Why can't certain nucleoside analogs be used to label bacterial RNA?

Question

A 2020 publication in Nucleic Acids Research ¹ includes the following passage:

A variety of nucleoside analogs have been developed for metabolic RNA labeling in various eukaryotic cells (9–16). Among them, 4-thiouridine (4SU) and 5-ethynyluridine (EU) are two most widely used noncanonical nucleosides that can be conjugated via thiol coupling chemistry and click chemistry, respectively (9,10). Although the metabolic RNA labeling technique has been instrumental for studying RNA dynamics in eukaryotes, it is not applicable to bacteria.

The authors give no citations or further explanations for the bolded claim. Indeed, I have found at least one publication that contradicts this statement,² showing 4SU labeling in E. coli:

We have defined conditions allowing the substitution of 13 +/- 2% of the uridine residues in bulk RNA by 4-thiouridine.

One explanation is that 4SU is a natural RNA modification in bacteria, as Rogers et al. ³ states that most bacterial tRNAs contain 4SU. So, while bacterial RNA can be labeled with exogenous 4SU, naturally occurring 4SU confounds analysis. Concerning EU, Blenkiron et al. ⁴ used this analog to label nascent RNAs from uropathogenic E. coli in order to track the fate of bacterial RNAs in membrane vesicles:

We then looked at whether we could track uptake of MV RNA cargo into treated host cells by labelling UPEC nascent RNA in the bacteria before it was packaged with MVs. This was achieved by adding an excess of a modified uracil, 5EU (5-ethynyl uridine), which was incorporated into the newly synthesised and MV associated RNA.

Additionally, the text of this research grant award abstract ⁵ suggests that EU is suitable for labeling both E. coli and B. subtilis:

We have been investigating whether bacteria can incorporate the “clickable” RNA analog 5-ethynyl uridine (EU) for subsequent attachment of fluorescent label (Alexa Fluor) for microscopic detection or a chemical tag (desthiobiotin) for streptavidin-mediated RNA capture. The results for both aspects so far are a qualified “yes”. Both Escherichia coli and Bacillus subtilis grown with EU can subsequently be detected by fluorescence microscopy, but E. coli becomes considerably more brightly labeled.

I can find no research stating EU / 5EU naturally occurs in bacterial RNA.

My question, in short: What aspects of bacterial metabolism or RNA composition justify the claim that 4SU and EU are "not applicable" for bacterial RNA labeling?

---

Edit, 1 March 2021 -- I'm adding a bounty to garner more attention. From further reading, I suspect the answer to be one of 3 possibilities.

1. Per David's comment, the bacterial cell wall may be impermeable to certain formulations of nucleoside analogs or the reagents subsequently used in thiol coupling / click chemistry.
2. Certain nucleoside analogs are toxic in bacteria to an extent not seen in eukaryotic cells, perhaps by a mechanism that causes the elongating RNA polymerase to stall.
3. 4SU and EU labeling are applicable to bacteria, and the claims of the authors are unsubstantiated.

---

References

1. Meng L, Guo Y, Tang Q, Huang R, Xie Y, Chen X. Metabolic RNA labeling for probing RNA dynamics in bacteria. Nucleic Acids Res. 2020 Dec 16;48(22):12566-12576.
2. Favre A, Bezerra R, Hajnsdorf E, Lemaigre Dubreuil Y, Expert-Bezançon A. Substitution of uridine in vivo by the intrinsic photoactivable probe 4-thiouridine in Escherichia coli RNA. Its use for E. coli ribosome structural analysis. Eur J Biochem. 1986 Nov 3;160(3):441-9.
3. Rogers KC, Crescenzo AT, Söll D. Aminoacylation of transfer RNAs with 2-thiouridine derivatives in the wobble position of the anticodon. Biochimie. 1995;77(1-2):66-74.
4. Blenkiron C, Simonov D, Muthukaruppan A, Tsai P, Dauros P, Green S, Hong J, Print CG, Swift S, Phillips AR. Uropathogenic Escherichia coli Releases Extracellular Vesicles That Are Associated with RNA. PLoS One. 2016 Aug 8;11(8):e0160440.
5. Click-chemistry labeling of RNA: A new tool for microbial ecology?

Answer 1

Meng et al. actually give their explanation in the first paragraph of their results section. The main reasons for their claim are that they did not detect any labeling with 4SU or EU and that both analogs suppress bacterial growth. Also they raise the issue with the naturally occurring 4SU in tRNAs.

Aiming to develop nucleoside analogs compatible with bacterial RNA labeling, we first set out to evaluate the two nucleoside analogs widely used in eukaryotic cells. E. coli cells were incubated with 4SU or EU at varied concentrations for 2 h, which resulted in no detectable labeling even at high concentrations (Supplementary Figure S1A, B). Moreover, severe suppression of bacterial growth was observed with both 4SU and EU (Supplementary Figure S1C, D), which was in agreement with previous observations that 4SU impaired cellular metabolism in E. coli (31). Cell viability was significantly impaired with 4SU at concentrations higher than 100 μM, while EU did not induce significant decrease in cell viability (Supplementary Figure S1E, F). In addition, tRNAs of bacteria contain endogenous 4SU generated via thiolation of uracil (32), which may interfere with metabolic labeling if exogenous 4SU would be used. Of note, although EU does not incorporate into DNA in mammalian cells (10), significant labeling of DNA was observed in EU-treated E. coli, presumably due to metabolic conversion of EU to EdUTP (Supplementary Figure S1G). Together, these results demonstrate that 4SU and EU are not applicable for bacteria.

This explanation is not necessarily incompatible with the results in the other references mostly because the strains and the test approaches are different.

In Favre et al. they use a different strain of E. coli,"an Escherichia coli pyrD strain" that is optimized for 4SU incorporation :

In Material and Methods :

pyrD used to control the level of 4-thiouridine incorporation

and in Results :

a) growth of E. coli AB1157 sfiA pyrD in the presence of 4-thiouridine

Preliminary studies showed that the level of s4U incorporation into the rRNA of strains MRE 600 or K12 AB1157 grown in a standard glucose medium in the presence of 4-thiouridine remained at the detection limit. We therefore use AB1157 pyrD mutant deficient in the pyrimidine biosynthesis pathway. To our growth medium we systematically add 40 μg/ml uridine because it is the minimal concentration required to supplement growth fully. The growth of the pyrD mutant was first examined at different temperatures and different exogeneous s4U concentrations from 40 μg/ml upwards in the conditions defined in Material and Methods, In Fig. 1 we report the growth curves obtained for two s4U/U ratios,1 and 2.5, and four temperatures: 25°C, 28°C, 30°C, 35°C. When the ratio is 1 there is practically no growth inhibition. With a ratio of 2.5 the growth inhibition becomes strongly dependent on temperature. The culture does not grow at all at 20°C, the growth is linear between 25°C and 30°C and for temperatures above 30°C it escapes inhibition, becoming exponential. As will be described below significant s4U incorporation occurs in conditions where growth is linear.

In Favre et al. as seen at the end of the paragraph above as well as in another publication (Bezerra and Favre, 1990) from the same lab shows that incorporation of 4SU depends on several factors.

Concerning the work of Blenkiron et al., it is harder to explain but they also use a different strain so that may impact the incorporation and they do not use the same approaches compared to Meng et al. to detect it. Confocal microscopy for Blenkiron et al. vs Dot blot and In-gel fluorescence for Meng et al..

Finally as far as the grant abstract is concerned, it's hard to conclude because the information is scarce but as Meng et al. they also observed an effect (although milder) of 5EU on bacterial growth and again, they used a different approach to observe the labeling based on fluorescence microscopy as Blenkiron et al..

---

References :

• Meng L, Guo Y, Tang Q, Huang R, Xie Y, Chen X. Metabolic RNA labeling for probing RNA dynamics in bacteria. Nucleic Acids Res. 2020 Dec 16;48(22):12566-12576.
• Favre A, Bezerra R, Hajnsdorf E, Lemaigre Dubreuil Y, Expert-Bezançon A. Substitution of uridine in vivo by the intrinsic photoactivable probe 4-thiouridine in Escherichia coli RNA. Its use for E. coli ribosome structural analysis. Eur J Biochem. 1986 Nov 3;160(3):441-9.
• Blenkiron C, Simonov D, Muthukaruppan A, Tsai P, Dauros P, Green S, Hong J, Print CG, Swift S, Phillips AR. Uropathogenic Escherichia coli Releases Extracellular Vesicles That Are Associated with RNA. PLoS One. 2016 Aug 8;11(8):e0160440.
• Bezerra R, Favre A. In vivo incorporation of the intrinsic photolabel 4-thiouridine into Escherichia coli RNAs. Biochemical and biophysical research communications. 1990 Jan 15;166(1):29-37.