Substrates used for in vitro and in vivo deadenylation. Full-length chimeric mRNAs AT-GMCSF and GC-GMCSF contain the Xenopus globin 5′-UTR (66 nt), human β-globin coding region (450 nt), the indicated 3′-UTR (62 nt), and a poly(A) tail of 100 nt. Poly(A)⁻ transcripts AT-UTR and GC-UTR contain a short leader sequence (25 nt) followed by the UTR sequence (62 nt). Poly(A)⁺ transcripts AT-UTR, GC-UTR, 0AU, 2AU, 3AU, and 5AU contain a short leader sequence (45 nt), followed by the UTR sequence (62 or 35 nt) and a poly(A) tail of 100 nt.

In vitro deadenylation assay. (A) The indicated [α-³²P]UTP labeled and polyadenylated substrates (see Fig. 1) containing either a GpppG (lanes 1–5,11–14) or m⁷GpppG cap (lanes 5–10,15–18) were incubated at 22°C in activated egg extract. Aliquots were removed for analysis at the times indicated and fractionated by 6% PAGE as described in Materials and Methods. (B) [α-³²P]UTP-labeled, m⁷GpppG capped, and polyadenylated transcripts containing different numbers of tandem AUUUA repeats (see Fig. 1) were cytoplasmically injected into mature/activated eggs (top panel) or incubated in the in vitro system (bottom panel). RNA was isolated at the times indicated. (C) Data from B were quantitated by PhosphorImager and plotted as the % poly(A)⁻ (calculated as the cpm in the p(A)⁻ position divided by the total cpm between the p(A)⁺ and p(A)⁻ markers) as a function of time. (D) Fifty fmoles [α-³²P]UTP-labeled AT-UTR p(A)⁺ (top panel) or GC-UTR p(A)⁺ (bottom panel) were incubated in the presence of increasing concentrations of unlabeled competitor RNAs for 30 min in the in vitro system. Unlabeled competitor AT-UTR or GC-UTR was added at 1× (50 fmoles), 10× (500 fmoles), 10²× (5 pmoles), and 10³× (50 pmoles) relative to substrate, while homopolymeric poly(A) or poly(C) (Sigma) was added at 1× (2.5 ng), 10× (25 ng), 10²× (250 ng), and 10³× (2.5 μg).

Purification of the ARE-binding protein ePAB. (A) The activated egg extract (HSS) was loaded onto heparin sepharose, washed (W, final wash) and eluted with 1 M KCl (E). Equivalent fractions of these samples were analyzed by SDS-PAGE, and the gel was stained with Coomassie (left panel). Comparable fractions were also crosslinked to [α-³²P]UTP-labeled AT-UTR or GC-UTR transcripts (right panel). Arrows denote ARE-specific crosslinked bands, one of which (*) is ePAB. (B) Colloidal blue (Sigma) staining of a 15% SDS gel analysis of the KCl elution profile from the AT-UTR (AT) and GC-UTR (GC) AADA columns. The molarity at the start of the elution of each fraction is indicated at the top. RNase A was used to elute proteins remaining bound to the columns following the 2M KCl elution. Arrows indicate the three proteins bound specifically to the AT-UTR column. (C) Fractions from the heparin sepharose column (E is the 1M elution, P-E is the same sample after pre-clearing on AADA beads) and the ATUTR AADA column (FT is the flow through at 0.1 M KCl, and KCl elution fractions are from 0.3 to 1M) were UV-crosslinked to a [α-³²P]UTP-labeled AT-UTR (AT) (lanes 1,3,5,7–14) or GC-UTR (GC) (lanes 2,4,6) probe. Purified ePAB in lane 14 is indicated by an arrow.

DNA and predicted amino acid sequences of the Xenopus ePAB gene. (A) Underlined are amino acids identified by microsequence analysis of tryptic peptides. The 5′ untranslated region may not be complete, but there are two upstream UGAs in frame with the initiating methionine, arguing that the coding region is complete. In the 3′ untranslated region, a putative polyadenylation signal (minor class, AUUAAA) is located 12 nucleotides upstream of the poly(A) tail. (B) Amino acid sequences of Xenopus ePAB (629 aa), Xenopus (Xl) PABP1 (633 aa, accession no. P20965) (Zelus et al. 1989), human (Hs) PABP1 (636 aa, accession no. P11940) (Grange et al. 1987), and human (Hs) iPABP (644 aa) (accession no. NM003819) (Yang et al. 1995) are compared. Shaded regions denote identity. RBDs I, II, III and IV extend from human PABP1 positions 6 to 90, 92 to 176, 191 to 269, and 288 to 372, respectively (Burd et al. 1991).

ePAB expression during Xenopus early development. (A) Extract from various stages of development [1/2-egg equivalent per lane: stage VI oocyte, mature egg, 4-cell embryo (1.5 h), 16-cell embryo (2.5 h), morula (3.5 h), blastula (5 h), mid-blastula (7 h), gastrula (9 h), early neurula (12 h), 1-day embryo, and 30-h embryo] and XTC cells were fractionated by 15% PAGE and immunostained with anti-ePAB (third panel), monoclonal anti-PABP1 (top panel), or monoclonal anti-HuR, identifying ElrA (second panel), antibodies. The blot that was initially probed with anti-PABP1 was later reprobed for ePAB; the asterisk (*) marks the remaining signal from anti-PABP1, therefore indicating the mobility of PABP1 relative to ePAB. The same samples were also UV-crosslinked to [α-³²P]ATP labeled AT-UTR p(A)⁺ (fourth panel) and GC-UTR p(A)⁺ (bottom panel). (B) Egg extract was crosslinked to: [α-³²P]ATP-labeled AT-UTR p(A)⁺ (lane 1), [α-³²P]ATP-labeled GC-UTR p(A)⁺ (lane 4), or [α-³²P]UTP-labeled AT-UTR (lane 7). Crosslinked samples (Total) were immunoprecipitated with anti-ePAB (α-ePAB, lanes 3,6,9) or preimmune (α-Pre, lanes 2,5, or 8) serum. The arrow denotes the ∼65-kD crosslinked band precipitated by anti-ePAB. (C) Stage VI oocytes were manually enucleated and 1/2-oocyte equivalents of the total (Tot), cytoplasmic (cyt), and nuclear (GV) fractions were immunostained with anti-ePAB, monoclonal anti-PABP1, and monoclonal anti-HuR, identifying ElrA, antibodies.

Effect of trans-acting factors on in vitro deadenylation. (A) [α-³²P]UTP labeled AT-UTR p(A)+ (top panel) or GC-UTR p(A)+ (bottom panel) was incubated under in vitro assay conditions with the addition of buffer alone (lanes 1–4), or equivalent concentrations (∼10 ng/μL HSS) of recombinant 6xHis PABP1 (lanes 5–8), 6xHis ePAB (lanes 9–12), or 6xHis HuR (lanes 13–16), and aliquots were removed for analysis at the times indicated. (B) Western blot analyses of egg extracts immunodepleted (depl.) for ePAB (lanes 1–4), ElrA (lanes 5–8), or DAN (lanes 9–12). The bands corresponding to these as well as molecular weight marker proteins and the heavy chain of immunoglobulin (h) are indicated. In each case, the supernatant (Sup or S) and pellet (Pellet or P) fractions were analyzed, with the corresponding preimmune serum (Pre) used in parallel. (C) [α-³²P]UTP-labeled AT-UTR p(A)⁺ (top panel) or GC-UTR p(A)⁺ (bottom panel) substrate was incubated in preimmune-depleted (lanes 1–4) or α-ePAB-depleted (lanes 5–25) extract supplemented with: buffer alone (lanes 5–8,17–19), or equivalent concentrations (∼10 ng/μL HSS) of recombinant 6xHis ePAB (lanes 9–12,23–25), 6xHis HuR (lanes 13–16), or 6xHis PABP1 (lanes 20–22). Aliquots were withdrawn for RNA analysis at the times indicated. (D) [α-³²P]UTP-labeled AT-GMCSF p(A)⁺ or GC-GMCSF p(A)⁺ substrate was incubated in either ElrA-depleted (lanes 9–16) or preimmune-depleted extract (lanes 1–8) under deadenylation assay conditions and aliquots withdrawn for RNA analysis at the times indicated. (E) [α-³²P]UTP-labeled AT-GMCSF p(A)⁺ or GC-GMCSF p(A)⁺ substrate was incubated in either DAN-depleted extract (lanes 9–16) or preimmune-depleted extract (lanes 1–8) under deadenylation assay conditions for the times indicated and analyzed as above.