Centrosome misorientation mediates slowing of the cell cycle under limited nutrient conditions in Drosophila male germline stem cells

Abstract

Drosophila male germline stem cells (GSCs) divide asymmetrically, balancing self-renewal and differentiation. Although asymmetric stem cell division balances between self-renewal and differentiation, it does not dictate how frequently differentiating cells must be produced. In male GSCs, asymmetric GSC division is achieved by stereotyped positioning of the centrosome with respect to the stem cell niche. Recently we showed that the centrosome orientation checkpoint monitors the correct centrosome orientation to ensure an asymmetric outcome of the GSC division. When GSC centrosomes are not correctly oriented with respect to the niche, GSC cell cycle is arrested/delayed until the correct centrosome orientation is reacquired. Here we show that induction of centrosome misorientation upon culture in poor nutrient conditions mediates slowing of GSC cell proliferation via activation of the centrosome orientation checkpoint. Consistently, inactivation of the centrosome orientation checkpoint leads to lack of cell cycle slowdown even under poor nutrient conditions. We propose that centrosome misorientation serves as a mediator that transduces nutrient information into stem cell proliferation, providing a previously unappreciated mechanism of stem cell regulation in response to nutrient conditions.

FIGURE 1:

Germline stem cell centrosome orientation is modulated by nutrient availability. (a) Schematic diagram of centrosome movement during the GSC cell cycle. (b) Centrosome misorientation frequencies in GSCs from flies cultured in rich or poor media. Numerical data are presented as the mean ± SD in all figures. n > 300/data point. Flies were cultured in the indicated media until the adult stage and then kept in/transferred to the indicated media at day 0. (c) An example of an apical testis tip from a fly cultured in poor media. Although many interphase GSCs (white circles) are misoriented, the mitotic spindle is correctly oriented (yellow circle). Blue, Vasa (germ cells); green, phosphorylated histone H3 (pH3, mitotic chromosomes); red, γ-tubulin (centrosomes). The hub is marked by an asterisk. Scale bar, 10 μm. (d) Bromodeoxyuridine incorporation in GSCs under poor or rich media conditions. n > 100/data point.

FIGURE 2:

Nutrient-poor media do not increase dedifferentiation of germline cells. (a) An example of an apical testis tip with a dedifferentiated GSC, as observed by LacZ expression. LacZ expression was induced by Bam-gal4/UAS-FLP–mediated activation of the LacZ gene, thus marking cells that had once committed to differentiation (Cheng et al., 2008). Green, Vasa (germ cells); red, LacZ; blue, 4′,6-diamidino-2-phenylindole. Hub, asterisk. Scale bar, 10 μm. (b) Frequency of LacZ-positive GSCs in rich and poor media (mean ± SD). (c) An example of apical testes tip containing dedifferentiating GSCs and spermatogonia that are still connected to one another. Green, Pavarotti-GFP (contractile ring and ring canal); red, adducin-like (spectrosome/fusome); blue, Vasa. (d) Frequency of dedifferentiation in rich and poor media (mean ± SD). Flies raised in rich media were transferred to new vials containing either rich or poor media, and testes were examined 2 or 3 d later. (e) Dedifferentiated GSCs do not recover proper centrosome orientation after a prolonged time period. Young adult hs-Bam flies (day 0) were subjected to five heat shocks (30 min each) over the course of 2 d (Sheng et al., 2009) and then cultured at 25°C. GSC centrosome orientation was scored at the indicated time.

FIGURE 3:

Insulin signaling regulates GSC centrosome orientation. (a) Centrosome misorientation frequency upon overexpression of a dominant-negative (K1409A) or a constitutively active form (Del, R418P, or A1325D) of InR using nos-gal4. n > 300/data point. Siblings from the same cross without the nos-gal4 driver served as controls. p values (Student's t test, two-tailed) are provided on each column compared with its control. Statistically significant values (p < 0.01) are highlighted with asterisks. (b) Akt functions downstream of InR in regulating centrosome orientation. n > 300/data point. (c) Local expression of dilp1, 2, 3, 5, and 6 in the testis reduces centrosome misorientation in poor media. Each dilp was expressed in the testis (nos-gal4>UAS-dilp).

FIGURE 4:

Apc2 mediates centrosome orientation in response to nutrients. (a, b) Representative Apc2 staining in apical testes tips from flies raised in rich (a) or poor (b) media. Cortical Apc2 localization is indicated with yellow lines. Cytoplasmic punctae of Apc2 are indicated with yellow arrowheads. Green, Apc2; blue, Vasa. (c) Quantification of Apc2 localization in rich or poor media. “Cortical” indicates Apc2 protein at the hub-GSC junction; “diffuse/puncta” indicates Apc2 in the cytoplasm or occasionally at the GSC cortex outside the hub-GSC interface. n = GSCs scored. (d) Pixel intensity analysis of Apc2 protein around the GSC cortex. Circumference of GSC was traced and the pixel intensity was analyzed (see Materials and Methods for detail). Fourteen GSCs from the rich media and 19 GSCs from the poor media were analyzed. Average pixel intensity for rich vs. poor media is shown at the far right. (e) Localization of GFP-Apc2 to the hub-GSC interface following mild expression (nos-gal4>UAS-GFP-Apc2; at 18°C) in poor media. Green, GFP-Apc2; blue, Vasa. (f) The frequency of GSC centrosome misorientation upon mild expression of GFP-Apc2. n > 300/data point. (g) The frequency of GSC centrosome misorientation in the InR mutant that expresses Apc2. n > 300/data point. (h) The frequency of GSC centrosome misorientation in the apc2 mutant. n > 300/data point. (i) The S-phase index (BrdU incorporation) of apc2 mutant GSCs. n > 300/data point.

FIGURE 5:

Apc2 localization is regulated by the insulin signaling pathway. (a–d) Examples of Apc2 staining in apical testes tips from flies that express a dominant-negative (K1409A) or constitutively active (A1325D) form of InR in rich vs. poor media. Cortical localization of Apc2 is indicated with yellow lines. Cytoplasmic punctae of Apc2 are indicated with yellow arrowheads. Green, Apc2; blue, Vasa (germ cells). (e) Quantification of Apc2 localization upon overexpression of a dominant-negative (K1409A) or constitutively active (A1325D) form of InR, Akt-RNAi, or Akt-HA in rich or poor media (mean ± SD). n > 100 testes/data point.

FIGURE 6:

An intact centrosome orientation checkpoint is required to slow the GSC cell cycle in poor media. (a) Centrosome misorientation frequency in GSCs mutant for cnn or expressing dominant-negative E-cadherin (dCR4h). n > 150/data point. p value (Student's t test, two-tailed) comparing centrosome misorientation in rich vs. poor media is shown. n.s., not statistically significant. (b) The S-phase index (BrdU incorporation frequency) in GSCs from checkpoint-defective mutants and those expressing InR^A1325D. n > 450/data point. p value (Student's t test, two-tailed) comparing BrdU incorporation in rich vs. poor media is shown. n.s., not statistically significant. (c) Changes in GSC numbers in testes from control genotypes, cnn mutants, or those expressing dCR4h or InR^A1325D after 10 d in rich vs. poor media. p value (Student's t test, two-tailed) comparing the GSC number in rich vs. poor media after 10 d is shown. n.s., not statistically significant. Asterisk indicates that a mild but statistically significant decrease in the GSC number was observed in this control group for unknown reasons. However, the degree of GSC loss was significantly (p < 0.01) more severe in InR^A1325D-expressing testis compared with the control. n = 70–100 testes/data point. (d) Model of the regulation of the GSC division rate by nutrient availability and insulin signaling (see the text for details).