The Length of the Combined 3′ Untranslated Region and Poly(A) Tail Does Not Control Rates of Glyceraldehyde-3-Phosphate Dehydrogenase mRNA Translation in Three Species of Parasitic Protists

Abstract

Experimental observations suggested that the length of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA 3′ end has a role in regulating rates of translation in the parasitic protists Trypanosoma brucei, Leishmania donovani, and Trichomonas vaginalis. Using a PCR assay for poly(A) tail length, we measured the size of the RNA 3′ end under different growth conditions in all three species. Our results showed that the combined 3′ untranslated region and poly(A) tail of GAPDH mRNA do not vary with different rates of translation.

Activities of the enzymes of the glycolytic pathway of the parasitic protists Trypanosoma brucei, Leishmania donovani, and Trichomonas vaginalis are finely adjusted to the needs of the cell (21–23). Cellular activities vary by a factor of 2 to 40 depending on growth conditions but rarely have a direct relationship to growth rate. Levels of RNA, both rRNA and different species of mRNA, are far more directly related to growth rate and can vary by an order of magnitude. The lack of correlation between steady-state mRNA levels and activities of the corresponding enzymes suggests that control of their expression occurs posttranscriptionally (23, 24). Since most of the enzymes studied are not regulated by low-molecular-weight effectors and no other posttranslational modification has been identified (3, 12, 13), activity can be used as a measure for abundance. Therefore, expression of the genes coding for the glycolytic enzymes must be regulated at the translational level.

The importance of translational control in the overall regulation of gene expression is increasingly recognized (10, 11, 14). A role of the mRNA 3′ end in regulation of translation has been suggested for many species (6), including expression of the glucose transporter (8) and procyclin genes in T. brucei (5, 7, 9, 18). In addition, the 3′ untranslated region (UTR) influences rates of turnover of hsp83 message of L. donovani (1). During oogenesis and early development in metazoans, a predominant form of translational regulation is a cytoplasmic change in the length of the poly(A) tail (27). These changes in tail length are controlled by sequences in the 3′ UTR. The modifications at the 3′ end of mRNAs seem to be under the strict control of the cell (28). Northern analysis of RNA isolated from cells grown in chemostats (23, 24) revealed different migration patterns of the same message isolated under different growth conditions. We therefore tested whether the length of the poly(A) tail and/or 3′ UTR of the message encoding the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) had a role in regulating its translation in these protists.

The length of the combined 3′ UTR and poly(A) tail was measured on the GAPDH mRNA found in T. brucei, L. donovani, and T. vaginalis grown at different growth rates and under different carbon regimens (23, 24). The organisms were grown in single-stage flow-controlled chemostats (25, 26) under glucose limitation or at excess glucose concentrations, in which case a component of the serum was rate limiting. Steady states were obtained at five growth rates, approximately 0.2, 0.4, 0.6, 0.8, and 0.95 times the maximum for each species, for both limited-glucose and excess-glucose cells, giving a total of 10 steady states per species. Message levels and GAPDH activities for these cultures have been reported recently (23, 24). The length of the combined 3′ UTR and poly(A) tail was measured using the ligase-mediated PCR poly(A) test (PAT) (16, 17). This PCR-driven assay amplifies a region between a specific site near the 3′ end of an RNA and an oligo(dT) primer that is targeted to the end of the poly(A) tail. The resulting amplicon represents the 3′ UTR and the poly(A) tail. Since the oligo(dT) anchor is targeted to the extreme end of the poly(A) tail, any change in length will manifest itself as a change in mobility of the resulting amplicon. Briefly, RNA samples from each steady state were incubated with oligo-p(dT)_12–18 for 30 min at 42°C in the presence of T4 DNA ligase. A fivefold molar excess of oligo(dT) anchor (5′-GCGAGCTCCGCGGGCCGCC-T12) was added, followed by a 2-h incubation at 12°C. This allows targeting of the anchor sequence to the region representing the extreme 3′ end of the poly(A) tail. First-strand cDNAs were then synthesized by reverse transcription and used as template for PCR. The samples were amplified using the oligo(dT) anchor and a specific primer for GAPDH (T. brucei, GTTATTCCCACCGTGTGGTG; L. donovani, GACCTGGTGCGCTACATGGC; and T. vaginalis, CAAACCCAGAAGGCCAAGGC), yielding a heterogeneous pool of amplicons that represent the length of the mRNA from the specific GAPDH primer site through the end of the poly(A) tail. The size of the amplicons from each sample was determined by agarose gel electrophoresis and ethidium bromide staining.

The combined length of the poly(A) tail and 3′ UTR was calculated by substracting the length of the region between the gene-specific primer site and the end of the open reading frame from the total measured length. This parameter varied within each sample by between 380 and 550 nucleotides (nt) in T. brucei, between 450 and 540 nt in L. donovani, and between 130 and 170 nt in T. vaginalis (Fig. 1). There was no meaningful difference among the size distributions of the 10 samples from each species. For a regulatory role, one would expect size differences of 50 to 600 nt (2, 4, 15, 27). No such variation was observed, even though the ligase-mediated PAT can detect differences of 10 nt (16). These data suggest that there is no direct influence of the growth rate on the size distribution of the 3′ UTR and poly(A) tail for the GAPDH message under conditions that greatly alter GAPDH activity. This finding does not contradict reduced rates of translation for individual mRNA copies caused by the progressive shortening of the poly(A) tail during the translation process (2, 4, 15). Messages with shorter 3′ UTRs and/or poly(A) tails may be translated at lower rates than those with longer ones. Nevertheless, the data indicate that the length of the 3′ end is not a mechanism for regulating rates of translation in the case studied here.

FIG. 1.

(A) PAT analysis of GAPDH mRNA. Agarose gels of the amplicons indicating the length of the combined poly(A) tail and 3′ UTR of the GAPDH messages of the three species grown under 10 different conditions. Growth rates vary by a factor of 5, and cellular activities vary up to 10-fold. As rates of turnover are below 2% per h, rates of translation vary by a factor of 50 maximally. No meaningful differences in size distributions between different samples of each species were detected. The lower-molecular-weight band in the L. donovani samples is an artifact due to internal hybridization, but as it also shows no modifications, the conclusions would not change if it were the primary signal. PAT cDNAs were prepared as described elsewhere using avian myeloblastosis virus reverse transcriptase (16). cDNAs were amplified under the following conditions: initial denaturation for 5 min at 93°C, followed by 35 cycles of 45 s at 93°C, 45 s at 60°C, and 75 s at 72°C. After cycling, the samples were incubated at 72°C for 7 min as a final extension. Molecular weight analysis was performed by electrophoresis on a 2% Metaphor (FMC) gel. For each species, lanes 1 to 5 are from cells grown under glucose limitation at approximately 0.2, 0.4, 0.6, 0.8, and 0.95 times the maximum growth rate. Lanes 6 to 10 are from excess glucose-grown cells at similar growth rates. The molecular weight estimations are derived from a PBR-MSP1 digest electrophoresis in an adjacent well. The reported 3′ UTR length takes the length of the coding region that was included in the amplicon into account. Numbers at right are nucleotides. (B) Agarose gel of control amplifications demonstrating reverse transcriptase dependence. T.b, T. brucei; L.d, L. donovani; T.v, T. vaginalis. Reverse transcriptase (rt) M-pBR322 digest was present (+) or absent (−). Numbers at left are as defined for panel A.

Regulation of translation by the length of the 3′ UTR has been demonstrated for mammalian and plant cells (19, 20) in addition to the examples from T. brucei and L. donovani mentioned above. The results of the present study do not exclude an effect of the length of the 3′ UTR and/or poly(A) tail on rates of translation of specific mRNAs. In fact, such an effect is highly likely, as translation initiation seems to involve interaction of translation initiation factors with both the 5′ and 3′ ends of messages (11, 14). However, the results of this study contradict the hypothesis that the length of the 3′ UTR and/or poly(A) tail has a role in controlling rates of translation for the specific cases of GAPDH message in the three protists studied. The reason for this is that, while rates of translation of the individual message copies may vary (23, 24), the size distribution of the combined 3′ UTR and poly(A) tail did not vary under those conditions that produced different translation rates.