Pleural relapse during hematopoietic remission in childhood acute lymphoblastic leukemia.

PURPOSE: This report describes an isolated pleural relapse during hematopoietic remission in a child previously treated for acute lymphoblastic leukemia (ALL). PATIENT AND METHODS: An 11-year-old boy had a cough and exertional dyspnea 34 months after an initial diagnosis of ALL and 10 months after completion of therapy. Imaging studies revealed a large left pleural effusion. Bone marrow and cerebrospinal fluid studies were negative for disease at this time. RESULTS: Histopathologic examination of biopsy samples revealed cells with morphologic features of acute lymphoblastic leukemia blasts. Immunophenotyping, cytogenetic, and gene rearrangement studies confirmed the presence of a leukemic blast cell population similar to that detected at initial diagnosis. An isolated extramedullary relapse in the pleura was diagnosed. The patient underwent successful reinduction therapy and subsequently a matched unrelated donor bone marrow transplant; he died of disseminated infection in the posttransplant period. CONCLUSIONS: Unusual extramedullary sites of relapse are recognized with increasing frequency as long-term survival in childhood ALL improves. The length of the disease-free interval before relapse is felt to be of prognostic significance. Isolated relapse to the pleural space has not previously been described. The mechanism for persistence of leukemic clones in patients who appear to be in hematopoietic remission is unknown.

In vitro analysis of translational rate and accuracy with an unmodified tRNA.

Escherichia coli tRNA(Phe) transcript lacking all the modified nucleosides was investigated in an in vitro translation system. To estimate the affinity of tRNA toward EF-Tu, Kd and K-1 were measured by the nuclease protection assay, and it was shown that the absence of modifications decreases ternary complex stability less than 2-fold. The activity of unmodified Phe-tRNA(Phe) on E. coli ribosomes was compared to modified Phe-tRNA(Phe) using the framework of the kinetic proofreading mechanism (Thompson & Dix, 1982) with both cognate and noncognate codons. Values of the individual rate constants in the elongation process showed that the modifications increased the accuracy of translation by (1) decreasing the rate of dipeptide synthesis and (2) increasing the rate of rejection with noncognate codons.

Codon choice and gene expression: synonymous codons differ in translational accuracy.

Ribosomes programmed by different synonymous codons also differ in discriminating among near-cognate aminoacylated tRNAs. In the initial step of the recognition reaction ribosomes programmed by UUC discriminate less well than ribosomes programmed by UUU against ternary complexes containing three types of Leu-tRNA, and ribosomes programmed by CUC discriminate less well than ribosomes programmed by CUU against ternary complexes containing Phe-tRNA. Furthermore, in the proofreading step ribosomes programmed by UUC discriminate less well than ribosomes programmed by UUU against two of three near-cognate Leu-tRNAs, and ribosomes programmed by CUC discriminate less well than ribosomes programmed by CUU against near-cognate Phe-tRNA. The codon-induced change in reaction rate with near-cognate ternary complexes is greater than that with cognate ternary complexes: the most efficient codon is, therefore, the least accurate. Because the efficient, but inaccurate, codon UUC is used preferentially in highly expressed mRNAs of Escherichia coli, maximization of translational accuracy apparently has not been significant in the evolution of this particular biased codon choice in E. coli.

Codon choice and gene expression: synonymous codons differ in their ability to direct aminoacylated-transfer RNA binding to ribosomes in vitro.

Phe-tRNA (anticodon GAA)--polypeptide-chain elongation factor Tu-GTP ternary complexes react faster with ribosomes programmed with UUC codons than with ribosomes programmed with UUU codons. A similar preference is shown by Leu-tRNA2 (anticodon GAG) complexes, which react faster with ribosomes programmed with CUC than with those programmed with CUU. The difference is seen in the rate of ternary-complex binding to the ribosome; no differences are seen in peptide-bond formation. Highly expressed mRNAs in Escherichia coli favor codons terminating in cytosine rather than uracil when both codons are read by a single tRNA with an anticodon beginning with guanine. The results suggest that intrinsic differences between the efficiencies of synonymous codons play an important role in modulating gene expression in E. coli.

Effect of replacing uridine 33 in yeast tRNAPhe on the reaction with ribosomes.

We have determined several kinetic parameters for the reaction of poly(U)-programmed ribosomes with ternary complexes of elongation factor Tu, GTP, and yeast Phe-tRNA analogs with different bases substituted for uridine in position 33. These analogs test whether disruption of the hydrogen bonds normally formed by uridine 33 and steric crowding in the anticodon loop are detrimental to tRNA function on the ribosome. Single-turnover kinetic studies of the reaction of these ternary complexes with ribosomes show that these Phe-tRNA analogs decrease the apparent rate of GTP hydrolysis (kGTP) and the ratio of peptide formed to GTP hydrolyzed. Thus, the substitution of uridine 33 affects not only the selection of a ternary complex by the ribosome but also the selection of an aminoacyl-tRNA in the proofreading reaction. The effects become greater as first one, and then the other, H-bond is disrupted. Steric crowding in the anticodon loop is also important, but does not have as great an effect on the rate constants. An analysis of the elementary rate constants which comprise the rate constant, kGTP, demonstrates that the reduction in kGTP results from a decreased rate of ternary complex association with the ribosome (k1) and that there is little or no effect on the rate of GTP cleavage (k2). An analysis of the rate constants involved in proofreading shows that all the modified (tRNAs have increased rates of aminoacyl-tRNA rejection (k4) but that the rate of peptide bond formation (k3) is unaffected.

The reaction of ribosomes with elongation factor Tu.GTP complexes. Aminoacyl-tRNA-independent reactions in the elongation cycle determine the accuracy of protein synthesis.

The fidelity of protein synthesis depends on the rate constants for the reaction of ribosomes with ternary complexes of elongation factor Tu (EF-Tu), GTP, and aminoacyl (aa)-tRNA. By measuring the rate constants for the reaction of poly(U)-programmed ribosomes with a binary complex of elongation factor (EF-Tu) and GTP we have shown that two of the key rate constants in the former reaction are determined exclusively by ribosome-EF-Tu interactions and are not affected by the aa-tRNA. These are the rate constant for GTP hydrolysis, which plays an important role in the fidelity of ternary complex selection by the ribosome, and the rate constant for EF-Tu.GDP dissociation from the ribosome, which plays an equally important role in subsequent proofreading of the aa-tRNA. We conclude that the fidelities of ternary complex selection and proofreading are fundamentally dependent on ribosome-EF-Tu interactions. These interactions determine the absolute value of the rate constants for GTP hydrolysis and EF-Tu.GDP dissociation. The ribosome then uses these rate constants as internal standards to measure, respectively, the rate constants for ternary complex and aa-tRNA dissociation from the ribosome. These rates, in turn, are highly dependent on whether the ternary complex and aa-tRNA are cognate or near-cognate to the codon being translated.

Elongation factor Tu.guanosine 3'-diphosphate 5'-diphosphate complex increases the fidelity of proofreading in protein biosynthesis: mechanism for reducing translational errors introduced by amino acid starvation.

Complexes of elongation factor Tu (EF-Tu) with guanosine 3'-diphosphate 5'-diphosphate (ppGpp) bind to ribosomes where they slow the incorporation of aminoacyl-tRNAs into protein by inhibiting both the binding of aminoacyl-tRNA.EF-Tu.GTP ternary complexes and the formation of peptide bonds. The latter action increases the time available for aminoacyl-tRNA rejection by the ribosome and, therefore, increases the effectiveness of proofreading. Synthesis of ppGpp and the formation of EF-Tu.ppGpp occur in vivo in response to amino acid starvation. Our finding, therefore, suggests an explanation for the otherwise puzzling observation that amino acid starvation has, at most, a moderate effect on the fidelity of protein synthesis in wild-type Escherichia coli. We suggest that an EF-Tu.ppGpp-induced increase in the effectiveness of proofreading buffers the overall translational fidelity of these cells against amino acid starvation-induced errors in initial selection of aminoacyl-tRNA ternary complexes.

The rate of cleavage of GTP on the binding of Phe-tRNA.elongation factor Tu.GTP to poly(U)-programmed ribosomes of Escherichia coli.

Interaction of Phe-tRNA.elongation factor Tu.GTP with poly(U)-programmed ribosomes containing an occupied P site can be described by a three-step kinetic mechanism. Initial binding is followed by the cleavage of GTP, and then a new peptide bond is formed. Rate constants controlling the first and third of these reactions are known, but only a lower limit for the rate constant of the cleavage step has been reported. We have determined this rate constant to be 20 s-1 at 5 degrees C, 30 s-1 at 15 degrees C, and 50 s-1 at 25 degrees C. This is much faster than the reverse step of the initial binding process and implies that the intrinsic accuracy of the ribosome in the initial selection step is sacrificed in favor of speed. The similarity of the kinetic and chemical mechanism of this GTP cleavage step with other nucleoside 5'-triphosphatases is discussed.

Effect of ppGpp on the accuracy of protein biosynthesis.

The maintenance of accuracy in protein biosynthesis in amino acid-starved rel+ strains of Escherichia coli has been attributed to an effect of ppGpp on the accuracy of aa-tRNA selection by the ribosome. It has been determined that concentrations of ppGpp characteristic of those found in amino acid-starved cells have no effect on the rate of reaction of poly(U)-programmed ribosomes with either the cognate (Phe) or the near-cognate (Leu2) ternary complexes. Neither the rate of GTP hydrolysis, which signals selection of the ternary complex, nor the rate of peptide formation, which signals the acceptance of the aa-tRNA after proofreading, is affected by the nucleotide. The results indicate that the effect of ppGpp in maintaining the accuracy of protein biosynthesis in cells starved for an amino acid is not due to a direct effect on the rate constants for substrate selection by the ribosome.

Accuracy of protein biosynthesis. A kinetic study of the reaction of poly(U)-programmed ribosomes with a leucyl-tRNA2-elongation factor Tu-GTP complex.

We have determined several kinetic parameters for the reaction of poly(U)-programmed ribosomes with the near-cognate ternary complex of leucyl-tRNA2, elongation factor Tu (EF-Tu), and GTP. From single-turnover experiments at 5 degrees C with the ribosomes present in excess we find that the apparent second order rate constant for GTP hydrolysis is 0.4 X 10(6) M-1 s-1, that the cleavage step is faster than 4 s-1 and that the apparent rate constant for peptide formation is 6 +/- 3 s-1. From multiple-turnover experiments at 5 degrees C, with the ternary complex present in excess we find that kcat for GTP hydrolysis is 0.4 s-1 and Km is 1.6 microM. For both kinds of experiment the ratio of peptide formed to GTP hydrolyzed is 0.05 +/- 0.2. Comparison of these results with those obtained with the cognate complex show that the ribosome distinguishes between the cognate and near-cognate complexes, not on the basis of the forward rate constants, but on the basis of the reverse and rejection rate constants, which differ for the two complexes by at least 4000- and 100-fold, respectively. Both in ternary complex selection and in proofreading, the frequency of errors observed is much higher than might be expected from these large differences in rate constant. The reason is that, even for the near-cognate species, the rates of ternary complex rejection and aminoacyl-tRNA rejection are not overwhelmingly greater than the forward rate constants for GTP hydrolysis and peptidyl-tRNA formation, respectively. The outcome of these reactions is, therefore, at least partially under kinetic rather than thermodynamic control, leading to the trapping of errors which would not be made in a slow process.

Effect of Mg2+ concentration, polyamines, streptomycin, and mutations in ribosomal proteins on the accuracy of the two-step selection of aminoacyl-tRNAs in protein biosynthesis.

Discrimination against the binding of noncognate aminoacyl (aa)-tRNAs by mRNA-programmed ribosomes is the outcome of two selection steps, one involving an aa-tRNA.EFTu.GTP complex, which occurs prior to and includes GTP hydrolysis, the other involving the aa-tRNA alone, which follows GTP hydrolysis. Conditions which lead to errors in protein synthesis have been found to influence the accuracy of one or both selection steps in a system measuring poly(U)-directed binding of Leu-tRNA2Leu. Streptomycin has a large effect only on the discrimination process following GTP hydrolysis, but the other pertubations of recognition studied, high [Mg2+], polyamines, the strA1 and ram1 mutations, affect both discrimination processes. The general result is consistent with the view that proofreading of aa-tRNA by ribosomes for the most part uses the same specificity determinants used in the initial selection of a ternary complex.

A GTPase reaction accompanying the rejection of Leu-tRNA2 by UUU-programmed ribosomes. Proofreading of the codon-anticodon interaction by ribosomes.

The characteristics of a GTPase reaction between poly(U)-programmed ribosomes, EFTu . GTP, and the near-cognate aminoacyl (aa)-tRNA, Leu-tRNA Leu 2, have been studied to assess the role of this reaction in proofreading of the codon-anticodon interaction. The reaction resembles the GTPase reaction with cognate aa-tRNAs and EFTu . GTP in its substrate requirements, in its involving EFTu . GTP . aa-tRNA ternary complexes, and in its requiring a free ribosomal A-site. The noncognate reaction differs from the cognate one in that aa-tRNA becomes stably bound to the ribosomes only 5% of the time; it therefore seems best characterized as an abortive enzymatic binding reaction. The rate of reaction is a significant fraction (4%) of that of the cognate aa-tRNA, indicating that recognition of ternary complexes by ribosomes involves a level of error greater than that of translation as a whole. The rejection of the noncognate aa-tRNA following GTP hydrolysis is therefore a vital step in the translation process and fulfills the criteria set for a proofreading reaction. Leu-tRNA Leu 2 which escapes rejection through proofreading, forms a stable complex with the ribosomal A-site, so it appears that the Leu-tRNA2 which was rejected never reached the A-site and that proofreading precedes full A-site binding.

Single turnover kinetic studies of guanosine triphosphate hydrolysis and peptide formation in the elongation factor Tu-dependent binding of aminoacyl-tRNA to Escherichia coli ribosomes.

The rates of GTP hydrolysis and peptide formation during the reaction of Phe-tRNA . elongation factor Tu . GTP complex with acetyl-Phe-tRNA polyuridylate-programmed ribosomes have been measured. The GTPase reaction is second-order up to reactant concentrations of 0.2 microM and has a rate constant of 5 X 10(6) M-1 s-1 at 5 degrees C and 5 mM Mg2+, pH 7.2. The formation of peptide shows a lag phase and has a rate constant of 0.4 S-1 under these conditions. The results of a series of experiments between 5 degrees C and 25 degrees C show that GTP hydrolysis and peptide formation have Arrhenius activation energies of 13.1 and 15.3 kcal mol-1, respectively. The results indicate that these reactions proceed in vitro at rates comparable to those observed for protein biosynthesis in vivo, and that peptide bond formation occurs after GTP hydrolysis.