected in mutant ileum compared to wild-type controls (Fig. 16A). To confirm these data, total Akt was immunoprecipitated from tissue lysates, and the phosphorylation state was determined by anti-phospho-Ser473 antibodies. Total Akt recovered from Brk mutant ileum showed increased Ser473 phosphorylation compared to wild-type animals (Fig. 16A). Ser-phosphorylation of Akt was detected in wild-type mice but it occurred at much lower levels compared to the knockout counterparts (Fig. 16A). Similarly, using a two-step in vitro kinase assay to determine Akt activity, only Akt precipitated from tissue homogenates of Brk deficient mice resulted in major phosphorylation of GSK-3, whereas GSK-3 phosphorylation levels in wild-type animals were markedly reduced (Fig. 16B).

↓44

Exposure of cells to ionizing radiation and a variety of other toxic stresses has been reported to induce simultaneous compensatory activation of MAPK signaling pathways. In addition to the radioprotective and growth-promoting signaling executed by the PI3-K/Akt pathway, ERK pathway signaling has been shown to be induced by radiation and to further stimulate an anti-apoptotic response (Dent, et al., 2003). To delineate the signaling pathways involved in the observed pro-survival response in Brk deficient mice, total tissue lysates were further analyzed by Western blotting. Lysates from distal ileum of untreated and irradiated (6 and 72 hour time-points) wild-type and Brk knockout mice were collected for Western blot analysis with anti-phospho-Erk1/2, anti-total-Erk1/2, anti-Brk and anti-β actin antibodies. Irradiation greatly stimulated activation of Erk1/2 MAPK in Brk deficient mice but not in wild-type counterparts at both time-points studied post treatment (Fig. 16C). Taken together with the previously described increased activation of Akt signaling in these mice, disruption of Brk expression stimulated anti-apoptotic Akt/PKB and Erk1/2 signaling in the gastrointestinal tract after γ-irradiation, contributing to and/or resulting in increased resistance to radiation-induced apoptosis.

Fig. 16: Pro-survival signaling in Brk deficient mice. Total cell lysates (TCL) from distal ileum of wild-type (WT) and knockout (Brk -/-) animals at 0, 6 and 72 hours after whole-body γ-irradiation (8 Gγ) were prepared. (A) Lysates from irradiated (6 hours) wild-type and knockout mice were analyzed by Western blotting and immunoprecipitation (IP). Endogenous Akt was immunoprecipitated and subjected to Western blotting with phospho-Akt specific antibodies. IP for IgG served as a control. Equal amounts of Akt were present in each immunoprecipitation as confirmed by immublotting for total Akt. Brk deficient mice exhibit increased phospho-Ser473 Akt. (B) Akt in vitro kinase assays. The lysates analyzed in (A) were subjected to Akt kinase assay. Endogenous Akt was immunoprecipitated with immobilized Akt antibody, and incubated with purified recombinant GSK-3 and ATP in kinase reaction buffer. Reactions were stopped at 0 (control or C) and 30 minutes and subjected to immunoblotting with Phospho-GSK-3β and total GSK-3β antibodies. Increased Akt kinase activity is detected in mice deficient for Brk. (C) Western blot analysis of TCL from untreated and irradiated (6 and 72 hours) wild-type and knockout mice with antibodies against phospho- and total Erk1/2. Expression of Brk and β-actin were examined as controls for genotype and protein loading respectively. Increased phosphorylation of Erk1/2 was observed in Brk deficient mice.

3.3.3 Induction of Brk in the intestine following ionizing radiation

It was previously reported that Brk protein levels are induced after DSS treatment. To determine whether irradiation induces Brk expression in intestinal epithelial cells in vivo, wild-type mice were exposed to whole-body γ-irradiation. The gastrointestinal tract was excised, total RNA and protein were prepared from distal ileum and analyzed by RNase protection assays and Western blotting. Tissue was also processed for immunohistochemistry.

↓45

Analysis of Brk mRNA levels using an anti-sense probe for Brk revealed no differences in Brk steady-state mRNA levels upon irradiation in wild-type animals (Fig. 17A). However, Brk protein levels rapidly increased after γ-irradiation in a time dependent manner detected by Western blotting with anti-Brk antibodies (Fig. 17B). To confirm these data, Brk protein expression in the small intestine of untreated and irradiated wild-type mice was furthermore examined by immunohistochemistry. Paraffin sections of the distal jejunum were immunostained with anti-Brk antibodies and counterstained with hematoxylin. As controls, sections were incubated with anti-rabbit IgG instead of anti-Brk antibodies. Untreated mice showed the previously described restriction of Brk protein expression to non-proliferating, terminally differentiated cells of the villus (Fig. 17C). In contrast, Brk protein expression was detected not only in the villus but also in proliferating cells of the crypt compartment in wild-type mice treated with ionizing radiation (Fig. 17C). The additional expression of Brk protein in cells, which usually exclude its presence, could account for the increased protein levels detected by Western blotting in irradiated wild-type mice. The rapid radiation-induced upregulation of Brk protein levels but not mRNA indicated that Brk regulation at the protein level is a determinant of cellular sensitivity to genotoxic stress.

Fig. 17: Brk protein expression is induced by γ-irradiation. (A) RNase protection assays were performed with total RNA from distal ileum of untreated and irradiated wild-type animals and 32P-labeled antisense probes specific for Brk and cyclophilin as control. (B) Western blot analysis of untreated and irradiated wild-type mice with lysates from distal ileum and antibodies against Brk and β-actin as a loading control. Increased Brk protein levels were detected in irradiated mice. (C) Immunohistochemical analysis of Brk expression on distal jejunum sections of untreated and irradiated wild-type animals. Immunostaining for IgG served as a control. Brk protein expression is restricted to the villus of untreated mice, whereas it is also detected in the crypt of irradiated mice. Size bar represents 50 µm.

3.3.4  Expression of Brk sensitizes cells to apoptosis in vitro

To understand the role of Brk in the regulation of differentiation and apoptosis, Rat1a fibroblasts stably overexpressing Brk were generated. Rat1a cells are immortal nontransformed rat embryo fibroblasts that are susceptible to oncogenic transformation and exhibit serum- and Akt-dependent susceptibility to a variety of apoptotic stimuli (Topp, 1981). They do not express endogenous Brk, but have been shown to take up DNA efficiently and to this extent, Rat1A cells were transfected with either retrovirus containing a BRK expression plasmid encoding for the wild-type or activated Brk tyrosine kinase (Brk WT or Brk Y-F) or the corresponding empty vector (pLXSN), followed by selection in G418. As expected, the correct Brk protein was expressed in the former but not in the latter stable populations (Fig. 18A). Ectopic Brk expression appeared to have no effect on mean cell size.

↓46

The growth properties of the established polyclonal Rat1A cell lines were examined by growth rate and FACS analysis. The cell cycle profiles of Rat1A fibroblasts overexpressing Brk were analyzed by flow cytometry and BrdU incorporation (Fig. 18B). Cells were synchronized in 0.1% FBS for 48 hours resulting in cell cycle arrest in the G1 phase of the cell cycle (Fig. 18B). Following cell cycle arrest, cells were released into normal growth medium containing 10% FBS and the amount of cells entering S-phase was monitored at various time points post stimulation. No difference was observed in the release into the cell cycle and the amount of cells entering S-phase in fibroblasts ectopically expressing wild-type or activated Brk and the control cell line (Fig. 18B). Interestingly, following cell cycle arrest by growth in 0.1% FBS for 48 hours, cell lines expressing wild-type or activated Brk exhibited more cells in the sub G1 phase of the cell cycle than the control line expressing vector only (Fig. 18B). The sub G1 phase represents apoptotic cells with nuclear fragmentation.

For growth rate determinations, cells of the respective stable populations were seeded at the same density, and cell numbers were determined daily for 5 days. No profound differences in growth were observed between fibroblasts overexpressing wild-type and activated Brk or the empty vector only (Fig. 18C). It appeared that their doubling time was similar. However, control cells grew to higher saturation density than wild-type and activated Brk expressing Rat1a cells, as was apparent with the growth curve at 5 days of growth (Fig. 18C). Rat1A cells undergo contact inhibition and G1 arrest when confluent. The decrease in cell number in Brk expressing populations once they reach confluence could be attributed to either a difference in cell morphology or increase in cell death compared to control populations. Brk has been shown to phosphorylate paxillin, and through this pathway it promotes cell motility and migration (Chen, et al., 2004). However, no clear difference in cell morphology was observed.

Fig. 18: Growth properties of stable Rat1A cells. (A) Western blot analysis of stable Rat1A fibroblasts. Total cell lysates of Rat1A cells stably overexpressing the expression vector pLXSN (Vector), wild-type Brk (Brk WT) and constitutively activated Brk (Brk Y-F) were analyzed by immunoblotting with anti-Brk antibodies and anti-β-actin antibodies as a control for protein loading. Rat1A cells do not express endogenous Brk. (B) Cell cycle analysis. Stable Rat1A populations were synchronized and arrested in G0 by serum starvation for 48 hours. Cells were released into G1 with the addition of 10% FBS, harvested at 0 and 16 hours post stimulation, and analyzed for cellular DNA content by flow cytometry. (C) Exponential growth curves of the indicated stable Rat1A populations. The total amount of cells per 6 cm plate was counted for 5 consecutive days. Values are means ± S.D. of 2 experiments with at least 3 plates per cell line and time point. No differences in growth and cell cycle were observed in Rat1A fibroblasts stably overexpressing Brk WT or Brk Y-F.

↓47

To investigate a possible role of Brk in apoptosis and cell survival in vitro, the effect of serum deprivation on stable Rat1A fibroblasts expressing wild-type and activated Brk or the expression vector alone was examined (Fig. 19A-D). Respective cells were seeded at the same density, serum starved in 0% FBS for 24 hours and fixed followed by DAPI staining. The amount of cells with condensed nuclei, typical of apoptotic cells, was counted and the data plotted (Fig. 19B). Clearly, overexpression of wild-type and a constitutively active form of Brk resulted in serum starvation-induced apoptosis. Serum deprivation resulted in rapid cell death of cell lines with ectopic Brk, with 15-20% cell death within 24 hours (Fig. 19B). In contrast, control cells stably expressing the empty expression vector exhibited nuclear condensation similar to untreated populations when exposed to serum deprivation (Fig. 19B). A background level of spontaneous apoptosis of approximately 5% was observed in all stable populations.

These data were confirmed by FACS analysis of untreated and serum starved Rat1A populations (Fig. 19C). The cell lines were treated as described above and the amount of apoptotic cells was determined by staining with propidium iodide and subjecting the cells to flow cytometry. Apoptotic cells are represented by cells in the sub G1 phase of the cell cycle due to nuclear fragmentation. All untreated cell populations exhibited a similar and minimal amount of cells in sub G1 consistent with the low background level of apoptosis observed earlier. However, after serum starvation, cells overexpressing Brk exhibited a significantly greater amount of cells in sub G1 compared to control cells (Fig. 19C). Consistent with the data acquired by DAPI staining, Rat1A fibroblasts with ectopic Brk seemed to have 3 times more apoptotic cells compared to the control cell population (Fig. 19B, C). Taken together these data suggested that Brk sensitizes Rat1A fibroblasts to serum starvation induced apoptosis.

Fig. 19: Brk sensitizes Rat1A fibroblasts to apoptosis induced by serum starvation or UV-irradiation/serum starvation. Stable Rat1A populations overexpressing vector alone (Vector), wild-type (Brk WT) and activated Brk (Brk Y-F) were plated at equal density and apoptosis was induced by 24 hours serum starvation or by serum starvation and ultraviolet irradiation (50 J/m2). Control cells were grown in 10% FBS. The percentage of cells with condensed chromatin was determined by DAPI staining. (A) Representative photomicrographs of DAPI-stained populations after 24 hours serum starvation. The arrows indicate cells with condensed chromatin (B) Histograms of percentage of apoptotic cells in untreated and serum starved populations. Percentage of apoptotic cells was determined microscopically after staining cells with DAPI. The values are means ± S.D. of three independent experiments. (C) FACS analysis of apoptosis. Untreated and serum starved populations were fixed, stained with propidium iodide (PI) and subjected to FACS analysis. The percentages of cells in sub G1 are indicated. (D) Cells were subjected to UV irradiation and at the time points indicated, chromatin condensation and the percentage of apoptotic cells were determined by DAPI staining. Values represent the average of three independent experiments ± S.D. Cell lines expressing Brk show an earlier onset of apoptosis and undergo apoptosis to a greater extent when compared to control cell lines (vector).

↓48

Another method to induce apoptosis in Rat1A fibroblasts in a controlled fashion is a combination of serum deprivation with a low dose of UV irradiation (Kennedy, et al., 1999). As has been previously shown, this combination accelerates growth factor-withdrawal-induced apoptosis and provides a convenient time frame for the analysis of the temporal sequence of apoptotic events. Serum deprivation alone induces apoptosis in Rat1A fibroblasts only after 24-36 hours. Furthermore, the dose of ultraviolet (UV) irradiation that is being used does not induce apoptosis in the presence of serum, whereas the combination of the two induces apoptosis after 5 hours (Kennedy, et al., 1999). To further characterize the sequential events leading to increased apoptosis in serum starved Brk expressing Rat1A cells, stable Rat1a fibroblasts were subjected to serum deprivation and UV exposure and analyzed. Following treatment, cells were fixed at the indicated time points, andthe amount of apoptotic cells was determined by DAPI staining (Fig. 19D). The results showed a background level similar to the previous experiment with approximately 5% apoptosis in all cell lines. However, an increased amount of apoptotic cells was observed already at 3 hours post treatment in wild-type and Brk Y-F expressing fibroblasts. In contrast, the number of apoptotic cells did not increase until later, at around 4 hours post treatment in the control cell line (Fig. 19D). This earlier onset of cell death in populations with ectopic Brk was coupled with an increased amount of cell death compared to control cells at all time points studied (Fig. 19D).

In conclusion, one clear phenotype emerged. Ectopic expression of Brk inhibited cell survival in conditions of serum starvation and UV exposure combined with serum starvation. However, Brk overexpression didn’t change the growth characteristics of these cells.Activation of apoptosis pathways may be the most important function of Brk in its tumor suppressor properties.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
08.11.2006