DNA topoisomerase IIIβ promotes cyst generation by inducing cyst wall protein gene expression in Giardia lamblia

Giardia lamblia causes waterborne diarrhoea by transmission of infective cysts. Three cyst wall proteins are highly expressed in a concerted manner during encystation of trophozoites into cysts. However, their gene regulatory mechanism is still largely unknown. DNA topoisomerases control topological homeostasis of genomic DNA during replication, transcription and chromosome segregation. They are involved in a variety of cellular processes including cell cycle, cell proliferation and differentiation, so they may be valuable drug targets. Giardia lamblia possesses a type IA DNA topoisomerase (TOP3β) with similarity to the mammalian topoisomerase IIIβ. We found that TOP3β was upregulated during encystation and it possessed DNA-binding and cleavage activity. TOP3β can bind to the cwp promoters in vivo using norfloxacin-mediated topoisomerase immunoprecipitation assays. We also found TOP3β can interact with MYB2, a transcription factor involved in the coordinate expression of cwp1-3 genes during encystation. Interestingly, overexpression of TOP3β increased expression of cwp1-3 and myb2 genes and cyst formation. Microarray analysis confirmed upregulation of cwp1-3 and myb2 genes by TOP3β. Mutation of the catalytically important Tyr residue, deletion of C-terminal zinc ribbon domain or further deletion of partial catalytic core domain reduced the levels of cleavage activity, cwp1-3 and myb2 gene expression, and cyst formation. Interestingly, some of these mutant proteins were mis-localized to cytoplasm. Using a CRISPR/Cas9 system for targeted disruption of top3β gene, we found a significant decrease in cwp1-3 and myb2 gene expression and cyst number. Our results suggest that TOP3β may be functionally conserved, and involved in inducing Giardia cyst formation.

Major Points 1. The change in localization observed for TOP3β in non-encysting cells (Fig 1) versus encysting cells in Sup Fig 3 is fascinating and this deserves more attention. That TOP3β localization changes from perinuclear to dispersed in the nucleus and is an important point that supports the authors assertion that TOP3β is involved in regulating differentiation. This same re-localization is not observed in Fig 1 perhaps because the cell is not at the right stage of encystation or is not actually encysting (no CWP1 staining here). The cell shown in Fig S3 is sort of mid-encystation based on the CWP1 staining. So I also wonder if the nuclear distribution changes with stage in the encystation response. This is potentially a really nice observation that if moved to Fig 1 will strengthen the paper because it shows TOP3β is specifically changing localization in actively differentiating cells. I would like the authors to include this result in Fig 1 and also quantify the number of cells that have dispersed nuclear versus perinuclear localization during encystation. It is also interesting that the mutants are more clearly nuclear localized during encystation. Since there is no quantification of this I don't know if this is due to selected images or a real response to encystation stimulus. Fig S5B on the other hand is very clear. I believe the authors, but this experiment should be repeated and hopefully a clearer result can be obtained. What also doesn't make sense is that in S5B there are two bands one that is found in every lane and then a unique shifted band near the top of the blot in the presence of the antibody. In Fig 5C the same band is observed in every well although it does appear darker in the lane where the antibody was labeled. As an alternative to repeating the blot, the authors could quantify the intensity of this band versus the same position in the control lanes to convince us that there is enrichment in their replicates.

The super-shift assay in Fig 5C is not very convincing. A similar experiment in
3. Regarding the CRISPR experiment I question the value of using CRISPR/CAS9 over homologous recombination. I am under the impression that CRISPR/CAS9 is not actually providing any benefit and that the authors are observing homologous recombination. After transfection we expect the provided repair template to enter one of the two nuclei based on Poxleitner et al 2008 and Carpenter et al 2012. Therefore, a single transfection should be able to edit two copies of the genome. It is not expected to generate a true knockout due to the two nuclei challenge. Here however, PCR (8B) and western blotting (8E) both point toward just a single allele being disrupted since genomic DNA and protein levels are pretty close to the 25% reduction level of losing a single allele. The authors mention the use of NHEJ inhibiting drugs which seem to be doing nothing since Ebneter et al showed that homologous recombination can be used to knockout genes in Giardia. I wonder why the authors bother to invoke the use of CRISPR since there is no perceivable benefit? Note that I do not dispute their mutant has a phenotype. My concern is that their pseudo use of CRISPR/CAS9 will confuse readers.
Minor Points 1. Fig S1 is corrupted or there was an error in its construction. The region around T328 is not shown and neither is the red arrow that is mentioned by the legend. Please fix this.
2. Sentence on Line 157 is confusing. It reads as if it is interesting that ESVs contain CWP1. To me what is interesting is that the localization of TOP3β is more dispersed in this encysting cell.
3. Sentence on 342 is awkward please re-write 4. On 411 change catalytic to catalytically 5. I would change the last sentence of the conclusion. I don't believe this study told us much about the evolution of TOP3β and whether TOP3β is really a viable drug target. The function of the gene wasn't shown to be essential since only a single allele of TOP3β was deleted according to the PCR in Fig 8B which roughly corresponds to the 30% reduction in TOPO3β protein levels. My recommendation: "Our study provides evidence for the important role of TOP3β in the authors assertion that TOP3β is involved in regulating differentiation. This same re-localization is not observed in Fig 1 perhaps because the cell is not at the right stage of encystation or is not actually encysting (no CWP1 staining here). The cell shown in Fig S3 is sort of mid-encystation based on the CWP1 staining. So I also wonder if the nuclear distribution changes with stage in the encystation response. This is potentially a really nice observation that if moved to Fig 1 will strengthen the paper because it shows TOP3β is specifically changing localization in actively differentiating cells. I would like the authors to include this result in Fig 1 and also quantify the number of cells that have dispersed nuclear versus perinuclear localization during encystation. It is also interesting that the mutants are more clearly nuclear localized during encystation. Since there is no quantification of this I don't know if this is due to selected images or a real response to encystation stimulus. Fig 5C is not very convincing. A similar experiment in Fig S5B on the other hand is very clear. I believe the authors, but this experiment should be repeated and hopefully a clearer result can be obtained. What also doesn't make sense is that in S5B there are two bands one that is found in every lane and then a unique shifted band near the top of the blot in the presence of the antibody. In Fig 5C the same band is observed in every well although it does appear darker in the lane where the antibody was labeled. As an alternative to repeating the blot, the authors could quantify the intensity of this band versus the same position in the control lanes to convince us that there is enrichment in their replicates. 3. Regarding the CRISPR experiment I question the value of using CRISPR/CAS9 over homologous recombination. I am under the impression that CRISPR/CAS9 is not actually providing any benefit and that the authors are observing homologous recombination. After transfection we expect the provided repair template to enter one of the two nuclei based on Poxleitner et al 2008 and Carpenter et al 2012. Therefore, a single transfection should be able to edit two copies of the genome. It is not expected to generate a true knockout due to the two nuclei challenge. Here however, PCR (8B) and western blotting (8E) both point toward just a single allele being disrupted since genomic DNA and protein levels are pretty close to the 25% reduction level of losing a single allele. The authors mention the use of NHEJ inhibiting drugs which seem to be doing nothing since Ebneter et al showed that homologous recombination can be used to knockout genes in Giardia. I wonder why the authors bother to invoke the use of CRISPR since there is no perceivable benefit? Note that I do not dispute their mutant has a phenotype. My concern is that their pseudo use of CRISPR/CAS9 will confuse readers.

The super-shift assay in
Minor Points 1. Fig S1 is corrupted or there was an error in its construction. The region around T328 is not shown and neither is the red arrow that is mentioned by the legend. Please fix this.
2. Sentence on Line 157 is confusing. It reads as if it is interesting that ESVs contain CWP1. To me what is interesting is that the localization of TOP3β is more dispersed in this encysting cell.
3. Sentence on 342 is awkward please re-write 4. On 411 change catalytic to catalytically 5. I would change the last sentence of the conclusion. I don't believe this study told us much about the evolution of TOP3β and whether TOP3β is really a viable drug target. The function of the gene wasn't shown to be essential since only a single allele of TOP3β was deleted according to the PCR in Fig 8B which roughly corresponds to the 30% reduction in TOPO3β protein levels. My recommendation: "Our study provides evidence for the important role of TOP3β in the differentiation of Giardia trophozoites into cysts, leading to greater understanding of the mechanism regulating cell differentiation and parasite transmission." Referee: 2 Comments to the Author(s) The authors have submitted a detailed and interesting study of the effect of DNA topoisomerase 3b (type 1) on encystation in Giardia. The title focuses on the effect of cyst generation related to cwp expression. However, I'm not convinced that there is enough causal information to warrant the use of "by inducing" in the title.
They have used transfection studies to study the effect of upregulation of TOP3b on expression of the cwps and myb2 on encystation. They have also used CRISPR/Cas9 to reduce expression of top3b and shown decreases in the above. (Of note, knockout is impossible with Giardia because of its polyploidy, so these approaches are reasonable).
Overall, the scientific component of the work seems solid, but I do have some questions regarding the data to be addressed by the authors: Thus, I will begin with the results section: For figure 1 and other figures, the authors simply use p<0.05, but unless <0.001, it would be helpful to see the numbers for CI and actual p value. For Fig 4A, the difference between top3b positive and negative for linear DNA is not very convincing. Can the authors provide quantitative scan data from several runs? Fig 5C, the mobility shifted lanes are not very convincing.
The bigger issues for correction relate to the writing: 1. The introduction has far too much of results in it. Almost 1/3 of the intro is results. 2. The biggest problem is the discussion. I felt like I was reading the results again when reading the discussion. There are two things that would greatly help. First, rather than reiterating results, explain the difficult parts and show how the results fit into other literature. Second, a figure with a model of what the authors think is happening would be very useful. 3. Although the writing is reasonably good, there are still occasional grammatical errors. If the journal has in-house copy editing, it should be ok. 4. A structured abstract would be nice if it fits with journal policy. 1 Dear Journal editor, Thank you very much for the critical review of our manuscript. We appreciate the comments and are submitting a manuscript revised according to the critiques and questions of the reviewers. The responses and the changes in the text are detailed below each critique. They are marked by //. The changed manuscript with highlight is shown below.
We hope that you find the revised manuscript suitable for publication in Open Biology. We are writing to inform you that the Editor has reached a decision on your manuscript RSOB-19-0228 entitled "DNA topoisomerase IIIβ promotes cyst generation by inducing cyst wall protein gene expression in Giardia lamblia", submitted to Open Biology.
As you will see from the reviewers' comments below, there are a number of criticisms that prevent us from accepting your manuscript at this stage. The reviewers suggest, however, that a revised version could be acceptable, if you are able to address their concerns. If you think that you can deal satisfactorily with the reviewer's suggestions, we would be pleased to consider a revised manuscript.
The revision will be re-reviewed, where possible, by the original referees. As such, please submit the revised version of your manuscript within four weeks. If you do not think you will be able to meet this date please let us know immediately.
To revise your manuscript, log into https://mc.manuscriptcentral.com/rsob and enter your Author Centre, where you will find your manuscript title listed under "Manuscripts with Decisions." Under "Actions," click on "Create a Revision." Your manuscript number has been appended to denote a revision.
You will be unable to make your revisions on the originally submitted version of the manuscript. Instead, please revise your manuscript and upload a new version through your Author Centre.
When submitting your revised manuscript, please respond to the comments made by the referee(s) and upload a file "Response to Referees" in "Section 6 -File Upload". You can use this to document any changes you make to the original manuscript. In order to expedite the processing of the revised manuscript, please be as specific as possible in your response to the referee(s).
Please see our detailed instructions for revision requirements https://royalsociety.org/journals/authors/author-guidelines/ Once again, thank you for submitting your manuscript to Open Biology, we look forward to receiving your revision. If you have any questions at all, please do not hesitate to get in touch.

Sincerely,
The Open Biology Team mailto: openbiology@royalsociety.org Reviewer(s)' Comments to Author(s): Referee: 1 Comments to the Author(s) Sun et al explore the role of TOP3β in regulating Giardia differentiation. They demonstrate that TOP3β localizes to the nucleus and binds to AT rich promoter sequences. They also show that over expression of this enzyme leads to up-regulation of CWP proteins and Myb2. Conversely 3 knockdown with CRISRP based allele deletion results in reduced CWP and Myb2 expression. Overall the results are clearly presented. I reviewed an earlier version of this paper and the authors addressed my major criticisms, so I recommend acceptance with some minor revisions I will outline below.
Major Points 1. The change in localization observed for TOP3β in non-encysting cells (Fig 1)  That TOP3β localization changes from perinuclear to dispersed in the nucleus and is an important point that supports the authors assertion that TOP3β is involved in regulating differentiation.
// The old Fig. S3B (finally moved to Fig. 2C) is a bit misleading, so we changed it to a new one.
The localization of TOP3 is still in the nuclear periphery. In Fig. 2B, The same "nuclear periphery" localization of cells in both the Veg (vegetative) and Enc (encyting) stages was also shown.
// The perinuclear localization of the Giardia TOP3 may have its specific function. We wrote "Similarly, DNA topoisomerase I from yeast and human have been found to have a perinuclear distribution that may help function in DNA replication with perinuclear anchors of chromosomes (35)." (Line154) This same re-localization is not observed in Fig 1 perhaps because the cell is not at the right stage of encystation or is not actually encysting (no CWP1 staining here).
// As we suggested above, there is no re-localization (we found the same "nuclear periphery" localization of TOP3 in Fig. 2B (including Veg and Enc) and It is also interesting that the mutants are more clearly nuclear localized during encystation.
Since there is no quantification of this I don't know if this is due to selected images or a real response to encystation stimulus.
// As we mentioned in paper (lines 159-165). TOP3m1 is also in nuclear periphery. TOP3m2 and m3 have less nuclear peripheral staining. We found consistent staining results for the same staining pattern, so we did not quantify it. What also doesn't make sense is that in S5B there are two bands one that is found in every lane and then a unique shifted band near the top of the blot in the presence of the antibody. In Fig 5C the same band is observed in every well although it does appear darker in the lane where the antibody was labeled. As an alternative to repeating the blot, the authors could quantify the intensity of this band versus the same position in the control lanes to convince us that there is enrichment in their replicates.
// In supershift assays, the anti-TOP3 was added in the reaction and the size of supershift complex (anti-TOP3 -TOP3 -DNA probe) will be bigger than the bound form (TOP3 -DNA probe) and may get stuck near the loading well (arrows in Fig. 5C and Fig. S5B). We quantified the region with arrows and showed the results in Fig. S5C. We added the text: "Both supershift results for the probes cwp1-45/-1 and cwp3 -30/+10F were significant as quantified in Fig. S5." (page 13) 2. Regarding the CRISPR experiment I question the value of using CRISPR/CAS9 over homologous recombination. I am under the impression that CRISPR/CAS9 is not actually providing any benefit and that the authors are observing homologous recombination.
// We hope our thorough analysis can provide a better method for the protozoan Giardia, so we must do a lot of work on the CRISPR/Cas9 system using mlf genes as a model target gene (Lin et al., 2019)(ref 11). Due to the tetraploid genome in two nuclei of Giardia, it could be hard to disrupt a gene completely in Giardia. We only generated knockdown but not knockout mutants. The potential of CRISPR/Cas9 system to complete knockout genes of interest in Giardia awaits further studies to explore. Using a strong promoter to drive the expression of cas9 gene could improve knockout efficiency. For that study, they wanted to ensure the presence of only a single integrated construct in a trophozoite before encysting experiment. They tested clonal populations with PCR, and tested for the absence of episomes by PCR, and analyzed with fluorescence in situ hybridization (FISH). Therefore, they found a single nuclei staining.
// In our TOP3 paper, our aim is to disrupt the top3 gene for functional analysis. We did not make clonal populations. We did not ensure the presence of only a single integrated construct in a trophozoite. We did not ensure the absence of episomes. Therefore, there could be some mixtures of successful and unsuccessful cells. Interestingly we still got 23% disruption efficiency.
The authors mention the use of NHEJ inhibiting drugs which seem to be doing nothing since Ebneter et al showed that homologous recombination can be used to knockout genes in Giardia.
// Cre-lox method in Ebneter et al 2016 is also a very good method. We don't have a chance to work on Cre-lox. We wrote in our CRISPR paper ( I wonder why the authors bother to invoke the use of CRISPR since there is no perceivable benefit? Note that I do not dispute their mutant has a phenotype. My concern is that their pseudo use of CRISPR/CAS9 will confuse readers. 7 // As we wrote above, there is always a need for gene disruption techniques in Giardia. That is why CRISPRi has also been developed. For our lab, disruption of top3 gene is the second successful example (the first one is mlf gene (Lin et al., 2019)(ref 11)). We wanted to check out how many genes can be successfully target-disrupted by CRISPR/Cas9.

Minor Points 1. Fig S1 is corrupted or there was an error in its construction. The region around T328 is not
shown and neither is the red arrow that is mentioned by the legend. Please fix this.

Sentence on Line 157 is confusing. It reads as if it is interesting that ESVs contain CWP1.
To me what is interesting is that the localization of TOP3β is more dispersed in this encysting cell.
// We tried to emphasize the role of TOP3 by writting "Interestingly, the CWP1 protein was stained in the ESVs of TOP3-HA positive stained cells (Fig. 2C), suggesting that TOP3 may function in inducing the ESV and thereby in inducing cyst formation." //As we mentioned above, the old The authors have submitted a detailed and interesting study of the effect of DNA topoisomerase 3b (type 1) on encystation in Giardia. The title focuses on the effect of cyst generation related to cwp expression. However, I'm not convinced that there is enough causal information to warrant the use of "by inducing" in the title.
They have used transfection studies to study the effect of upregulation of TOP3b on expression of the cwps and myb2 on encystation. They have also used CRISPR/Cas9 to reduce expression of top3b and shown decreases in the above. (Of note, knockout is impossible with Giardia because of its polyploidy, so these approaches are reasonable).
Overall, the scientific component of the work seems solid, but I do have some questions regarding the data to be addressed by the authors: Thus, I will begin with the results section: For figure 1 and other figures, the authors simply use p<0.05, but unless <0.001, it would be helpful to see the numbers for CI and actual p value. (1). 56 The small genome suggests Giardia as a simplified life form of evolutionary interest (5). It 57 contains most pathways for life event but with fewer conserved components as compared with 58 yeast (5). Giardia is also a good model for studying single cell differentiation as its easy 59 transition between the trophozoite and cyst forms in vitro (1, 2). After sensing encystation stimuli, 60 trophozoites perform coordinated synthesis of the three cyst wall proteins (CWPs) which are 61 transported through encystation secretory vesicles (ESVs) to form a protective cyst wall (1,2). 62 Signaling molecules and transcription factors, including CDK2, MYB2 (Myb1-like protein in the 63 Giardia genome database), WRKY, PAX1, and E2F1, may play a role in inducing the cwp gene 64 expression (6, 7, 8, 9, 10). We also found that a myeloid leukemia factor (MLF) protein plays an 65 important role in inducing Giardia differentiation into cysts (11). We used our newly developed 66 CRISPR/Cas9 system in G. lamblia for targeted disruption of mlf gene expression to analyze 67 MLF (11). 68 Topoisomerases are essential enzymes that can overcome the topological problems of 69 4 chromosomes during DNA replication, transcription, recombination, and mitosis (12,13). They 70 are involved in cell growth, tissue development, and cell differentiation (12,13,14). The type I 71 topoisomerases function by cutting one strand of DNA, but type II topoisomerases cut two 72 strands of DNA (12,13  Giardia TOP3 (residues 645 to 973) (Fig. S1). The C-terminal zinc ribbon domains of bacterial 126 type IA topoisomerases are important for DNA binding and interaction of RNA polymerase (18, 127 31, 32). Giardia TOP3 also has a conserved Tyr (residue 328), corresponding to the 128 catalytically important Tyrosines of E. coli topoisomerase I (residue 319) and human TOP3 129 (residue 336) ( Fig. 1A; Fig. S1)(20, 33). E. coli topoisomerase III has a unique insertion which is 130 a decatenation loop, which is and not found in Giardia and other eukaryotic TOP3 (Fig. S1)(34). 131 The full-length of Giardia TOP3 has 28.73% identity and 41.32% similarity to that of human 132 TOP3 (Calculated from Fig. S1). A phylogenic tree obtained from the alignment of the 133 7 topoisomerase III proteins from various organisms revealed that Giardia TOP3 (15190) is 134 similar to TOP3 from other organisms, and that Giardia TOP3 (7615) is similar to TOP3 135 from other organisms (Fig. S2). 136 Encystation-induced expression of the top3 gene and perinuclear localization of the 137 TOP3 protein. RT-PCR and quantitative real-time PCR analysis showed that the top3 mRNA 138 increased by ~1.75-fold in 24h encysting cells (Fig. 1B). Western blot analysis with anti-TOP3 139 antibody revealed that the TOP3 level significantly increased during encystation (Fig. 1C). 140 To determine the expression of TOP3 protein, we prepared construct pPTOP3, in which the 141 top3 gene is controlled by its own promoter and contains an HA epitope tag (~1kDa) at its C 142 terminus (Fig. 1D), and stably transfected it into Giardia. Similar to the expression pattern of the 143 endogenous TOP3 protein (Fig. 1C), the level of TOP3 with the HA tag significantly increased 144 during encystation (Fig. 1E). 145 Change of localization of the TOP3 mutants. To further understand the function of Giardia 146 TOP3, we analyzed the effect of mutation of TOP3. The type I topoisomerases use an 147 important Tyr of the cleavage domain as the active-site residue to create a transient 148 single-stranded DNA break by transesterification (13). We tried to understand whether Tyr 328 of 149 TOP3, which corresponds to Tyr 336 of the human TOP3, is also important for its activity (Fig.  150 2A, Fig. S1). Interestingly, the wild-type TOP3-HA was located to the nuclear periphery that 151 partly overlapped with DAPI and slightly to the cytoplasm (Fig. 2B). The perinuclear staining 152 pattern of TOP3-HA is not the endoplasmic reticulum (ER) staining. The typical ER staining of 153 Giardia as shown in BIP staining contains reticulum shape of interconnected network in the 154 cytoplasm and slightly perinuclear staining that did not overlap with DAPI (Fig. S3). Similarly, 155 8 DNA topoisomerases I from yeast and human have been found to have a perinuclear distribution 156 that may help function in DNA replication with perinuclear anchors of chromosomes (35). 157 Interestingly, the CWP1 protein was stained in the ESVs of TOP3-HA positive stained cells 158 (Fig. S32C), suggesting that TOP3 may function in inducing the ESV and thereby in inducing 159 cyst formation. We found that mutation of the Tyr 328 to Phe did not change the localization of 160 TOP3 to the nuclear periphery (TOP3m1, Fig. 2A, Fig. 2C2D). We also found that deletion of 161 the C-terminal 332 amino acids corresponding to the zinc ribbon domain (residues 642-973, 162 pPTOP3m2, Fig. 2A, Fig. 2D2E) resulted in a decrease in perinuclear localization, but an 163 increase in cytosolic localization. Deletion of the C-terminal 552 amino acids corresponding to 164 the zinc ribbon domain and a part of topoisomerase domain (residues 422-973, pPTOP3m3, Fig.  165 2A, Fig. 2E2F) also decreased perinuclear localization, but increased localization to cytosolic 166 vesicles. The background staining was very low as observed with wild-type WB trophozoites 167 (Fig. 2F2G). TOP3m2 and TOP3m3 have lower but still some ability to localize to nuclear 168 periphery (Fig. 2D2E, Fig. 2E2F). The results suggest that the Zinc ribbon domain may play a 169 partial role in the perinuclear localization. 170 Overexpression of TOP3 induced the expression of the cwp1-3 and myb2 genes. We 171 further investigated the effect of the Giardia TOP3 on encystation. We found a significant 172 increase in the CWP1 and MYB2 levels in the TOP3-overexpressing cell line relative to the 173 control cell line (Fig. 1D, Fig. 3A) (36). The mRNA expression of the endogenous top3 plus 174 vector-expressed top3 significantly increased in the TOP3-overexpressing cell line relative to 175 the control cell line (Fig. 3B, Fig. 3C). The mRNA expression of cwp1-3 and myb2 also 176 increased in the TOP3-overexpressing cell line (Fig. 3B, Fig. 3C). In previous studies, we 9 obtained consistent cyst number data for Giardia growth stage due to spontaneous differentiation 178 (37). We found that the cyst number significantly increased in the TOP3-overexpressing cell 179 line (Fig. 3D). Similar results were obtained during encystation (Fig. S3). These findings suggest 180 that overexpression of TOP3 can increase expression of cwp1-3 and myb2 and cyst formation. 181 We further investigated the role of TOP3 by mutation analysis. We found that the levels of 182 TOP3m1 and TOP3m2 were similar to that of wild-type TOP3 during vegetative growth, but 183 TOP3m3 was expressed at a lower level (Fig. 3A). We also found that the CWP1 level 184 significantly decreased in the TOP3m1-and TOP3m3-expressing cell lines relative to the 185 wild-type TOP3-expressing cell line (Fig. 3A). The CWP1 level also significantly decreased in 186 the TOP3m2-expressing cell line, but with a lower effect than in the TOP3m1-and 187 TOP3m3-expressing cell lines (Fig. 3A). We further analyzed whether the transcript levels were 188 changed. As shown by RT-PCR analysis, the mRNA expression of top3m2-HA and 189 top3m3-HA increased compared with that of wild-type top3-HA during vegetative growth, but 190 the mRNA expression of top3m1-HA decreased (Fig. 3B). The mRNA expression of cwp1-3 191 and myb2 significantly decreased in the TOP3m1-m3-expressing cell lines relative to the 192 wild-type TOP3-expressing cell line (Fig. 3B, Fig. 3C). The level of cyst formation 193 significantly decreased in the TOP3m1-m3-expressing cell lines relative to the wild-type 194 TOP3-expressing cell line (Fig. 3D). Similar results were obtained during encystation (Fig. S3). 195 The findings suggest a decrease in encystation-inducing activity of TOP3m1-m3. 196 Oligonucleotide microarray assays confirmed up-regulation of cwp1-3 and myb2 expression in 197 the TOP3-overexpressing cell line to ~1.5 to ~11.9-fold of the levelsthat in the control cell line 198 10 (Fig. 3E). The ran mRNA expression in the TOP3-overexpressing cell line slightly decreased 199 (~0.8-fold) (Fig. 3E). We found that 93 and 40 genes were significantly up-regulated (2-fold) 200 and down-regulated (1/2)(p<0.05) in the TOP3-overexpressing cell line relative to the vector 201 control, respectively (Table S1). The top3 mRNA expression increased by ~2.1-fold (p<0.05) in 202 the TOP3-overexpressing cell line (Fig. 3E). 203 TOP3 has DNA cleavage activity. The type I topoisomerases have ability to bind to and 204 cleave single-stranded DNA (12,13,38). Drosophila TOP3 cleaves DNA by forming a covalent 205 topoisomerase-DNA complex (39). To test DNA cleavage activity of TOP3, we expressed 206 TOP3 in E. coli and purified it to >95% homogeneity. We performed DNA cleavage assays with 207 purified recombinant TOP3 and pBluescript SK(+) plasmid. As shown in Fig. 4A, TOP3 has 208 DNA cleavage activity. 209 Norfloxacin, a type II topoisomerase inhibitor, also inhibits E. coli topoisomerase I (type IA) 210 at higher concentrations, resulting in anti-bacteria activity (25). Norfloxacin can inhibit 211 topoisomerases by stabilizing covalent topoisomerase-DNA cleavage complexes (26). To 212 understand whether norfloxacin can inhibit Giardia TOP3 DNA cleavage activity, we also 213 performed DNA cleavage assays with norfloxacin. As shown in Fig. 4B, the addition of 214 norfloxacin increased the amount of linear DNA, suggesting that norfloxacin can stabilize the 215 TOP3-DNA cleavage complex. We also tried to understand whether the products are from the 216 covalent TOP3-DNA cleavage complex. In a normal condition of the cleavage assay, proteinase 217 K was included to stop the reaction by removing TOP3 from the cleavage complex (Fig. 4C). 218 When proteinase K was not included, the TOP3-DNA cleavage complex can not enter the gel 219 ( Fig. 4C), suggesting that TOP3 can form a cleavage complex with DNA. The results indicate 220 11 that TOP3 may function as a topoisomerase in Giardia. 221 Norfloxacin has anti-Giardia effect. Norfloxacin is an inhibitor of type IA and type II 222 topoisomerases with anti-bacteria activity (25). We found that norfloxacin increased DNA 223 cleavage activity of TOP3, indicating that norfloxacin can trap the cleavage complex of TOP3 224 (Fig. 4B). We also found that treatment with norfloxacin significantly reduced Giardia 225 trophozoites growth (Fig. 4D). The half-maximal inhibitory concentration (IC50) of norfloxacin 226 on Giardia was 497M (Fig. 4D). The addition of 497M norfloxacin also decreased cyst 227 formation by 67% (Fig. S3). The results from the topoisomerase inhibitor norfloxacin suggest 228 that TOP3 may regulate Giardia growth and differentiation into cysts. 229 TOP3 mutants have a lower cleavage activity. To understand which regions are important 230 for cleavage activity, the specific TOP3 mutants were expressed in E. coli, and purified ( Fig.  231 5A), and tested by cleavage assays. We found a decrease in cleavage activity of TOP3m1 (with 232 a mutation of the catalytically important Tyr 328) and TOP3m3 (with a deletion of C-terminal 233 552 amino acids) and slight decrease in cleavage activity of TOP3m2 (with a deletion of the 234 C-terminal 332 amino acids) ( Fig. 2A, Fig. 5B). 235 TOP3 has DNA-binding activity and its mutants have lower DNA-binding activity. We 236 further tested DNA-binding activity of TOP3. Electrophoretic mobility shift assays were 237 performed with the purified TOP3 protein and double-stranded DNA sequences from the 238 5'-flanking region of the cwp genes. Incubation of a labeled double-stranded DNA probe, 239 cwp1-45/-1, with TOP3 resulted in the formation of retarded bands (Fig. 5C). The binding 240 specificity was confirmed by competition and supershift assays (Fig. 5C). The formation of the 241 shifted cwp1-45/-1 bands was competed by a 200-fold molar excess of unlabeled cwp1-45/-1, 242 12 but not by the same excess of a nonspecific competitor, 18S-30/-1 (Fig. 5C), suggesting that 243 TOP3 did not bind to GC rich sequence. The bound form on cwp1-45/-1 could be supershifted 244 by an anti-TOP3 antibody (Fig. 5C). The results suggest that Giardia TOP3 can bind to the 245 cwp1 promoter (-45/-1 region). TOP3 was also shown to bind to the cwp3 promoter, 246 cwp3-30/+10 (Fig. 5D). To understand which regions of TOP3 are important for DNA binding, 247 the specific mutants were tested for their DNA-binding activity. There was only a slight decrease 248 in the DNA-binding activity of TOP3m1, but there was a far more decrease in the DNA-binding 249 activity of TOP3m2 and TOP3m3 (Fig. 5D). 250 TOP3 was also shown to bind to cwp1-90/-46, and within this region it weakly bound to the 251 5'-region (cwp1-90/-69), the middle region (cwp1-78/-58), or the 3'-region (cwp1-68/-46) (Fig.  252   6). We found that TOP3 bound strongly to the cwp2-60/-31 and cwp3-30/+10 probes, and 253 weakly to the cwp2-30/+8 and cwp3-60/-31 probes (Fig. 6), suggesting that TOP3 can bind to 254 other encystation-induced promoters, cwp2 and cwp3. TOP3 also bound to a well characterized 255 ran core AT-rich promoter, ran-51/-20 (40) and weakly to ran-30/-1 (Fig. 6). TOP3 did not bind 256 to the 18S-30/-1, and 18S-60/-31 probes, which do not contain AT-rich sequence (Fig. 6). 257 Interestingly, TOP3 also weakly bound to a poly(A) sequence and a poly(A) sequence with a T 258 or TT insertion (Fig. S4), indicating that the TOP3 binding sequence contains AT-rich 259 sequences. The results suggest that TOP3 can strongly bind to the cwp1-3 and ran AT-rich 260 promoter regions. 261 We also performed DNA-binding assays with single-stranded DNA probes. We found that 262 TOP3 can bind to the single-stranded DNA of cwp1-3 promoters (cwp1-45/-1F, cwp1-90/-46F, 263 cwp2-60/-31F, cwp3-30/+10F) (Fig. S5), but it can not bind to the single-stranded DNA of 18S 264 13 promoter (18S-30/-1F), which does not contain AT-rich sequence (Fig. S5). TOP3 bound to 265 cwp3-30/+10F could be supershifted by an anti-TOP3 antibody (Fig. S5). Both supershift 266 results for the probes cwp1-45/-1 and cwp3 -30/+10F were significant as quantified in Fig. S5. 267 The results suggest that Giardia TOP3 can bind to the single-stranded DNA of the cwp1-3 268

promoters. 269
Recruitment of TOP3 to the top3, cwp1-3 and myb2 promoters. We further used 270 norfloxacin-mediated topoisomerase immunoprecipitation assay (28, 29), a method similar to 271 ChIP assays, to study the association of TOP3 with the specific promoters. The addition of 272 norfloxacin may increase the cleavage complex formation and thereby could increase ChIP 273 sensitivity (29). We found that TOP3 was associated with its own promoter and the cwp1-3, 274 myb2, and ran promoters during encystation (Fig. 7A, Fig. 7B). However, TOP3 was not 275 associated with the U6 snRNA promoter (transcribed by pol III), or 18S ribosomal RNA 276 promoter (transcribed by pol I) which has no TOP3 binding site (Fig. 7A). 277 Interaction between MYB2 and TOP3 associated complexes. It is possible that TOP3 278 may regulate encystation-induced cwp genes by interacting with other transcription factors. We 279 further tried to understand whether TOP3 can interact with the encystation-induced MYB2 280 transcription factor (7). We performed co-immunoprecipitation experiments using the 281 TOP3-overexpressing cell line. The TOP3-HA protein (~108 kDa) was expressed in the 282 pPTOP3 stable cell line but not in the control cell line (5'5N-Pac) (Fig. 7C) as detected by 283 anti-HA antibody in Western blots (Fig. 7C). Overexpression of TOP3 in the pPTOP3 cell line 284 also can be confirmed by the anti-TOP3 antibody (Fig. 7C). We found that TOP3 285 overexpression resulted in an increase in the MYB2 level (Fig. 7C). However, the ISCS level 286 14 decreased by TOP3 overexpression (Fig. 7C). We lysed the cells and immunoprecipitated 287 TOP3-HA with anti-HA antibody. Western blots of immunoprecipitates probed with anti-HA 288 and anti-MYB2 indicate that MYB2 co-precipitates with TOP3-HA in the absence or presence 289 of DNase (Fig. 7D). The anti-HA antibody did not immunoprecipitate TOP3-HA and MYB2 in 290 the control cell line (Fig. 7D), nor did it immunoprecipitate ISCS in the pPTOP3 cell line (Fig.  291   7D). Far Western blot analysis confirmed the interaction between the N-terminal region of 292 MYB2 (MYB2-N) and TOP3-HA (Fig. 7E). The results suggest an interaction between MYB2 293 and TOP3 in a complex. 294 Targeted disruption of the top3 gene reduced expression of cwp1-3 and myb2. To further 295 understand the function of TOP3, we analyzed the effect of top3 gene disruption. We 296 developed a CRISPR/Cas9 system to disrupt the mlf gene (11). We further adapted this system to 297 the study the role of TOP3. The CRISPR/Cas9 constructs were transfected into G. lamblia and 298 TOP3td stable transfectants were established under puromycin selection (Fig. 8A). Scr7, an 299 inhibitor of NHEJ, was added in the first replenishment of puromycin containing medium to 300 increase knock-in efficiency via homologous recombination (11). The replacement of the top3 301 gene with the puromycin acetyltransferase (pac) gene was confirmed by PCR and sequencing 302 analysis of genomic DNA (Fig. 8B, Fig. 8C, Fig. S6). The results show a successful disruption of 303 the top3 gene by about 23% and a partial replacement of the top3 gene with the pac gene ( Fig.  304   8B, Fig. 8C). It has been shown that G418 has cytotoxicity on mammalian cells (41). The 305 toxicity of G418 may be mediated by blocking polypeptide synthesis during translation 306 elonglation (41). Inhibition of protein synthesis further results in oxidative stress and cell death 307 (41). G418 also inhibits the growth of Giardia and can be used to select transfected cells (37). 308 15 We used G418 to test the drug sensitivity of the TOP3td cell line and found that TOP3td cell 309 line exhibited increased sensitivity to it compared to the control cell line (Fig. 8D). We also 310 found that the level of cyst formation significantly decreased in the TOP3td cell line relative to 311 the control cell line during vegetative growth (Fig. 8E). 312 Western blot analysis confirmed the decrease of the TOP3 protein level in the TOP3td cell 313 line relative to the control cell line (Fig. 8F). We found that the CWP1 level also significantly 314 decreased in the TOP3td cell line relative to the control cell line (Fig. 8F). We further found the 315 mRNA expression of top3, cwp1-3, or myb2 significantly decreased in the TOP3td cell line 316 relative to the control cell line (Fig. 8G, Fig. S7). Similar results were obtained during 317 encystation (Fig. S8). The findings suggest a decrease in expression of cwp1-3 and myb2, drug 318 sensitivity, and cyst formation by targeted disruption of the top3 gene. 319 We further tried to analyze results without puromycin. After selection, puromycin was 320 removed from the TOP3td cell line to obtain the TOP3td -pu cell line. We found a successful 321 disruption of the top3 gene by about 31% and a partial replacement of the top3 gene with the 322 pac gene (Fig. S9). The level of cyst formation significantly decreased in the TOP3td -pu cell 323 line relative to the control -pu cell line (Fig. S9). The levels of CWP1 and top3, cwp1-3, and 324 myb2 mRNA also significantly decreased in the TOP3td -pu cell line relative to the control -pu 325 cell line (Fig. S9). Similar results were obtained during encystation (Fig. S10). The findings 326 suggest a decrease in expression of cwp1-3 and myb2 and cyst formation by targeted disruption 327 of the top3 gene without puromycin. 328 16 Discussion 329 The type I topoisomerase are involved in cell growth, tissue development, and cell 330 differentiation (19; 20; 21). In this study, we identified and characterized a type IA 331 topoisomerase, TOP3, from Giardia. TOP3 has DNA-binding and cleavage activity of 332 topoisomerases (Fig. 4A, Fig. 5C), as its catalytically important domains and residues are 333 conserved. This suggests that the type IA topoisomerases may have evolved before divergence of 334 Giardia from the main eukaryotic line of descent. The presence of at least one type II 335 topoisomerase and one type IA topoisomerase suggests that they play the necessary roles in 336 different organisms (42). Similarly, Giardia also has one type II topoisomerase and two type IA 337 topoisomerases (orfs 16975, 15190, and 7615)(28). the cyst wall, cwp1-3 (Fig. 6, Fig. S4, Fig. 9). Similarly, Drosophila topoisomerase IIIβ prefers to 355 bind to AT-rich DNA sequences (39). We hypothesize that TOP3, MYB2, and other 356 transcription factors, can bind to the AT-rich elements or the proximal upstream regions and form 357 complexes ( Fig. 9) (7). This interaction may recruit RNA polymerase II to activate cwp1-3 358 transcription (Fig. 9). The overexpressed TOP3 increased the levels of CWP1 and MYB2 and 359 cwp1-3 and myb2 mRNA and cyst formation (Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D). Results from 360 the CRISPR/Cas9 system suggest a decrease in expression of CWP1, cwp1-3 and myb2, and cyst 361 formation by targeted disruption of the top3 gene (Fig. 8E, Fig. 8F, Fig. 8G). Furthermore, the 362 addition of an inhibitor of type IA topoisomerases, norfloxacin, inhibited cell growth and cyst 363 formation (Fig. 4D, Fig. S3). The results suggest a positive role of TOP3 in inducing cwp1-3 364 and myb2 gene expression and Giardia encystation. 365 Many Giardia gene promoters have the AT-rich initiator elements responsible for promoter 366 activity and transcription start site selection (40,57,58). We have identified several transcription 367 factors involved in the transactivation of the cwp genes, and they can bind to the AT-rich 368 elements or the proximal upstream regions of the cwp promoters (7, 8,9,10,59,60,61). It has been 369 reported that type IA topoisomerases can bind to DNA and activate gene expression (31,32, 50). 370 In this study, we found that TOP3 can also bind to the AT-rich elements of the cwp promoters in 371 vitro (Fig. 6). Norfloxacin-mediated topoisomerase immunoprecipitation assays confirmed the 372 18 association of TOP3 with its own promoter and the cwp1-3 and myb2 promoters but not with 373 the U6 promoter in vivo (Fig. 7A, Fig. 7B). ChIP assays confirmed the binding of 374 encystation-induced transcription factors E2F1and MYB2 to the cwp and myb2 gene promoters 375 previously (7,10). E2F1and MYB2 may interact together to activate expression of the cwp genes 376 (10). We also found that MYB2 is co-immunoprecipitated with TOP3 (Fig. 7D). Treatment with 377 DNase did not prevent the immunoprecipitation of Myb2 with TOP3 (Fig. 7D), suggesting that 378 the interaction depends on protein-protein interaction but not DNA. Far Western blot analysis, a 379 non-antibody method, was further used to confirm this interaction between the MYB2-N and 380 TOP3-HA to avoid the nonspecific problem (Fig. 7E). The MYB2-N can be phosphorylated by 381 CDK2, which is involved in inducing encystation (6). The interaction of TOP3 and MYB2, and 382 other transcription factors binding to the AT-rich elements or the proximal upstream regions (7), 383 may be required for promoter activity and accurate transcription start site selection. TOP3 can 384 bind to AT-rich elements of both the constitutive ran gene and encystation-induced cwp genes 385 (Fig. 6). However, overexpressed TOP3 can induce the CWP1 level but not the RAN level (Fig.  386   3A). This could be due to a lack of cooperation of the encystation-specific transcription factors to 387 transactivate the constitutive ran gene. Similar results were found in studies of other 388 transcription factors (7,8,9,10,60,61). Interestingly, expression of all three TOP3 mutants 389 (TOP3m1-3) have led to less expression levels of CWP1, cyst formation, and cwp1-3 and myb2 390 mRNA relative to the wild-type TOP3 (Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D). However, they it 391 still have leds to more expression levels of CWP1, cyst formation, and cwp1-3 and myb2 mRNA 392 relative to the vector control (Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D). It is possible that the mutants 393 that were relatively overexpressed, may still interact with transcription factors, such as Myb2, to 394 19 activate expression of the cwp genes. 395 We also found that the top3β promoter contains the MYB2 binding sequences (Fig. S10) The DNA topoisomerase domain (residues 159 to 610) of Giardia TOP3 is near the central 406 region (Fig. 1A). A zinc ribbon domain is located in the C terminus (residues 645 to 973) (Fig.  407   S1). Studies suggest that the C-terminal zinc ribbon domains of E. coli topoisomerase I and 408 Drosophila topoisomerase IIIα are important for DNA binding (31, 32). The C-terminal zinc 409 ribbon domain of E. coli topoisomerase I interacts with RNA polymerase to help bring to the 410 transcription site for relaxation reaction (18). We found that Ddeletion of the C-terminal 332 411 amino acids (residues 642-973) corresponding to the zinc ribbon domain of TOP3 (residues 412 642-973) resulted in reduction in DNA-binding activity, but only slight decrease in cleavage 413 activity (TOP3m2) (Fig. 5D, Fig. S1), suggesting that the zinc ribbon domain is important for 414 DNA-binding. Deletion of C-terminal 552 amino acids (residues 422-973) corresponding to the 415 zinc ribbon domain and a part of topoisomerase domain of TOP3 (residues 422-973) resulted in 416 20 reduction in the cleavage activity (TOP3m3) (Fig. 1A, Fig. 5B), indicating that the 417 topoisomerase domain is important for cleavage activity. We also found that a mutation of the 418 catalytically important Tyr 328 resulted in a decrease in cleavage activity (TOP3m1) (Fig. 1A, 419 Fig. 5B). Tyr327 of topoisomerase III in Leishmania donovani is also catalytically important 420 (63). Interestingly, mutation of the Tyr 328 to Phe did not change its perinuclear localization in 421 both vegetative and encysting cells (TOP3m1) (Fig. 2C2D). Mutation of this important Tyr also 422 resulted in a significant decrease in the levels of CWP1, cyst formation, and cwp1-3 and myb2 423 mRNA (Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D, Fig. 5B, Fig. 5D), suggesting a correlation of DNA 424 cleavage activity and in vivo function. 425 Typically, nuclear localization signal (NLS) is a region rich with basic amino acids. Two 426 putative NLS motifs were predicted in TOP3 using the PSORT program (http://www.psort.org/), 427 including RKHR at 970, and RRAAQPKRHGPRGRKHR at 957. We also found that deletion of 428 the C-terminal 332 (residues 642-973) or 552 amino acids (residues 422-973) resulted in a 429 decrease but not complete loss of perinuclear localization (TOP3m2 or m3) (Fig. 2D2E, Fig.  430 2E2F), suggesting that the C-terminal zinc ribbon domain may play a partial role in the 431 perinuclear localization and that other NLS motifs may be present in TOP3. Deletion of the 432 C-terminal 332 amino acids (TOP3m2) resulted in a significant decrease in the levels of CWP1, 433 cyst formation, and cwp1-3 and myb2 mRNA (TOP3m2) (Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D), 434 but the effect is lower than the TOP3m1 and TOP3m3. As Shown in Fig. 2D2E, TOP3m2 435 still has some ability to localize to nuclear periphery. Interestingly, DNA cleavage activity was 436 less affected in this mutant (TOP3m2) (Fig. 5B), suggesting again-a correlation of DNA 437 cleavage activity and in vivo function. The results suggest that TOP3 may enhance the 438 21 encystation-induced expression of cwp1-3 and myb2 through its cleavage activity. 439 Drosophila topoisomerase IIIβ prefers to bind to AT-rich DNA sequences (39). We also found 440 that Giardia TOP3 can bind to the AT-rich promoter elements of the cwp genes (Fig. 6), and 441 that it may bind to and up-regulate the cwp gene promoters to induce Giardia encystation. We 442 found that 93 and 40 genes were significantly up-regulated and down-regulated in the 443 TOP3-overexpressing cell line relative to the vector control, respectively (Table S1). In addition, 444 targeted disruption of the top3 gene resulted in a decrease in expression of cwp1-3 and myb2 445 and cyst formation using the CRISPR/Cas9 system (Fig. 8), suggesting TOP3 may induce 446 Giardia encystation. We also found that chemosensitivity of the TOP3td cell line significantly 447 increased by the addition of G418 (Fig. 8D), suggesting that TOP3 may affect many genes 448 involved in cell growth to survive antibiotic stress. 449 Norfloxacin, which belongs to quinolones, is an inhibitor of the type II topoisomerases, 450 including topoisomerase II (DNA gyrase) and topoisomerase IV, and (25). it It also inhibits type 451 IA topoisomerases at higher concentrations (25). Quinolones can stabilize the topoisomerase 452 II-DNA complex and prevent religation of DNA, resulting in anti-topoisomerase activity (23, 26,453 64). A model suggests that quinolones can form complexes with DNA and topoisomerase IV and 454 create barriers to DNA replication (65). We found that the addition of norfloxacin increased the 455 cleavage complex formation of Giardia TOP3 (Fig. 4B), and that the addition of norfloxacin 456 significantly decreased cell growth and cyst formation (Fig. 4D, Fig. S3 lamblia genome database). The TOP3 coding region with 300 bp of 5'-flanking region was 499 cloned and the nucleotide sequence was determined. To isolate the cDNA of the top3 gene, we 500 performed RT-PCR with the top3-specific primers using total RNA from G. lamblia. For 501 RT-PCR, 5g of DNase-treated total RNA from vegetative and 24h encysting cells was mixed 502 with oligo (dT)12-18 and random hexamers and Superscript II RNase H-reverse transcriptase 503 (Invitrogen). Synthesized cDNA was used as a template in subsequent PCR with primers top3F 504 and top3R. Genomic and RT-PCR products were cloned into pGEM-T easy vector (Promega) 505 and sequenced (Applied Biosystems, ABI). Comparison of genomic and cDNA sequences 506 showed that the top3 gene contained no introns. 507 Genomic DNA extraction, PCR and quantitative real-time PCR analysis. Synthetic 508 oligonucleotides used are shown in Table S2. Genomic DNA was isolated from trophozoites 509 using standard procedures (Sambrook, 1989). For PCR, 250 ng of genomic DNA was used as a 510 template in subsequent PCR. PCR analysis of top3 (XM_001709742.1 , orf 15190), cwp1 511 (U09330, orf 5638), cwp2 (U28965, orf 5435), and ran (U02589, orf 15869) genes was 512 performed using primers top3F (PCR1F) and top3R (PCR1R), PCR2F and PCR2R, cwp1F 513 and cwp1R, cwp2F and cwp2R, ranF and ranR, respectively. For quantitative real-time PCR, 514 SYBR Green PCR master mixture was used (Kapa Biosystems). PCR was performed using an 515 Applied Biosystems PRISMTM 7900 Sequence Detection System (Applied Biosystems). 516 Specific primers were designed for detection of the top3, cwp1, cwp2, and ran genes: 517 top3realF and top3realR; cwp1realF and cwp1realR; cwp2realF and cwp2realR; ranrealF and 518 25 ranrealR. Two independently generated stably transfected lines were made from each construct 519 and each of these cell lines was assayed three separate times. The results are expressed as a 520 relative expression level over control. Student's t-tests were used to determine statistical 521 significance of differences between samples. 522 RNA extraction, RT-PCR and quantitative real-time PCR analysis. Synthetic 523 oligonucleotides used are shown in Table S2. Total RNA was extracted from G. lamblia cell line 524 during vegetative growth or encystation using TRIzol reagent (Invitrogen). For RT-PCR, 5g of 525 DNase-treated total RNA was mixed with oligo (dT)12-18 and random hexamers and Superscript 526 II RNase Hreverse transcriptase (Invitrogen). Synthesized cDNA was used as a template in 527 subsequent PCR. Semi-quantitative RT-PCR analysis of top3 (XM_001709742.1, orf 15190), 528 top3-ha, cwp1 (U09330, orf 5638), cwp2 (U28965, orf 5435), cwp3 (AY061927, orf 2421), 529 myb2 (AY082882, orf 8722), ran (U02589, orf 15869), and 18S ribosomal RNA (M54878, orf 530 r0019) gene expression was performed using primers top3865F and top3926R, top3865F and 531 HAR, cwp1F and cwp1R, cwp2F and cwp2R, cwp3F and cwp3R, myb2F and myb2R, ranF and 532 ranR, 18SrealF and 18SrealR, respectively. For quantitative real-time PCR, SYBR Green PCR 533 master mixture was used (Kapa Biosystems). PCR was performed using an Applied Biosystems 534 PRISMTM 7900 Sequence Detection System (Applied Biosystems). Specific primers were 535 designed for detection of the top3, top3-ha, cwp1, cwp2, cwp3, myb2, ran, and 18S ribosomal 536 RNA genes: top3realF and top3realR; top3HAF and HAR; cwp1realF and cwp1realR; 537 cwp2realF and cwp2realR; cwp3realF and cwp3realR; myb2realF and myb2realR; ranrealF and 538 ranrealR; 18SrealF and 18SrealR. Each primer pairs were determined for amplification 539 efficiency ~95% based on the slope of the standard curve. Two independently generated stably 540 26 transfected lines were made from each construct and each of these cell lines was assayed three 541 separate times. The results are expressed as a relative expression level over control. Student's 542 t-tests were used to determine statistical significance of differences between samples. 543 Plasmid construction. Synthetic oligonucleotides used are shown in Table S2. All constructs 544 were verified by DNA sequencing with a BigDye Terminator 3.1 DNA Sequencing kit and an 545 Applied Biosystems 3100 DNA Analyzer (Applied Biosystems). Plasmid 5'5N-Pac was a gift 546 from Dr. Steven Singer and Dr. Theodore Nash (Singer et al., 1998). Plasmid pgCas9 has been 547 previously described (11). To make construct pPTOP3, the top3 gene and its 300 bp of 548 5'-flanking region were amplified with oligonucleotides top3NF and top3MR, digested with 549 NheI and MluI, and cloned into NheI and MluI digested pPop2NHA (72). To make construct 550 pPTOP3m1, the top3 gene was amplified using two primer pairs top3m1F and top3MR, 551 and top3m1R and top3NF. The two PCR products were purified and used as templates for a 552 second PCR. The second PCR also included primers top3NF and top3MR, and the product 553 was digested with NheI and MluI and cloned into the NheI and MluI digested pPop2NHA (72). 554 To make construct pPTOP3m2 or pPTOP3m3, the top3 gene was amplified using primers 555 top3NF and top3m2MR or top3m3MR, digested with NheI and MluI, and cloned into NheI 556 and MluI digested pPop2NHA (72). 557 The 620-bp 5'-flanking region of the top3 gene was amplified with oligonucleotides 558 top35HF and top35NR, digested with HindIII/NcoI and cloned into HindIII/NcoI digested 559 5'5N-Pac, resulting in TOP35. The 700-bp 3'-flanking region of the top3 gene was amplified 560 with oligonucleotides top33XF and top33KR, digested with XbaI/KpnI and cloned into 561 XbaI/KpnI digested TOP35, resulting in TOP353. We used gene synthesis services from IDT 562 27 to obtain the fragment top3-guide. The NCBI Nucleotide Blast search was used to avoid the 563 potential off-target effects of guide sequence. The top3-guide was digested with KpnI/EcoRI 564 and cloned into KpnI/EcoRI digested TOP353, resulting in pTOP3td.