Protein Kinase D1 (PKD1) Signaling Induces Growth-Promoting Effects in Murine Enteroids.

The mammalian intestine, covered by a single layer of epithelial cells, renews every 4–6 days throughout adult life. This high rate of turnover plays an essential role in the organization, maintenance, function, and restoration of intestinal tissue integrity. The ultimate source of cells are multipotent intestinal stem cells (ISCs) that reside in the lower regions of intestinal crypts. Despite major advances in the identification of rapidly dividing ISCs and their regulation through growth factors and developmental cues from the microenvironment,1, 2, 3, 4 the intracellular signal transduction mechanisms remain incompletely understood.

Protein kinase D1 (PKD1), an evolutionarily conserved protein kinase family, has emerged as a key node in cellular signaling., By using transgenic mice that overexpress PKD1 in the intestine (PKD1-Tg), we showed that PKD1 induces crypt cell hyperproliferation and leads to a change in tissue architecture, manifested by an increase in the size and total number of epithelial cells in intestinal crypts., These findings prompted us to hypothesize that PKD1 controls the number and function of ISCs, a proposition that remained untested. Here, we determined the impact of PKD1 activity on the function of ISCs by assessing the capacity of isolated intestinal crypts to form 3-dimensional enteroids.,

We isolated crypts from PKD1-Tg and non-Tg littermates and plated them in Matrigel (Corning, Tewksbury, MA) containing R-spondin, noggin, and epidermal growth factor. Initially, we verified that PKD1 expression and activity were markedly more pronounced in lysates of PKD1-Tg enteroids, as shown by Western blotting with antibodies that detect total PKD1 and catalytically active PKD1, phosphorylated at the activation loop Ser744/748 or at the C-terminal autophosphorylation site Ser916 (Figure 1A). Enteroids generated from PKD1-Tg mice show a significant increase in area compared with enteroids generated from crypts isolated from non-Tg mice (Figure 1B and Supplementary Figure 1). After 7 days, the area of enteroids formed from PKD1-Tg mice was 58 ± 2.3 × 103 μm2 compared with 42 ± 3.5 × 103 μm2 in enteroids from non-Tg mice (Figure 1C). Similarly, the number of crypt-like buds, which contain the stem cell compartment, was 8.1 ± 0.49 in enteroids from PKD1-Tg and 5.5 ± 0.31 in those from non-Tg mice (Figure 1D). Importantly, we obtained similar results with enteroids passaged 4 times in culture (Figure 1B–D), or with enteroids prepared from primary crypts from PKD1-Tg and non-Tg mice (Figure 1E–G). Given that the impact of PKD1 overexpression on enteroid morphology persists after serial subcultures, it is likely to be intrinsic to the epithelial cells.

Figure 1

PKD1 overexpression increases the area and complexity of enteroids. (A) Western blot analysis of PKD1 expression and activity. (B–G) Representative images and quantification of the size and bud number of enteroids derived from wild-type (WT, open bars) and PKD1-Tg (TG, solid bars) crypts in either (B–D) passage 4 enteroids or (E–G) primary enteroids. Values are means ± SEM of at least 30 enteroids. Similar results were obtained in (C and D) 18 and (F and G) 7 independent experiments. Scale bars: 100 μm. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. ∗P < .05; ∗∗P < .01; TG vs WT.

Supplementary Figure 1

( Representative low-magnification images of enteroids derived from the ileum of wild-type (WT) or PKD1 Tg (A) at passage 4 or (B) primary enteroids. These images complement the high-magnification images of enteroids shown in Figure 1. (C and D) Impact of PKD1 overexpression on abundance of ISCs and Paneth cells. Additional representative (C) olfactomedin 4 (Olfm4) and (D) lysozyme staining from WT and PKD1-Tg mice–derived enteroids at passage 4. Hoechst staining is overexposed to show nuclei in the enteroids. Boxes represent the portion of enteroids used for quantification and shown in Figure 2. Similar results were obtained in 4 independent experiments.

Because PKD1 overexpression increases enteroid size and complexity, we next determined whether acute PKD inhibition produces the opposite effect. The structurally unrelated PKD family inhibitors CRT0066101 (Supplementary Figure 2) and kb NB 142-70 (Supplementary Figure 3) markedly inhibited the increase in enteroid area and budding of crypt-like structures in enteroids derived from either PKD1-Tg or non-Tg mice in a dose-dependent manner.

Supplementary Figure 2

The selective PKD family inhibitor CRT0066101 (CRT) reduced the size and the number of crypt-like buds in enteroids from PKD1-Tg and non-Tg mice. (A) Crypts isolated from the ileum of PKD1-Tg and non-Tg mice were cultured in the 3-dimensional system, as described in the Supplementary Materials and Methods section, either in the absence or presence of increasing concentrations of the selective PKD family inhibitor CRT0066101. For enteroid area measurements, 5–8 (4×) or 10–12 (10×) nonoverlapping images were acquired using an inverted microscope (Axio Observer.A1; Zeiss). Enteroid surface areas were defined manually by outlining the perimeter with the polygon function in ImageJ software (National Institutes of Health). The same images were used to count the number of buds per enteroid. Note that addition of CRT0066101 strikingly decreased the area of enteroids grown for 6 days in a dose-dependent manner (typical images at low and high magnification are shown in panel A, quantification is shown in panel B). At a concentration as low as 0.1 μmol/L, CRT0066101 significantly reduced the area of enteroids isolated from PKD1-Tg and non-Tg littermates and abolished the increase in area shown by enteroids of PKD-Tg. The addition of CRT0066101 also reduced the number of crypt-like buds from the enteroids (images are shown in panel A, quantification is shown in panel C). Similar effects of CRT0066101 on the area and number of crypt-like buds were obtained with enteroids treated for 3 days with CRT0066101 (typical images at low and high magnification are shown in panel D, quantification is shown in panels E and F) instead of 6 days in culture. Statistical analysis was as described in the Materials and Methods section. ∗P < .05, ∗∗P < .01, as compared with untreated control. The bars represent the means ± SEM; (B and C) n = 50 and (E and F) n = 30. Similar results were obtained in 3 independent experiments. Scale bars: 100 μm. (B) Insert: Western blot of PKD1 activity. Enteroids were treated on day 2 of culture without or with 1 μmol/L CRT0066101 for 24 hours, the cultures then were lysed and analyzed by Western blot with phospho-PKD Ser916 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Note that CRT0066101 markedly inhibited PKD activity (scored by autophosphorylation on Ser916) when added to the growing enteroids. Similar results were obtained in 5 independent experiments. p-PKD, phospho-Protein Kinase D; WT, wild-type.

Supplementary Figure 3

Impact of kb NB 142-70, a PKD family inhibitor structurally unrelated to CRT0066101, on enteroid size and complexity. Crypts isolated from PKD1-Tg and non-Tg mice were cultured in the 3-dimensional system in the absence or presence of increasing concentrations of kb NB 142-70, as described in the Materials and Methods section. Measurements of the enteroid area and number of buds were performed as described in the Materials and Methods section and in Supplementary Figure 2. The addition of kb NB 142-70 markedly decreased the area of enteroids grown for 6 days in a dose-dependent manner (typical images at low and high magnification are shown in panel A, quantification is shown in panel B). (C) The addition of kb NB 142-70 also reduced the number of crypt-like buds from the enteroids. Similar effects of kb NB 142-70 on the area and number of crypt-like buds were obtained with enteroids treated for 3 days (typical images at low and high magnification are shown in panel D, quantification is shown in panels E and F) instead of 6 days in culture. ∗P < .05, ∗∗P < .01, as compared with untreated control. Bars represent the means ± SEM; n = 40. Scale bars: 100 μm. Similar results were obtained in 3 independent experiments. WT, wild-type.

Next, we investigated the effect of PKD1 activity on the proliferation of enteroid cells, as scored by 5-ethynyl-2′-deoxyuridine incorporation. Consistent with crypt hyperproliferation in vivo,, enteroids grown from crypts isolated from PKD1-Tg mice showed a 54% and 50% increase in the proportion of cells in DNA synthesis on days 3 and 6 of culture, respectively (Figure 2A and B). Conversely, PKD inhibition with CRT0066101 blunted DNA synthesis (Figure 2A and B) in a reversible manner (Supplementary Figure 4). Thus, the increase in cell proliferation likely explains the increase in size of enteroids derived from PKD1-Tg mice.

Figure 2

Impact of PKD1 overexpression on cell proliferation and abundance of ISCs and Paneth cells. (A) 5-Ethynyl-2′-deoxyuridine (EdU) incorporation in enteroids derived from wild-type (WT, open bars) and PKD1-Tg (TG, solid bars) crypts in the absence or presence of CRT0066101 (1.0 μmol/L, added 24 h before EdU). (B) Values are means ± SEM; n = 10 buds. Similar results were obtained in 3 (day 3) and 4 (day 6) independent experiments. Supplementary Figure 4 shows a different assay of EdU incorporation. (C–F) Representative images of (C and E) olfactomedin 4 (Olfm4) and (D and F) lysozyme staining from (C and D) WT (open bars) and PKD1-Tg (solid bars) mice–derived enteroids at passage 4, or (E and F) freshly isolated crypts. Bars represent means ± SEM; n = 7–8 buds. Similar results were obtained in 4 independent experiments. ∗P < .05; ∗∗P < .01; TG vs WT.

Supplementary Figure 4

The inhibitory effect of CRT0066101 on DNA synthesis in enteroids is reversible. (A) Fluorescent ratios of green (5-ethynyl-2′-deoxyuridine)/blue (Hoechst 33342) were determined with Photoshop (Adobe, San Jose, CA) using the histogram function to determine the mean fluorescent values of green and blue from confocal images. All images were captured with identical exposures in each experiment. The values shown represent the mean ratio ± SEM; n = 10 buds. Similar results were obtained in 4 independent experiments. ∗P < .05; ∗∗P < .01, PKD1-TG (TG, solid bars) vs Wild Type (WT, open bars). (B) Typical images of enteroids from PKD1-Tg mice grown for 1, 3, or 6 days in the absence (control [cont] or presence of 1 μmol/L CRT0066101 for 1 or 3 days (CRT day 1 and CRT day 3, respectively). Parallel enteroids exposed to 1 μmol/L CRT0066101 for 3 days were washed and transferred to medium without the inhibitor for 3 additional days (CRT days 1–3, day 6). Similar results were obtained in 4 independent experiments. Scale bars: 100 μm. (C) In line with the results shown, addition of CRT0066101 (1 μmol/L) markedly inhibited the incorporation of EdU into epithelial cells of the crypt-like domain of enteroids derived from PKD1-Tg mice (compare panels B and A at day 3). After 3 days of exposure to CRT0066101, the media of parallel cultures of enteroids were replaced with fresh media supplemented with or without 1 μmol/L CRT0066101. Enteroids continuously exposed to CRT0066101 for 6 days show a striking inhibition of EdU incorporation (compare panels E and C at day 6). In contrast, enteroids exposed to CRT0066101 for 3 days followed by 3 days without the inhibitor showed a striking recovery of DNA synthesis (compare panels D and C at day 6). (D) Percentage of EdU-labeled cells shown in panel C. Values are means ± SEM; n = 10 buds. Similar results were obtained in 3 independent experiments. (E) Ratios of green/blue fluorescence. Values are means ± SEM; n = 10 buds. Similar results were obtained in 3 independent experiments. Collectively, the results indicate that CRT0066101 reversibly arrests DNA synthesis in enteroids of PKD1-Tg mice.

We hypothesized that PKD1 hyperactivity increases the number and function of ISCs in the crypt-like structures, thereby enhancing enteroid size and complexity. Immunofluorescence analysis for olfactomedin 4, a marker of rapidly dividing ISCs,, showed an increase in ISCs in the crypt-like domains of enteroids generated from PKD1-Tg mice, as compared with those derived from non-Tg mice (Figure 2C and Supplementary Figure 1). Paneth cells, interspersed among ISCs in the crypts of the small intestine, produce and secrete growth-promoting factors that contribute to form a microenvironment propitious for ISC proliferation. Therefore, PKD1 could increase the number of ISCs in the crypt-like domains of enteroids through an increase in Paneth abundance. Against this possibility, our results show that PKD1 overexpression lead to a reduced quantity of Paneth cells in the crypt-like domains of enteroids (Figure 2D and Supplementary Figure 1). We verified a similar increase in olfactomedin 4–positive cells and a decrease in Paneth cells in primary crypts isolated from PKD1-Tg mice (Figure 2E and F), indicating that the increase in ISCs and the decrease in Paneth cell numbers found in enteroids is not an artifact of the culture conditions. Collectively, our results show that PKD1-mediated signaling increases enteroid area and complexity and modulates the balance of ISCs and Paneth cells. The findings indicate that PKD1 contributes to the regulation of ISC proliferation and differentiation.