Supplemental Material

This article contains the following supporting material:

• Supplemental Figure S1 - Inhibition of cell growth by cyclin A shRNA.
  (A) Cyclin A shRNA suppresses the expression of exogenous cyclin A. HtTA1 cells were cotransfected with constructs expressing FLAG-tagged cyclin A and CDK2. Control vector or a plasmid expressing cyclin A shRNA was cotransfected as indicated. Cell-free extracts were prepared and subjected to immunoblotting with an antibody against the FLAG tag.
  (B) Cyclin A-depleted cells fail to form colonies. Cells were transfected with empty vector or cyclin A shRNA plasmids. The cells were selected transiently with blasticidine as described in Materials and Methods, and 500 cells were plated on a 60-mm plate and allowed to grow for 10 days. Colonies were visualized after staining with crystal violet.
• Supplemental Figure S10 - Down-regulation of cyclin A reduced the activities of CDC25A and CDC25B.
  Cells were transfected with control vector or plasmids expressing shRNA against cyclin A or cyclin B. The cells were treated with buffer or nocodazole for 16 hr before harvested. Cell-free extracts were prepared and the phosphatase activities of CDC25A or CDC25B were detected by the activation of WEE1-treated cyclin B-CDC2 as described in Materials and Methods.
• Supplemental Figure S11 - Cyclin shRNA-induced G₂ arrest cannot be overcome by overexpression of CDC25s or PLK1.
  (A) Neither CDC25A nor CDC25B can drive mitosis in the absence of cyclin A or cyclin B. Control, cyclin A shRNA, or cyclin B shRNA was coexpressed with empty vector, HA-CDC25A, or HA-CDC25B as indicated. The cells were trapped in mitosis with nocodazole for 12 hr before harvested (except lanes 1-3). Cell-free extracts were prepared and the expression of the indicated proteins was detected by immunoblotting. The recombinant CDC25A and CDC25B were detected with an antibody against HA.
  (B) Neither wild type CDC25C nor CDC25C^S216A can drive mitosis in the absence of cyclin A or cyclin B. The experiment was performed similar to panel (A) except that HA-tagged CDC25C or CDC25C^S216A were used.
  (C) PLK1 cannot drive mitosis in the absence of cyclin A or cyclin B. The experiment was performed similar to panel (A) except that PLK1 or PLK1^T210D were used.
• Supplemental Figure S12 - Uncoupling the DNA integrity checkpoints by CDC25.
  (A) Overexpression of CDC25A and CDC25B can uncouple the DNA damage checkpoint and replication checkpoint. HtTA1 cells were transfected with control vector or plasmids expressing HA-tagged CDC25A or CDC25B. The cells were exposed to buffer, Adriamycin (ADR), or hydroxyurea (HU) for 18 hr as indicated. Cell-free extracts were prepared and the expression of HA-tagged proteins and phospho-histone H3^Ser10 was detected by immunoblotting.
  (B) CDC25C^S216A can uncouple the DNA damage checkpoint and replication checkpoint. The experiment was performed as in panel (A) except that plasmids expressing FLAG-tagged CDC25C or CDC25C^S216A were transfected. Cell-free extracts were prepared and the expression of FLAG-tagged proteins and phospho-histone H3^Ser10 was detected by immunoblotting.
• Supplemental Figure S2 - Knockdown of cyclin A does not affect cyclin B accumulation.
  HtTA1 cells were transfected either control vector or cyclin A shRNA plasmids, and subjected to a double thymidine synchronization. Nocodazole was added after the release from the second thymidine to trap cells in mitosis. The cells were harvested at the indicated time points. Extracts were prepared and the expression of cyclin A, cyclin B, and phospho-histone H3^Ser10 was detected by immunoblotting.
• Supplemental Figure S3 - Rescue of shRNA-induced G2 arrest with cyclin A.
  Cyclin A shRNA-expressing plasmids were cotransfected with GFP-histone H2B and an increasing amount of plasmids expressing FLAG-cyclin A as indicated. The cyclin A construct contained silence mutations that rendered resistance to the shRNA. The cell cycle distribution of transfected (GFP-positive) cells was analyzed by flow cytometry.
• Supplemental Figure S4 - Conditional rescue of cyclin A in stable knockdown cell lines reveals the requirement of cyclin A for G2/M control.
  (A) Schematic diagram of cyclin A/shRNA construct. Part of the cyclin A/shRNA in pKAR1 is shown to scale: TRE, tetracycline response element; P_minCMV, minimal immediate early cytomegalovirus (CMV) promoter; Puromycin, puromycin-resistant gene (which carried its own SV40 early promoter and SV40 polyadenylation signal), SV40 pA, SV40 polyadenylation signal; and P_U6, mouse U6 RNA promoter.
  (B) Repression of cyclin A by doxycycline in stable cyclin A/shRNA-expressing cells. Stable cyclin A/shRNA-expressing HtTA1 cells were either mock-treated (lane 2) or treated with doxycycline (lanes 3 and 4, 0.2 ng/ml and 2 ng/ml, respectively) for 36 hr. Cell-free extracts were prepared and the expression of cyclin A and cyclin B was detected by immunoblotting. Extracts from control cells were loaded in lane 1 and CDC2 analysis was included to assess protein loading and transfer.
  (C) Knockdown of cyclin A delays the cell cycle at G₂/M. Stable cyclin A/shRNA-expressing cells were either mock-treated or treated with doxycycline for 36 hr. The cells were then fixed, stained with propidium iodide, and processed for flow cytometry analysis. The positions of the 2N and 4N DNA contents are indicated.
• Supplemental Figure S5 - Mitotic slippage does not occur in cyclin A knockdown cells.
  HtTA1 cells were transfected with empty vector (lane 3) or cyclin A shRNA plasmids (lane 4). The cells were trapped in mitosis with nocodazole for 12 hr before harvested. Cell-free extracts were prepared and the expression of the indicated proteins was detected by immunoblotting. Extracts from untreated control cells, mitotic cells (treated with nocodazole for 16 hr), and G₁ cells (at 6 hr after released from nocodazole block) were loaded as controls.
• Supplemental Figure S6 - Intact nuclear envelope in cells arrested by shRNAs against cyclin A and cyclin B.
  HtTA1 cells were transfected with control vectors or plasmids expressing shRNA against cyclin A or cyclin B. Transfected cells were transiently selected for a cotransfected puromycin-resistant gene-expressing plasmid for 36 hr. The cells were then fixed and subjected to immunostaining with a polyclonal antibody against lamin A/C, followed by a TRITC-conjugated anti-rabbit IgG secondary antibody (red). Nuclei were counter-stained with Hoechst 33258 (blue). Representative images are shown and the arrows indicate the position of a mitotic cell without the nuclear envelope.
• Supplemental Figure S7 - Down-regulation of cyclin A does not induce redistribution of p21^CIP1/WAF1 to cyclin B-CDC2.
  (A) Binding of ectopically expressed p21^CIP1/WAF1 to cyclin B-CDC2 does not increase after cyclin A knockdown. Control or cyclin A shRNA was cotransfected with either control vector or FLAG-p21^CIP1/WAF1-expressing constructs in the presence of doxycycline. At 48 hr after transfection, the expression of FLAG-p21^CIP1/WAF1 was induced by the removal of doxycycline. Cell lysates were prepared after 18 hr and subjected to immunoprecipitation with a p21^CIP1/WAF1 antiserum. The presence of the indicated proteins in total lysates and in the immunoprecipitates were detected by immunoblotting.
  (B) Binding of endogenous p21^CIP1/WAF1 to cyclin B-CDC2 does not increase after cyclin A knockdown in U2OS cells. Stable cyclin A/shRNA-expressing U2OS/Tet-On cells were either mock-treated or treated with doxycycline for 36 hr (the shRNA-resistant cyclin A was expressed only in the presence of doxycycline in these cells). Cell lysates were prepared and subjected to immunoprecipitation with either normal rabbit serum (NRS) or p21^CIP1/WAF1 antiserum. The presence of the indicated proteins in total lysates and in the immunoprecipitates were detected by immunoblotting.
• Supplemental Figure S8 - Figure S8. Cyclin A shRNA-induced cell cycle arrest is independent of the DNA integrity checkpoint pathways.
  (A) Expression of cyclin A shRNA does not activate CHK1. HtTA1 cells were transfected with empty vector (lane 4) or cyclin A shRNA plasmids (lane 5). Cell-free extracts were prepared and subjected to immunoblotting with antibodies against CHK1, cyclin A, cyclin B, and CDC2. Lysates from cells treated with buffer, Adriamycin (ADR), or hydroxyurea (HU) for 16 hr were loaded as controls.
  (B) Knockdown of cyclin A or cyclin B does not affect the DNA damage checkpoint machinery. Cells were transfected empty vector, cyclin A shRNA, or cyclin B shRNA plasmids. The cells were then either mock-treated or irradiated with 15 Gy of IR. Cell lysates were prepared at 3 hr after irradiation and subjected to immunoblotting with antibodies against phospho-ATM^Ser1981, phospho-CHK1^Ser345, and phospho-CHK2^Thr68.
  (C) Mitotic failure in cyclin-disrupted cells are insensitive to caffeine. Cells were transfected empty vector, cyclin A shRNA-, or cyclin B shRNA-expressing plasmids, and then treated with buffer, Adriamycin (16 hr), caffeine (12 hr), or Adriamycin (16 hr) followed by caffeine (12 hr). Cell-free extracts were prepared and the expression of the indicated proteins was detected by immunoblotting.
• Supplemental Figure S9 - Cyclin A knockdown-induced cell cycle arrest is independent of the p38 stress response pathway.
  (A) Expression of cyclin A shRNA does not activate p38. HtTA1 cells were transfected with empty vector or cyclin A shRNA plasmids. Cell-free extracts were prepared and subjected to immunoblotting with antibodies against phospho-p38^{Thr180/Thr182} and total p38. Lysates from cells treated with nocodazole for the indicated time were loaded as controls.
  (B) Inhibition of p38 does not uncouple the cyclin A shRNA-induced cell cycle arrest. Cells transfected with control vector or cyclin A shRNA plasmids were treated with buffer or SB203580 as indicated. Nocodazole was added at the same time to trap cells in mitosis. After 14 hr, cell lysates were prepared and the indicated proteins were detected by immunoblotting.