Development of a tryptophan-based dual selection system reveals the spatial organization of S-layer assembly during cytokinesis in Sulfolobus acidocaldarius

Article activity feed

1. 

   Review Commons

   Oct 12, 2026

   Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Reply to the reviewers

   Reviewer #1 (Evidence, reproducibility and clarity (Required)):

   Summary:

   In this manuscript, Sherman Foo and Buzz Baum report the development of a tryptophan auxotrophy-based selection system in S. acidocaldarius. Using classical pop-in and pop-out methodology, the authors isolated a strain ΔtrpBA using a rich medium BDN medium (Brock medium supplemented with NZ-amine, Sucrose and uracil and an uracil auxotroph. To get stringent selection, they developed a defined medium BDN, based on analysis of amino acid composition of NZ-amine A and other related substrates. They then analyzed the possibility to use 5-FAA for counterselection and found that 5-FAA …

   Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Reply to the reviewers

   Reviewer #1 (Evidence, reproducibility and clarity (Required)):

   Summary:

   In this manuscript, Sherman Foo and Buzz Baum report the development of a tryptophan auxotrophy-based selection system in S. acidocaldarius. Using classical pop-in and pop-out methodology, the authors isolated a strain ΔtrpBA using a rich medium BDN medium (Brock medium supplemented with NZ-amine, Sucrose and uracil and an uracil auxotroph. To get stringent selection, they developed a defined medium BDN, based on analysis of amino acid composition of NZ-amine A and other related substrates. They then analyzed the possibility to use 5-FAA for counterselection and found that 5-FAA did not appear to function as an effective counterselection agent. Next, they successfully constructed a tryptophan complementation system using the pSVAara plasmid backbone, replacing the pyrEF cassette with the trpBAD genes from Saccharolobus solfataricus (sso0888 -sso0890).

   Exploiting both uracil and tryptophan auxotrophies, the authors further demonstrated the use of this system by dual-plasmid co-transformation and co-expression of the surface layer proteins and a dominant-negative mutant of the AAA-ATPase Vps4 and confirmed that the accumulation of surface layer lattice forming protein SlaA at the midzone of division-arrested cells together with its membrane anchor SlaB. The localization of SlaA and SlaB at the midcell was in agreement with the observations previously reported by the same group.

   This is a good summary of our paper.

   Major comments:

   1. The dual plasmids differ only at the selection markers, pryEF and trpBA. The stability of them in one cell needs to be addressed.

   Thank you for raising this point. We have now addressed plasmid stability by following the maintenance of both plasmids during growth under selective and non-selective conditions. As shown in our new figure Fig. S4C, a strain carrying both plasmids was first grown in selective BCS medium and subsequently transferred to BCS, BCS ura+, BCS trp+, or BCS ura+ trp+. Cultures were sampled after 10 and 20 generations and plated on selective medium to quantify plasmid maintenance.

   As expected, the two plasmid were efficiently maintained under selection. When selection for either plasmid was relaxed by supplementation, plasmid loss increased progressively with the number of generations as expected. After 20 generations, approximately 5% of the initial CFU remained under BCS ura+, approximately 21% under BCS trp+, and approximately 14% under BCS ura+ trp+. Thus, both plasmids can be maintained together under selection, while their maintenance is reduced at a similar rate only when the corresponding selective pressure is removed. The presence or loss of one plasmid does not appear to affect the stability of the other.

   1. Relevant to the above, the authors should test the plasmid stability after culture with reasonable generations. Is it possible to construct a stably transmitted dural plasmid with different ori?

   We agree that an alternative plasmid architecture with a different origin of replication could potentially improve long-term plasmid maintenance. However, constructing and validating such a system would require the identification and testing of additional compatible origins of replication, including screening of available native or cryptic plasmids, followed by construction and characterization of new plasmid backbones in S. acidocaldarius. This would constitute a substantial plasmid-development study beyond the scope of the present work.

   Importantly, the dual-plasmid experiment was intended primarily as a proof-of-concept demonstrating that the two selection systems can be combined in the same cell, rather than as a requirement for all applications of the toolbox. Where long-term maintenance of both selectable markers is required, an alternative strategy would be to incorporate the second selection marker stably into the genome. This is addressed in the discussion. We have therefore focused the present study on demonstrating the functionality and compatibility of the two selection systems, while experimentally characterizing the stability of the dual-plasmid configuration under selective and relaxed selection conditions (Fig. S4C).

   1. Lines 214-223, I am curious why the authors used ΔtrpAB as the auxotroph instead of using ΔtrpF directly, because the latter could be easily subject to 5-FAA counter selection test.

   Thank you for this question. As described earlier in the Results (revised manuscript line 120), we initially selected trpBA as the deletion target because we wanted to establish tryptophan auxotrophy without disrupting the production of upstream metabolic intermediates generated by other genes in the tryptophan biosynthetic pathway, such as trpF, trpE and trpD. In particular, indole-3-glycerol phosphate and other intermediates of this pathway can serve as precursors for the biosynthesis of secondary metabolites, including benzoxazinones and indole diterpene alkaloids.

   More generally, the principle of the selection system is not restricted to *trpBA *or trpF. Other genes in an amino acid biosynthetic pathway could in principle be targeted, provided that the selected gene is non-redundant and its deletion produces the desired auxotrophic phenotype.

   Minor comments:

   1. Lines 71-84, please add an important reference, Zhao et al., Rational design of unrestricted pRN1 derivatives and their application in the construction of a dual plasmid vector system for Saccharolobus islandicus, 2024.

   We thank the reviewer for this excellent suggestion. We agree that this study is highly relevant to our discussion of archaeal genetic tool development. We have now included this reference in the indicated section, highlighting the recent development and of a second stably maintained vector system for* Sc. islandicus*.

   1. In the title, the spatial organization of S-layer assembly during cytokinesis has already been reported by the authors previously. The title could be improved by slight modification.

   We have now updated the title to avoid overstating this aspect as a completely novel finding and to shift the focus to the selection system instead.

   Reviewer #1 (Significance (Required)):

   Archaea of the order Sulfolobales are model microbes for studying the biology of archaea and the origin and evolution of eukaryotic-like characteristics. Lack of sufficient selection markers has been a limitation for genetic analysis of these organisms. Although dual selection system has been developed in Saccharolobus islandicus, another model archaeon of Sulfolobales, there is no such system for Sulfolobus acidocaldarius.

   The study by Foo and Baum explored a method for the formulation of defined Brock-based dropout medium, which enables stringent selection not only for trypan auxotrophic selection, but also will be potentially useful for other amino acid-based selection in general. Therefore, this study expanded the genetic toolkit available for S. acidocaldarius and provided additional evidence for spatial regulation of S-layer assembly during archaeal cytokinesis. The dual selection system was based the same plasmid backbone which could have problem in plasmid stability. Use of different compatible plasmid could improve the dual selection system. In general, the developed system will facilitate the study of cell biology and molecular biology using the model archaeon Sulfolobus acidocaldarius.

   We appreciate the reviewer's clear summary and are glad that this new selection method will be a valuable asset to the archaeal cell biology community.

   ---

   Reviewer #2 (Evidence, reproducibility and clarity (Required)):

   In this manuscript, the authors develop a new selection method in the model thermoacidophilic archaeon S. acidocaldarius that allows the simultaneous expression of two proteins of interest. Namely, they expressed either the S-layer proteins SlaA or SlaB fused to an HA tag simultaneously with a dominant negative Vps4 protein point mutant tagged with a His6 tag that allows cell cycle arrest. Using this technique, they go on to show that S-layer proteins accumulate at division site in dividing cells. The paper is well written, the data well presented, and the novel selection method will likely be a useful new tool in this model species for archaeal cell biology. However, a few points should be addressed to further support some of the claims made, convince readers of the usefulness of the new strain, and increase the overall clarity of the manuscript.

   Thank you for this summary of our results. Below, we address each of the specific points raised to further support our claims and enhance the manuscript's overall clarity and utility.

   Major points:

   • Line 115: Was any RT-qPCR or transcriptomics done to confirm that the expression of the downstream genes in the operon remained the same?
   • Line 227: Why express the trpBAD cassette if you only deleted trpBA? Does this mean you disrupted the expression of the downstream trpD gene (saci_1423) in the tryptophan biosynthesis operon?

   To address both points, yes, RT-qPCR was performed to test the expression of the downstream trpD gene. We have now included this result in Fig. S3D.

   We found that trpD expression was reduced but not abolished in the ΔtrpBA strain when compared to the wild-type strain DSM639. However, we also note that this is similar to the commonly used MW001 (ΔpyrE) strain, where the downstream pyrF gene expression was reduced compared to the wild-type strain. In both cases, complementation with the corresponding Sa. solfataricus gene(s) and the downstream gene (*trpD *or *pyrF) *was used to rescue its auxotrophic phenotype.

   • Line 227: Why did you choose to complement with the genes from Sa. solfataricus instead of using the native genes from S. acidocaldarius? Please clarify.

   This was done to minimize the risk of unwanted homologous recombination events between the expression vector and the host genome. Because the trpB 5’ UTR and trpD sequences are present on the plasmid backbone, using the native S. acidocaldarius sequence could lead to unintended recombination at these genomic flanking regions, destabilizing the system. We have now included this information in the text at line 230.

   • Line 236: Why is the promoter so much leakier on pSF251-SlaA-HA vs. pSVAaraFX-SlaA-HA? pSF251 is a derivative of pSVAaraFX-HA so it seems strange that the promoter would behave so differently on the two plasmids. While the arabinose promoter is somewhat leaky, it should not be this leaky according to reference 22 and in comparison to the non-induced controls of pSVAaraFX-SlaA-HA. The high level of expression in the non-induced control suggests the slaA-HA gene is being constitutively expressed. This could potentially be driven by the promoter of trpBAD cassette you introduced if you did not add a transcription terminator downstream.
   • Line 238: Why is the expression of SlaA-HA so much higher in BDM when using the pSF251-SlaA-HA plasmid vs. the pSVAara-Fx-SlaA-HA plasmid? It appears to express just as well as SlaA-HA did in BSN media when expressed on pSVAara-Fx-SlaA-HA. Do you think the promoter controlling the trpBAD casette is somehow driving stronger and more constitutive expression of genes downstream trpBAD?
   • Line 243: The authors may want to re qualify this. While the system can support inducible protein expression, the existing plasmid seems to express genes constitutively even when they are placed under inducible promoters. While the expression does increase upon induction, it goes from "high" to "very high" rather than "low" to "high" which may not be readily useful in typical applications of inducible promoters.

   Thank you for these comments. We have repeated and re-quantified the SlaA-HA expression experiments and believe that the apparent difference between pSF251-SlaA-HA and the parental pSVAaraFX-SlaA-HA construct is partly due to differences in Western blot presentation in the original manuscript.

   Firstly, pSF251 is progressively lost when cells are propagated in the absence of the corresponding selection in BNS. This is also supported by our observation in Fig. S4C, following supplementation of tryptophan in BCS. Thus, reduced SlaA-HA signal in both the uninduced and induced BNS samples in Fig. 3D is likely to reflect loss of the pSF251 plasmid rather than intrinsically low expression from the construct.

   Secondly, direct visual comparison of the original Fig. 3D with Fig. S2C is complicated by differences in Western blot exposure and contrast. In particular, the presentation of Fig. 3D was adjusted to make the relatively weak BNS signal visible, due to the point raised above. This more saturated presentation results in the impression that uninduced SlaA-HA expression with pSF251 in BDM may seem much higher than it actually is.

   We therefore consider the quantified values to provide a more appropriate comparison than direct visual comparison of band intensities between the two figures. To further support this, we have now included re-quantified data using triplicate experiments to show the quantified relative expression with means ±SD indicated in Fig. 3D and S2C. The re-quantified data show that SlaA-HA expression from pSF251 increases approximately two-fold following arabinose induction, confirming that the construct remains arabinose responsive. Furthermore, the blot in Figure 3D has been replaced with a more representative image featuring lower saturation levels.

   We also note that the magnitude of basal expression and the induction response can depend on the gene of interest and the properties of the encoded protein. The LacS protein used in previous characterization of the arabinose-inducible system is distinct from SlaA and may differ substantially in abundance and stability. A demonstration of this effect can be found in Fig. S4D, where the expression of SlaA and SlaB are significantly different, despite being expressed from the same plasmid. We have therefore avoided interpreting the basal expression observed with SlaA-HA as a universal property of the promoter system.

   With respect to the possibility of transcriptional read-through from the introduced S. solfataricus trpBAD cassette, we did not include a transcriptional terminator downstream of the cassette and therefore cannot exclude a contribution from transcription originating at the trpBAD promoter. However, we have not directly tested this mechanism and have therefore avoided attributing the observed basal expression specifically to trpBAD promoter read-through.

   Importantly, the primary purpose of pSF251 in this study is to introduce trpBAD as an alternative selectable marker, rather than to systematically characterize or optimize the arabinose-inducible promoter. Our data demonstrate that the resulting vector retains arabinose-responsive expression, while also highlighting that basal expression can be substantial and should be considered when selecting and optimizing constructs for individual applications. Alternative regulatory systems, including copper-, maltose- or xylose-responsive promoters, may provide additional options where tighter regulation is required, but systematic comparison of promoter systems is beyond the scope of this study. This caveat has now been included in the discussion.

   We have therefore edited the relevant text to state that pSF251 supports arabinose-responsive protein expression, rather than implying that it provides a tightly regulated low-to-high induction system.

   • Line 275: Since both plasmids have the same replication strategy as well as possessing extended regions of homology, one could wonder whether there are any issues with plasmids undergoing recombination with one another? Or if there are any issues with inheritance of the plasmids? While we certainly see cells that are positive for both proteins being expressed (an HA tagged S-layer protein and a His6 tagged Vsp4 protein), what is the percentage of cells that are positive for both? And how stable is this?

   This concern was also raised by Reviewer 1. We have now included new data in Fig. S4C to show the co-maintenance of both plasmids under selective and non-selective conditions. These experiments demonstrate that both plasmids can be maintained together under selection, while plasmid loss occurs progressively when the corresponding selective pressure is relaxed. Please see our response to Reviewer 1 above for a detailed description of these experiments.

   We have also examined simultaneous protein expression from the two plasmids at the single-cell level by flow cytometry. Specifically, we analysed strains overexpressing either SlaA-HA or SlaB-HA together with Vps4WkB-His6 (Fig. S4D). These experiments demonstrate that proteins encoded by the two plasmids can be expressed simultaneously in individual cells. However, the proportion of cells expressing each protein is dependent on the particular gene of interest. For example, SlaA expression is readily detected in the SlaA-HA/Vps4WkB-His6 strain, whereas SlaB expression is substantially lower in the corresponding SlaB-HA/Vps4WkB-His6 strain. Thus, the proportion of double-positive cells reflects gene-specific expression properties and cannot be considered a general characteristic of the dual-plasmid system.

   Importantly, the flow cytometry measurements report protein expression at the single-cell level and do not directly establish the presence or integrity of both plasmids in each individual cell. We therefore consider the plasmid stability experiment in Fig. S4C to be the appropriate assay for assessing inheritance of the two plasmids, while the flow cytometry and Western blotting experiments (Fig. S4D, Fig 4A) demonstrate that simultaneous expression from the two plasmids is possible in individual cells.

   • Line 379: Figure 3D actually shows the contrary (cells start to lose the pSF plasmid if grown in BSN). Please correct accordingly.

   Thank you for pointing this out. Indeed in the original Results (line 238), we also explicitly noted that lower expression levels were observed in BNS, potentially due to loss of the pSF251 vector in the presence of tryptophan in the rich medium, which relieves selection pressure. This limitation is further clarified with the addition of the plasmid stability data in Fig. S4C.

   Our intention in the statement at line 379 was not to suggest that pSF251 plasmids can be stably maintained during prolonged growth in BNS without selection. Rather, we intended to highlight the flexibility provided by having independent trpBAD and pyrEF selection systems. For example, the trpBAD marker can be used to select for genome-engineering events, while a pyrEF-based plasmid can subsequently be used for plasmid-based overexpression under conditions where higher protein yields are required. Conversely, the two selection systems can also be combined where simultaneous selection for a genomic modification and plasmid-borne construct is desirable.

   To avoid confusion, we have now edited this section to, “The availability of independent trpBAD and pyrEF selection systems provides flexibility for combining genome engineering and plasmid-based approaches. For example, auxotrophic strains can first be selected under defined conditions using trpBAD, while pyrEF-based plasmids can subsequently be used for overexpression experiments requiring higher protein yields. The two selection systems can also be combined to enable simultaneous selection of genomic modifications and plasmid-borne constructs.”

   • Line 407: How does SlaB contribute to the recruitment or stabilization of SlaA if SlaA is also enriched in the portion of the extended division bridge that SlaB is absent from? Please discuss.

   We agree that our original wording could be interpreted as implying that SlaB is required for recruitment or stabilization of SlaA at the division midzone, which is not directly supported by our observations.

   Our data show enrichment of SlaA or SlaB at the division midzone in division-arrested cells, consistent with the possibility that SlaB contributes to SlaA recruitment or stabilization, as suggested previously by the accumulation of exogenously added SlaA at the division bridge (Foo* et al.*, 2025). However, as the reviewer points out, SlaA remains enriched within portions of extended division bridges where SlaB signal is reduced. Thus, SlaB is unlikely to be the sole determinant of SlaA localization at the division midzone.

   One possibility is that SlaB is dynamically remodelled during division-bridge maturation. In cells with extended division bridges, SlaB signal appears reduced at the bridge midzone but enriched at the flanking regions, while SlaA remains enriched within the bridge. Alternatively, additional S-layer anchoring components, including Saci_1846 identified previously (Foo* et al.*, 2025), could contribute to SlaA retention or organization independently of SlaB. Distinguishing between these possibilities will require further experiments.

   We have therefore revised the text to present the contribution of SlaB to SlaA localization as a model rather than an established mechanism, and to acknowledge the possibility of additional or dynamically regulated S-layer anchoring mechanisms in the discussion.

   Foo, Sherman, Ido Caspy, Alice Cezanne, Tanmay A. M. Bharat, and Buzz Baum. “A Self-Assembling Surface Layer Flattens the Cytokinetic Furrow to Aid Cell Division in an Archaeon.” Proceedings of the National Academy of Sciences 122, no. 25 (2025): e2501044122. https://doi.org/10.1073/pnas.2501044122.

   -Figure 4: It is unclear what the sample size is for microscopy. If only N=10 cells were imaged in one single experiment for each strain, then this does not fit standard criteria for scientific reproducibility. Unless N=10 cells were imaged across multiple independent experiments (e.g. 3 or 4 cells in three independent experiments)? Either way, one would expect that more cells could be imaged and added to this analysis, especially given that the authors enriched for dividing cells in their assays. It is also important that cells be selected in a blinded fashion.

   Thank you for raising this point. We have repeated the microscopy analysis and increased the sample size to 25 cells per strain, with three independent experiments performed. The revised analysis and corresponding figure have been updated accordingly.

   Because the purpose of this analysis was specifically to examine S-layer organization at the division site, dividing cells were selected according to its morphology. Importantly, cells were first identified and selected using an the independent channel (ConA staining) without assessing SlaA/SlaB localization. Only after the cells had been selected were the SlaA/SlaB fluorescence channels examined and quantified. Thus, selection of cells for analysis was based on an independent marker of cell morphology rather than the S-layer localization phenotype being quantified. This approach also resulted in inclusion of cells with both short and extended division bridges. Consequently, the midzone enrichment of SlaB is less distinct in the population-level profile shown in Fig. 4E than for SlaA, reflecting heterogeneity in division-bridge morphology rather than selection based on the S-layer localization phenotype.

   We have clarified this procedure in the revised Materials and Methods.

   • Figure 4: "endogenous S-layer at the midzone": Here the authors image the localization of tagged, overexpressed copies of SlaA and SlaB, not endogenous proteins. Please correct.

   We have now edited this phrase accordingly.

   Minor points:

   • Line 145: While the cells clearly grow in the absence of added tryptophan, it appears that they have a more severe/noticeable growth defect when grown on solid vs. liquid media. Is there a reason why this might be so?

   The more pronounced growth defect observed on solid media compared to liquid media may be due to differences in pH between the two cultivation formats, dictated by technical constraints in solid media preparation. Gelrite, the gelling agent used, fails to polymerize at the acidic pH of 3.0 at high temperatures, and is thus kept at a pH of 5.0, which imposes an additional stress on *Sulfolobus *cells.

   • Line 153: have been > has been

   We have corrected this typo accordingly.

   • Line 216: Would 5-FAA not be expected to just abiotically isomerize into the product of trpF under the acidic media conditions?

   This is an interesting hypothesis. However, the literature and our experimental data demonstrate that this is unlikely to be occurring.

   Firstly, an older study has shown that anthranilic acid derivatives do not undergo spontaneous skeletal or positional isomerization in acidic environments. At highly acidic values (approximating pH 2), anthranilic acids are remarkably stable against complex structural modifications (Stevens et al., 1952). The introduction of a highly electronegative fluorine atom at the C-5 position is expected to further exert a powerful electron-withdrawing inductive effect on the benzene ring, protecting the aromatic ring against the electrophilic intermediates required for skeletal or positional rearrangements.

   Second and more importantly, if 5-FAA were abiotically converting into a functional downstream tryptophan intermediate, this shortcut would bypass the genetic block in our ΔtrpF strain, rescuing its growth. Instead, we observe that 5-FAA is uniformly toxic to both the wild-type and the ΔtrpF strain (Fig. S3C). This lethality rules out abiotic conversion into a tryptophan intermediate or thermal degradation of 5-FAA. Instead, it indicates that 5-FAA possesses a toxicity mechanism in S. acidocaldarius that operates independently of the TrpF pathway. This phenomenon is precedented in hyperthermophilic archaea; for example, in Pyrococcus furiosus, deletion of downstream tryptophan biosynthesis genes similarly failed to yield 5-FAA resistance (Farkas* et al.*, 2012).

   Stevens, W. H., J. M. Pepper, and M. Lounsbury. “The Decarboxylation of Anthranilic Acid.” Canadian Journal of Chemistry 30, no. 7 (1952): 529–40. https://doi.org/10.1139/v52-065.

   Farkas, Joel, Karen Stirrett, Gina L. Lipscomb, et al. “Recombinogenic Properties of Pyrococcus Furiosus Strain COM1 Enable Rapid Selection of Targeted Mutants.” Applied and Environmental Microbiology 78, no. 13 (2012): 4669–76. https://doi.org/10.1128/AEM.00936-12.

   • Line 228: Was a transcription terminator added downstream of sso0890?

   No. Please see above for the full response regarding transcriptional read-through and protein expression levels.

   • Line 375: This is only true for one of the two plasmids. The expression is quite high on your pSF plasmid.

   We agree that the original statement was too broad. We have rephrased the sentence to clarify that the response to arabinose induction is lower, rather than total expression levels.

   Reviewer #2 (Significance (Required)):

   S. acidocaldarius has previously been made accessible to genetic manipulation including in-frame markerless gene deletion and exogenous expression of proteins of interest from an expression vector, pSVA-AraFX, under the control of inducible promoters (arabinose, xylose, sucrose) that allow the timely expression of target proteins in the cell. Existing tools rely on uracil auxotrophy (i.e. the parental strain is deficient for uracil synthesis, and exogenous plasmid DNA carries the missing uracil synthesis gene, allowing selection of transformants). Here, the authors develop a tryptophan auxotrophy-based selection system in S. acidocaldarius by deleting the genes trpBA. They then modified the pSVA-AraFX plasmid by replacing the uracil synthesis genes pyrEF by the tryptophan synthesis genes trpBA, allowing double expression of proteins in S. acidocaldarius. The technique is used in the present work to show localization of S-layer proteins in dividing cells as a proof on concept. This novel selection method will likely be a useful new tool for archaeal cell biologists.

   We thank the reviewer for their positive assessment and for recognizing the value of our tryptophan auxotrophy-based selection system.

   ---

   Reviewer #3 (Evidence, reproducibility and clarity (Required)):

   The manuscript written by Foo and Baum describes the development of a tryptophan auxotrophy-based selection system for the archaeal model organism Sulfolobus acidocaldarius. The authors formulated a defined Brock-based dropout medium and demonstrated its use for tryptophan-based selection. They further combined this system with the established pyrEF marker to co-express two plasmid-borne constructs and used this approach to examine the localization of S-layer proteins in division-arrested cells.

   The development of an additional selectable marker in S. acidocaldarius has great potential for sophisticated genetic analyses, including chromosomal marker exchange, simultaneous gene expression and complementation, and so on. I believe that this work is a valuable contribution to the Sulfolobus research community. Overall, the manuscript is clearly written, the experiments are generally well designed, and the results are clearly presented and discussed. I have several comments that I hope will help the authors strengthen the manuscript.

   This is an accurate summary of our results.

   Major concerns 1.The dual-plasmid system is one of the major advances claimed in this study. However, the two plasmids appear to share the same backbone and differ mainly in their selection markers. The compatibility and stability of the two plasmids within the same cells have not been adequately validated. The authors should evaluate the dual-transformation efficiency and plasmid retention over multiple generations under dual-selection conditions. Plasmid-specific analyses are also needed to confirm that both plasmids remain intact and to exclude potential recombination or rearrangement.

   Thank you for the comment, this point was also raised by reviewers 1 and 2. We have now included new data in Fig. S4C to show the co-maintenance of both plasmids under selective and non-selective conditions. These experiments demonstrate that both plasmids can be maintained together under selection, while plasmid loss occurs progressively only when the corresponding selective pressure is relaxed. Please see our response to Reviewer 1 above for a more detailed description of these experiments.

   2.I am not fully convinced that the current data support the authors' claim of stringent tryptophan-based selection. The spot assays do demonstrate that an established ΔtrpBA culture cannot grow on BDM without tryptophan. However, in Sulfolobus transformation, a spot assay is not equivalent to a transformation plating assay and does not directly demonstrate the performance of the selection marker during transformation. Although the authors present this approach as a further development of the S. acidocaldarius genetic system, they do not show representative transformation plates or report transformation efficiency, background colony formation, or false-positive frequency. These data are needed to establish the stringency and practical utility of the proposed selection system.

   Thank you for raising this important point. We agree that the spot assays alone do not directly assess the performance of the* trpBAD* selection marker during transformation. We have therefore performed transformation plating assays and subsequently genotyped individual colonies to directly assess the stringency of tryptophan-based selection.

   Specifically, we performed three independent transformations of and genotyped 15 colonies from each transformation. This was done for the case of single (trpBAD) and double (*trpBAD *and pyrEF) transformations, plated on BDM or BCS. These analyses allowed us to directly determine the frequency of genuine transformants among colonies recovered. We have now included the resulting false-positive frequencies in the revised manuscript (lines 234, 281 and 294). These results demonstrate that background colony formation under the selection conditions is low and provide direct evidence for the stringency and practical utility of the trpBAD selection system during transformation.

   We have not presented CFU per µg DNA as a general transformation-efficiency metric. Although we recorded the number of colonies recovered following transformation, this value reflects not only DNA uptake but also recovery during the outgrowth period post electroporation and subsequent growth under selection. In addition, constructs containing different genes of interest can differ in plasmid size and in their effects on plasmid replication, copy number, gene expression, cellular fitness and recovery. Consequently, CFU per µg DNA would not provide a directly comparable measure of transformation efficiency across different constructs.

   Instead, we have focused on the frequency of genuine transformants among colonies recovered under selection, determined by genotyping independent colonies. We believe this provides a more direct assessment of the parameter relevant to the reviewer’s concern, namely the stringency and specificity of the proposed tryptophan-based selection system.

   Specific comments 1.The S-layer localization experiments provide a useful demonstration of the dual-selection system. However, the authors should clarify that SlaA and SlaB were examined separately in Vps4E209Q-arrested cells rather than simultaneously during unperturbed cytokinesis. Accordingly, terms such as "co-enrichment," "endogenous S-layer," and "spatial regulation" should be used with appropriate caution.

   We have combed through the manuscript text and updated the terms accordingly to more accurately reflect the conditions and results presented in the paper.

   2.The authors should provide additional information about key reagents, including the manufacturers and catalog numbers for casamino acids, tryptophan, Gelrite, and L-arabinose. This information is important for reproducing the defined and selective media, particularly because the composition of complex products such as casamino acids may vary among suppliers.

   This information is now included in the Materials and Methods.

   3.Plasmid sequences, annotated maps, primer sequences, and the complete BDM recipe should be deposited in a stable repository or provided as supplementary files. The statement that "all data are available in the manuscript" is not sufficient to ensure the reproducibility of a newly developed genetic system.

   We have now included the annotated plasmid map and sequence of pSF251, and primer sequences used as supplementary files. The full BDM recipe can be found in the Materials and Methods, which we have now updated to include the full recipe for Brock media.

   4.The sentence "The deletion of the arginine decarboxylase have been described" should be corrected to "Deletion of the arginine decarboxylase has been described."

   We have edited this typo accordingly.

   5.Figure S4C reports the percentages of HA-positive and His6-positive cells, but the gating strategy and negative controls used to establish the thresholds are not shown. Appropriate uninduced and single-expression controls should be included or described.

   We appreciate the reviewer's request for clarity on the gating approach. Because the data are presented as a scatter plot, the boundary thresholds for HA-positive and His6-positive populations were established as a horizontal gate on the Y-axis, set directly above the wild-type parental control population. As noted earlier in the manuscript, the uninduced sample exhibits a degree of leaky/basal expression rather than a completely silent state. Because of this, the uninduced population displays a significant shift in fluorescence compared to the negative wild-type control. This is highly dependent on the protein of interest as can be clearly observed between the expression of the soluble SlaA-HA vs the membrane protein HA-SlaB (Fig. S4D).

   As such, we chose to use the wild-type control to demonstrate the presence of the proteins of interest in this proof-of concept work. This Y-axis gate was positioned such that fewer than 5% of the negative wild-type control cells fall above the line. Because the fully induced populations exhibit an unambiguous upward shift in fluorescence intensity along the Y-axis, this clear boundary provides a highly reproducible separation between baseline cellular autofluorescence and protein overexpression.

   We have added a statement to the Materials and Methods section to explicitly clarify this gating threshold rationale in light of the basal expression levels.

   6.In line 227 and Figure 3A, why trpD was introduced into pSF251?

   This point was raised by reviewer 2 as well, please see above for our full response. In short, RT-qPCR analysis (Fig. S3D) revealed that our ΔtrpBA deletion reduced downstream trpD expression. This was similar to what we found in the MW001 background strain where, pyrF expression was reduced following deletion of pyrE. In both cases, complementation was carried out with the corresponding Sa. solfataricus gene + the downstream gene (trpD or pyrF)

   7.In line 268, I am not sure if "results not shown" should be allowed to use. The authors may need to rephrase these sentences or provide supporting data.

   We have revised this section to remove the statement regarding reduced protein expression, which was not directly supported by data presented in the manuscript. In addition, we now provide direct evidence that the ΔtrpBA mutant can be efficiently selected in BCS as well (line 294).

   Reviewer #3 (Significance (Required)):

   Sulfolobus acidocaldarius is an emerging model archaeon for studying fundamental cellular processes. However, the limited availability of selectable markers restricts more sophisticated genetic analyses in this organism. The development of an additional selectable marker should facilitate chromosomal gene exchange and recombination, simultaneous gene expression and complementation, and experiments combining genetic perturbations with reporter constructs. This work therefore represents a valuable technical contribution to the Sulfolobus research community and may enable more advanced genetic and cell-biological studies in this important archaeal model.

   We appreciate this supportive assessment of our work's significance.

   Read the original source
2. 

   Review Commons

   Oct 12, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   The manuscript written by Foo and Baum describes the development of a tryptophan auxotrophy-based selection system for the archaeal model organism Sulfolobus acidocaldarius. The authors formulated a defined Brock-based dropout medium and demonstrated its use for tryptophan-based selection. They further combined this system with the established pyrEF marker to co-express two plasmid-borne constructs and used this approach to examine the localization of S-layer proteins in division-arrested cells.

   The development of an additional selectable marker in S. acidocaldarius has great potential for sophisticated genetic analyses, including …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   The manuscript written by Foo and Baum describes the development of a tryptophan auxotrophy-based selection system for the archaeal model organism Sulfolobus acidocaldarius. The authors formulated a defined Brock-based dropout medium and demonstrated its use for tryptophan-based selection. They further combined this system with the established pyrEF marker to co-express two plasmid-borne constructs and used this approach to examine the localization of S-layer proteins in division-arrested cells.

   The development of an additional selectable marker in S. acidocaldarius has great potential for sophisticated genetic analyses, including chromosomal marker exchange, simultaneous gene expression and complementation, and so on. I believe that this work is a valuable contribution to the Sulfolobus research community. Overall, the manuscript is clearly written, the experiments are generally well designed, and the results are clearly presented and discussed. I have several comments that I hope will help the authors strengthen the manuscript.

   Major concerns

   1.The dual-plasmid system is one of the major advances claimed in this study. However, the two plasmids appear to share the same backbone and differ mainly in their selection markers. The compatibility and stability of the two plasmids within the same cells have not been adequately validated. The authors should evaluate the dual-transformation efficiency and plasmid retention over multiple generations under dual-selection conditions. Plasmid-specific analyses are also needed to confirm that both plasmids remain intact and to exclude potential recombination or rearrangement. 2.I am not fully convinced that the current data support the authors' claim of stringent tryptophan-based selection. The spot assays do demonstrate that an established ΔtrpBA culture cannot grow on BDM without tryptophan. However, in Sulfolobus transformation, a spot assay is not equivalent to a transformation plating assay and does not directly demonstrate the performance of the selection marker during transformation. Although the authors present this approach as a further development of the S. acidocaldarius genetic system, they do not show representative transformation plates or report transformation efficiency, background colony formation, or false-positive frequency. These data are needed to establish the stringency and practical utility of the proposed selection system.

   Specific comments

   1. The S-layer localization experiments provide a useful demonstration of the dual-selection system. However, the authors should clarify that SlaA and SlaB were examined separately in Vps4E209Q-arrested cells rather than simultaneously during unperturbed cytokinesis. Accordingly, terms such as "co-enrichment," "endogenous S-layer," and "spatial regulation" should be used with appropriate caution.
   2. The authors should provide additional information about key reagents, including the manufacturers and catalog numbers for casamino acids, tryptophan, Gelrite, and L-arabinose. This information is important for reproducing the defined and selective media, particularly because the composition of complex products such as casamino acids may vary among suppliers.
   3. Plasmid sequences, annotated maps, primer sequences, and the complete BDM recipe should be deposited in a stable repository or provided as supplementary files. The statement that "all data are available in the manuscript" is not sufficient to ensure the reproducibility of a newly developed genetic system.
   4. The sentence "The deletion of the arginine decarboxylase have been described" should be corrected to "Deletion of the arginine decarboxylase has been described."
   5. Figure S4C reports the percentages of HA-positive and His6-positive cells, but the gating strategy and negative controls used to establish the thresholds are not shown. Appropriate uninduced and single-expression controls should be included or described.
   6. In line 227 and Figure 3A, why trpD was introduced into pSF251?
   7. In line 268, I am not sure if "results not shown" should be allowed to use. The authors may need to rephrase these sentences or provide supporting data.

   Significance

   Sulfolobus acidocaldarius is an emerging model archaeon for studying fundamental cellular processes. However, the limited availability of selectable markers restricts more sophisticated genetic analyses in this organism. The development of an additional selectable marker should facilitate chromosomal gene exchange and recombination, simultaneous gene expression and complementation, and experiments combining genetic perturbations with reporter constructs. This work therefore represents a valuable technical contribution to the Sulfolobus research community and may enable more advanced genetic and cell-biological studies in this important archaeal model.

   Read the original source
3. 

   Review Commons

   Oct 12, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   In this manuscript, the authors develop a new selection method in the model thermoacidophilic archaeon S. acidocaldarius that allows the simultaneous expression of two proteins of interest. Namely, they expressed either the S-layer proteins SlaA or SlaB fused to an HA tag simultaneously with a dominant negative Vps4 protein point mutant tagged with a His6 tag that allows cell cycle arrest. Using this technique, they go on to show that S-layer proteins accumulate at division site in dividing cells. The paper is well written, the data well presented, and the novel selection method will likely be a useful new tool in this model species …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   In this manuscript, the authors develop a new selection method in the model thermoacidophilic archaeon S. acidocaldarius that allows the simultaneous expression of two proteins of interest. Namely, they expressed either the S-layer proteins SlaA or SlaB fused to an HA tag simultaneously with a dominant negative Vps4 protein point mutant tagged with a His6 tag that allows cell cycle arrest. Using this technique, they go on to show that S-layer proteins accumulate at division site in dividing cells. The paper is well written, the data well presented, and the novel selection method will likely be a useful new tool in this model species for archaeal cell biology. However, a few points should be addressed to further support some of the claims made, convince readers of the usefulness of the new strain, and increase the overall clarity of the manuscript.

   Major points:

   • Line 115: Was any RT-qPCR or transcriptomics done to confirm that the expression of the downstream genes in the operon remained the same?
   • Line 227: Why express the trpBAD cassette if you only deleted trpBA? Does this mean you disrupted the expression of the downstream trpD gene (saci_1423) in the tryptophan biosynthesis operon?
   • Line 227: Why did you choose to complement with the genes from Sa. solfataricus instead of using the native genes from S. acidocaldarius? Please clarify.
   • Line 236: Why is the promoter so much leakier on pSF251-SlaA-HA vs. pSVAaraFX-SlaA-HA? pSF251 is a derivative of pSVAaraFX-HA so it seems strange that the promoter would behave so differently on the two plasmids. While the arabinose promoter is somewhat leaky, it should not be this leaky according to reference 22 and in comparison to the non-induced controls of pSVAaraFX-SlaA-HA. The high level of expression in the non-induced control suggests the slaA-HA gene is being constitutively expressed. This could potentially be driven by the promoter of trpBAD cassette you introduced if you did not add a transcription terminator downstream.
   • Line 238: Why is the expression of SlaA-HA so much higher in BDM when using the pSF251-SlaA-HA plasmid vs. the pSVAara-Fx-SlaA-HA plasmid? It appears to express just as well as SlaA-HA did in BSN media when expressed on pSVAara-Fx-SlaA-HA. Do you think the promoter controlling the trpBAD casette is somehow driving stronger and more constitutive expression of genes downstream trpBAD?
   • Line 243: The authors may want to re qualify this. While the system can support inducible protein expression, the existing plasmid seems to express genes constitutively even when they are placed under inducible promoters. While the expression does increase upon induction, it goes from "high" to "very high" rather than "low" to "high" which may not be readily useful in typical applications of inducible promoters.
   • Line 275: Since both plasmids have the same replication strategy as well as possessing extended regions of homology, one could wonder whether there are any issues with plasmids undergoing recombination with one another? Or if there are any issues with inheritance of the plasmids? While we certainly see cells that are positive for both proteins being expressed (an HA tagged S-layer protein and a His6 tagged Vsp4 protein), what is the percentage of cells that are positive for both? And how stable is this?
   • Line 379: Figure 3D actually shows the contrary (cells start to lose the pSF plasmid if grown in BSN). Please correct accordingly.
   • Line 407: How does SlaB contribute to the recruitment or stabilization of SlaA if SlaA is also enriched in the portion of the extended division bridge that SlaB is absent from? Please discuss. -Figure 4: It is unclear what the sample size is for microscopy. If only N=10 cells were imaged in one single experiment for each strain, then this does not fit standard criteria for scientific reproducibility. Unless N=10 cells were imaged across multiple independent experiments (e.g. 3 or 4 cells in three independent experiments)? Either way, one would expect that more cells could be imaged and added to this analysis, especially given that the authors enriched for dividing cells in their assays. It is also important that cells be selected in a blinded fashion.
   • Figure 4: "endogenous S-layer at the midzone": Here the authors image the localization of tagged, overexpressed copies of SlaA and SlaB, not endogenous proteins. Please correct.

   Minor points:

   • Line 145: While the cells clearly grow in the absence of added tryptophan, it appears that they have a more severe/noticeable growth defect when grown on solid vs. liquid media. Is there a reason why this might be so?
   • Line 153: have been > has been
   • Line 216: Would 5-FAA not be expected to just abiotically isomerize into the product of trpF under the acidic media conditions?
   • Line 228: Was a transcription terminator added downstream of sso0890?
   • Line 375: This is only true for one of the two plasmids. The expression is quite high on your pSF plasmid.

   Significance

   S. acidocaldarius has previously been made accessible to genetic manipulation including in-frame markerless gene deletion and exogenous expression of proteins of interest from an expression vector, pSVA-AraFX, under the control of inducible promoters (arabinose, xylose, sucrose) that allow the timely expression of target proteins in the cell. Existing tools rely on uracil auxotrophy (i.e. the parental strain is deficient for uracil synthesis, and exogenous plasmid DNA carries the missing uracil synthesis gene, allowing selection of transformants). Here, the authors develop a tryptophane auxotrophy-based selection system in S. acidocaldarius by deleting the genes trpBA. They then modified the pSVA-AraFX plasmid by replacing the uracil synthesis genes pyrEF by the tryptophan synthesis genes trpBA, allowing double expression of proteins in S. acidocaldarius. The technique is used in the present work to show localization of S-layer proteins in dividing cells as a proof on concept. This novel selection method will likely be a useful new tool for archaeal cell biologists.

   Read the original source
4. 

   Review Commons

   Oct 12, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   Summary:

   In this manuscript, Sherman Foo and Buzz Baum report the development of a tryptophan auxotrophy-based selection system in S. acidocaldarius. Using classical pop-in and pop-out methodology, the authors isolated a strain ΔtrpBA using a rich medium BDN medium (Brock medium supplemented with NZ-amine, Sucrose and uracil and an uracil auxotroph. To get stringent selection, they developed a defined medium BDN, based on analysis of amino acid composition of NZ-amine A and other related substrates. They then analyzed the possibility to use 5-FAA for counterselection and found that 5-FAA did not appear to function as an effective …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   Summary:

   In this manuscript, Sherman Foo and Buzz Baum report the development of a tryptophan auxotrophy-based selection system in S. acidocaldarius. Using classical pop-in and pop-out methodology, the authors isolated a strain ΔtrpBA using a rich medium BDN medium (Brock medium supplemented with NZ-amine, Sucrose and uracil and an uracil auxotroph. To get stringent selection, they developed a defined medium BDN, based on analysis of amino acid composition of NZ-amine A and other related substrates. They then analyzed the possibility to use 5-FAA for counterselection and found that 5-FAA did not appear to function as an effective counterselection agent. Next, they successfully constructed a tryptophan complementation system using the pSVAara plasmid backbone, replacing the pyrEF cassette with the trpBAD genes from Saccharolobus solfataricus (sso0888 -sso0890).

   Exploiting both uracil and tryptophan auxotrophies, the authors further demonstrated the use of this system by dual-plasmid co-transformation and co-expression of the surface layer proteins and a dominant-negative mutant of the AAA-ATPase Vps4 and confirmed that the accumulation of surface layer lattice forming protein SlaA at the midzone of division-arrested cells together with its membrane anchor SlaB. The localization of SlaA and SlaB at the midcell was in agreement with the observations previously reported by the same group.

   Major comments:

   1. The dural plasmids differ only at the selection markers, pryEF and trpBA. The stability of them in one cell needs to be addressed.
   2. Relevant to the above, the authors should test the plasmid stability after culture with reasonable generations. Is it possible to construct a stably transmitted dural plasmid with different ori?
   3. Lines 214-223, I am curious why the authors useed ΔtrpAB as the auxotroph instead of using ΔtrpF directly, because the latter could be easily subject to 5-FAA counter selection test.

   Minor comments:

   1. Lines 71-84, please add an important reference, Zhao et al., Rational design of unrestricted pRN1 derivatives and their application in the construction of a dual plasmid vector system for Saccharolobus islandicus, 2024.
   2. In the title, the spatial organization of S-layer assembly during cytokinesis has already been reported by the authors previously. The title could be improved by slight modification

   Significance

   Archaea of the order Sulfolobales are model microbes for studying the biology of archaea and the origin and evolution of eukaryotic-like characteristics. Lack of sufficient selection markers has been a limitation for genetic analysis of these organisms. Although dual selection system has been developed in Saccharolobus islandicus, another model archaeon of Sulfolobales, there is no such system for Sulfolobus acidocaldarius.

   The study by Foo and Baum explored a method for the formulation of defined Brock-based dropout medium, which enables stringent selection not only for trypan auxotrophic selection, but also will be potentially useful for other amino acid-based selection in general. Therefore, this study expanded the genetic toolkit available for S. acidocaldarius and provided additional evidence for spatial regulation of S-layer assembly during archaeal cytokinesis. The dual selection system was based the same plasmid backbone which could have problem in plasmid stability. Use of different compatible plasmid could improve the dual selection system. In general, the developed system will facilitate the study of cell biology and molecular biology using the the model archaeon Sulfolobus acidocaldarius.

   Read the original source
5. 

   Version published to 10.64898/2026.06.29.735232 on bioRxiv

   Jun 29, 2026