Construction of a Yeast Cell-Based Assay System to Analyze SNAP25-Targeting Botulinum Neurotoxins

Herein, we describe a yeast cell-based assay system to analyze SNAP25-targeting botulinum neurotoxins (BoNTs). BoNTs are protein toxins, and, upon incorporation into neuronal cells, their light chains (BoNT-LCs) target specific synaptosomal N-ethylmaleimide-sensitive attachment protein receptor (SNARE) proteins, including synaptosomal-associated protein 25 (SNAP25). BoNT-LCs are metalloproteases, and each BoNT-LC recognizes and cleaves conserved domains in SNAREs termed the SNARE domain. In the budding yeast Saccharomyces cerevisiae, the SNAP25 ortholog Spo20 is required for production of the spore plasma membrane; thus, defects in Spo20 cause sporulation deficiencies. We found that chimeric SNAREs in which SNARE domains in Spo20 are replaced with those of SNAP25 are functional in yeast cells. The Spo20/SNAP25 chimeras, but not Spo20, are sensitive to digestion by BoNT-LCs. We demonstrate that spo20∆ yeasts harboring the chimeras exhibit sporulation defects when various SNAP25-targeting BoNT-LCs are expressed. Thus, the activities of BoNT-LCs can be assessed by colorimetric measurement of sporulation efficiencies. Although BoNTs are notorious toxins, they are also used as therapeutic and cosmetic agents. Our assay system will be useful for analyzing novel BoNTs and BoNT-like genes, as well as their manipulation.


Introduction
Botulinum neurotoxins (BoNTs) are protein neurotoxins that target nerve cells and inhibit neurotransmitter release [1]. BoNTs are traditionally classified into seven serotypes, BoNT/A to/G [2]. The neurotoxins are predominantly produced by Clostridium species, but BoNT-like genes are also found in other organisms [3,4]. Although BoNTs are notorious toxins, they are also used as therapeutic agents [5]. Structurally, BoNTs are composed of a heavy chain (HC) and light chain (LC) interconnected via a disulfide bond [2]. The HC is subdivided into two functionally distinctive domains termed the H C and H N ; these domains are required for binding to target cells and translocation of the LC into the cytosol, respectively. In the cytosol of nerve cells, LC is released from HC by cleavage of the disulfide bond [6]. LC is a metalloprotease that can cleave the synaptic vesicle fusion machinery; this activity causes an abrogation of neurotransmitter release [1].
The synaptosomal N-ethylmaleimide-sensitive attachment protein receptor (SNARE) proteins serve as membrane fusion machinery in eukaryotic cells [7]. Various SNAREs are deployed in the intracellular vesicle transport system, and the localization of each SNARE protein is strictly regulated. When membranes fuse, SNARE proteins present in the opposing membranes form a tight complex called the trans-SNARE complex or SNAREpin [8][9][10]. SNARE proteins have one or two α-helical heptad repeats called the SNARE domains [11]. For the formation of the trans-SNARE complex, four SNARE domains provided by three or four SNARE proteins generate a tight coiled-coil interaction [12]. Formation of the trans-SNARE complex can induce membrane fusion by pulling the opposing membranes into close proximity [8][9][10]13]. SNARE proteins form fusogenic trans-SNARE complexes

Yeast Strains, Growth Media, and Cell Culture
The S. cerevisiae strains used in this study are listed in Table S3. To construct the spo20∆ diploid strain, a DNA fragment for the gene disruption was amplified by PCR using pFA6a-His3MX6 [27] as a template and HXO26 and HXO27 as primers. The PCR fragment was integrated into AN117-4B and AN117-16D [28], and the resulting haploid strains were crossed to generate diploid cells. To construct the sec9∆ mutant, pRS316TEF-Sec9 was first transformed into YPH499 [29], and then genomic SEC9 was disrupted in the transformants.
For the deletion of SEC9, the PCR cassette for the gene deletion was generated by PCR using CSL1 and CSL2 as primers, and pFA6a-His3MX6 as a template.
To induce sporulation, yeast cells were cultured in 5 mL of YPAD liquid medium (yeast extract 10 g/L, peptone 20 g/L, adenine 30 mg/L, and glucose 20 g/L) overnight. Then, 0.1 mL of the culture was transferred into 5 mL of YPAcetate liquid medium (yeast extract 10 g/L, peptone 20 g/L, and potassium acetate 20 g/L) and cultured overnight. Cells were then centrifugated and resuspended in sporulation medium (2% potassium acetate) at a concentration of 3 × 10 7 cells/mL and cultured for 24 h, 48 h, or 72 h at 30 • C.

Growth Assay
To analyze cell growth on plates, yeast cells were transformed with appropriate plasmids and the colonies were streaked on SD (6.7 g/L yeast nitrogen base, 2 g/L dropout mix without appropriate selectable supplements, and 20 g/L glucose) plates. After 3 day incubation at 30 • C, the cells were streaked on new SD plates and grown for 1 day at 30 • C. For complementation assays of Sec9/SNAP25 chimeras, sec9∆ cells harboring pRS316TEF-Sec9 were transformed with appropriate plasmids, and the transformants were grown on SD plates for 3 days. Then, the cells were streaked on SD plates supplemented with 1 mg/mL 5-fluoroortic acid (5-FOA) and grown for 2 days at 30 • C.
To draw growth curves, yeast cells were precultured in SD medium overnight to an OD 660 of approximately 1.0. Cells were then centrifugated and resuspended in 20 mL of SD medium to an OD 660 of approximately 0.1. OD 660 was measured every 8 h.

Sporulation Assay
To count spores, cells in sporulation medium (0.1 mL) were collected by centrifugation, washed with water once, and resuspended in 100 µL of water. Under the light microscope, the number of asci were counted among cells found in certain area (field of view). This analysis was performed in several areas, and at least 500 cells were counted. Asci and other cells were discriminated by their morphology; asci were recognized as cells containing 2-4 smooth and spherical spores.
For the colorimetric assay, spores were generated in sporulation medium for 48 h as described above. Then, 0.2 mL of the spore suspension was taken into a 96-well plate, and fluorescence quantification was performed with a microplate reader (Synergy H4, BioTek, Winooski, VT, USA) at an excitation wavelength of 285 nm and an emission wavelength of 425 nm. A 2% potassium acetate solution was used as a blank control. Measured values were directly used as indices to represent sporulation efficiencies.

Western Blotting
Cell extracts were made by disruption with glass beads in urea buffer (8 M urea and 1 mM phenylmethylsulphonyl fluoride) followed by 10 min of centrifugation at 21,500× g. Protein concentrations in the supernatants were determined using a Nano-Drop (Thermo-Scientific, Shanghai, China). Then, 100 µg of proteins were subjected to 10% SDS-PAGE and transferred to PVDF membranes (Bio-Rad, Shanghai, China). Mouse anti-FLAG (Transgen, Beijing, China), mouse anti-GFP (Transgen, Beijing, China), mouse anti-HA (Transgen, Beijing, China), or mouse anti-actin (Transgen, Beijing, China) antibodies were used as primary antibodies at 1:5000 dilutions. Goat anti-mouse IgG-HRP (Transgen, Beijing, China) was used as a secondary antibody at 1:5000 dilution. Signals were visualized by Clarity Western ECL Substrate (Bio-Rad, Shanghai, China). Tanon-5200Multi (Tanon, Shanghai, China) was used to obtain Western blot images. Quantification of the signals were performed with ImageJ software [30]. Intensities of target protein band were normalized to actin (the intensity of the target protein band was divided by the relative intensity of the corresponding actin band). Then, relative intensities of target protein bands were calculated.

Microscopy
Microscopy images were obtained using a Nikon C2 Eclipse Ti-E (Nikon, Shanghai, China) inverted microscope with a DS-Ri camera equipped with NIS-Element AR software.

Statistics
All experiments were performed with three or four independent samples; cells from different colonies were cultured on different days. Differences between the analyzed samples were considered significant at p < 0.05. Statistical significance was determined using a two-tailed unpaired Student's t-test calculated with Prism 9.0.0 software.

Production of Functional Chimeras of Yeast and Human SNAP25
Deletion of SEC9 causes a lethal phenotype in yeast cells, which was not complemented by expression of SNAP25 ( Figure S1a,b). To establish a yeast system to analyze BoNT-LCs targeting SNAP25, we constructed Sec9/SNAP25 chimeras. Sec9 has two SNARE domains; the N-and C-terminal SNARE domains are referred to as SNARE(N) and SNARE(C), respectively ( Figure S1c). To this end, we first constructed a chimera in which SNARE(C) in Sec9 was replaced with that of SNAP25 ( Figure S1c). To detect the chimera, green fluorescent protein (GFP) was added to the N-terminus (the fusion protein is referred to as GFP-Sec9/SNAP25(C)). The sec9∆ cells harboring GFP-Sec9 grew, showing that the addition of GFP at the N-terminus of Sec9 did not abrogate the function of Sec9 ( Figure S1d). While the levels of GFP-Sec9/SNAP25(C) expressed in sec9∆ cells were similar to those of GFP-Sec9 ( Figure S1e), the cells harboring GFP-Sec9/SNAP25(C) failed to grow ( Figure S1d). Sec9/SNAP25(C) chimera without any tagging did not complement sec9 deletion either ( Figure S1f). To assess which amino-acid residues in the SNARE domain are required for Sec9 function in the SNARE(C) region, we constructed a series of chimeras in which various parts of the SNARE(C) region were replaced with the corresponding regions of SNAP25 (schematic diagram is shown in Figure S1f). The results of this analysis showed that only seven amino acids of SNARE(C) in Sec9 were exchanged to produce a functional chimera ( Figure S1f). Thus, the function of the SNARE domain in Sec9 was hardly substituted by that of SNAP25.
S. cerevisiae has another SNAP25 ortholog Spo20 [24]. Thus, we examined whether the SNARE domains of SNAP25 were functional in Spo20. We constructed a Spo20/SNAP25 chimera in which SNARE(C) was replaced with the human ortholog ( Figure 1a). Instead of GFP, a hemagglutinin (HA) tag was added to the N-terminus of the chimera since the small epitope rarely compromises protein functions. The fusion protein was termed HA-Spo20/SNAP25(C). HA-Spo20/SNAP25(C) was placed under the control of the sporulationspecific SPO20 promoter, and the expression plasmid was integrated into the genome of spo20∆ cells. While spo20∆ cells exhibited sporulation deficiency, the defect was recovered by expression of HA-Spo20/SNAP25(C) (Figure 1b and Figure S2). Compared to the spo20∆ mutant harboring HA-Spo20, those harboring HA-Spo20/SNAP25(C) took a longer time until the sporulation rate reached a plateau ( Figure 1b). Nevertheless, spo20∆ mutant cells harboring HA-Spo20 and HA-Spo20/SNAP25(C) formed spores at comparable levels when the cells were incubated in sporulation media for 48 h (Figure 1b). Thus, in this study, sporulation efficiencies were measured after 48 h of incubation in sporulation medium.

The Spo20/SNAP25(C) Chimera Is Targeted by BoNT/C-LC
To examine whether Spo20/SNAP25 chimeras were targeted by BoNT-LCs, BoNT/C-LC was expressed under the control of the constitutive TEF1 promoter in spo20∆ cells harboring HA-Spo20/SNAP25(C) (Figure 2a). In HA-Spo20-harboring cells, sporulation efficiency was not altered by the expression of BoNT/C-LC ( Figure 2b). However, in HA-Spo20/SNAP25(C)-harboring cells, sporulation efficiencies were decreased by 40% by expression of BoNT/C-LC ( Figure 2a). To verify that the decrease in sporulation efficiency was attributable to the proteolytic activity of BoNT/C-LC, a catalytically inactive mutant, BoNT/C E230Q -LC [31], was expressed in HA-Spo20/SNAP25(C)-harboring cells ( Figure 2c). As shown in Figure 2c, sporulation efficiency was not decreased by expression of the catalytically inactive BoNT/C-LC mutant. It should be noted that expression levels of BoNT/C E230Q -LC were decreased by 83% compared to those of wildtype ( Figure S3a). To further verify that HA-Spo20/SNAP25(C) was cleaved by BoNT/C-LC, we performed Western blotting to detect the chimera. However, HA-Spo20/SNAP25(C) expressed from the plasmid integrated into the genome was not detected ( Figure S4a). Thus, Spo20/SNAP25(C) fused to a 3× HA tag at the C-terminus (Spo20/SNAP25(C)-3HA) was overexpressed from a multicopy vector in spo20∆ cells. As shown in Figure S4a, Spo20/SNAP25(C)-3HA was detected in sporulating cells. The expression levels of Spo20/SNAP25(C)-3HA in spo20∆ cells reached the highest value after 12 h of incubation in sporulation medium ( Figure S4b). In BoNT/C-LC-expressing cells but not in BoNT/C E230Q -LC-expressing cells, the protein levels of Spo20/SNAP25(C)-3HA were decreased, presumably due to digestion by the neurotoxin (Figure 2d). The levels of Spo20-3HA were not altered by the expression of BoNT/C-LC (Figure 2e).

Yeast Cells Harboring the Spo20/SNAP25 Chimera Can Be Used to Assay BoNT-LCs
Spores contain a fluorescent molecule, bisformyldityrosine, in the spore wall [32]. We found that the quantities of spores were correlated with the fluorescent levels of dityrosine ( Figure S5a). While spores were counted under a microscope in the experiments described above, the obtained result suggests that sporulation efficiency can be quantified more readily and unbiasedly by the colorimetric measurement. To test this possibility, the sporulation efficiencies of the cells described in Figure 2a Figure S5b). Thus, hereafter, the colorimetric assay was employed to measure sporulation efficiencies.
To further examine whether Spo20/SNAP25 chimeras can be used to analyze BoNT-LCs, BoNT/A-LC, which is another SNAP25 targeting BoNT-LC, was expressed in spo20∆ cells harboring HA-Spo20/SNAP25(C) from the TEF1 promoter (Figure 3a). The expression of BoNT/A-LC caused a 65% decrease in sporulation efficiency in the HA-Spo20/SNAP25(C)-harboring cells (Figure 3a). A severe sporulation defect was not observed when a catalytically inactive BoNT/A-LC mutant, BoNT/A E224Q -LC, was expressed in the HA-Spo20/SNAP25(C)-harboring cells, although its expression caused a slight decrease (by 15%) in sporulation efficiency (Figure 3a). Notably, expression levels of the wildtype and mutant BoNT/A-LC were comparable (Figure 3a and Figure S3b). Levels of Spo20/SNAP25(C)-3HA were decreased by expression of BoNT/A-LC but not by BoNT/A E224Q -LC (Figure 3b). In spo20∆ cells harboring HA-Spo20, sporulation efficiencies were not altered by the expression of BoNT/A-LC (Figure 3c). The expression of BoNT/A-LC did not affect the levels of Spo20-3HA detected in spo20∆ cells (Figure 3d).
further verify that HA-Spo20/SNAP25(C) was cleaved by BoNT/C-LC, we performed Western blotting to detect the chimera. However, HA-Spo20/SNAP25(C) expressed from the plasmid integrated into the genome was not detected ( Figure S4a). Thus, Spo20/SNAP25(C) fused to a 3× HA tag at the C-terminus (Spo20/SNAP25(C)-3HA) was overexpressed from a multicopy vector in spo20∆ cells. As shown in Figure S4a, Spo20/SNAP25(C)-3HA was detected in sporulating cells. The expression levels of Spo20/SNAP25(C)-3HA in spo20∆ cells reached the highest value after 12 h of incubation in sporulation medium ( Figure S4b). In BoNT/C-LC-expressing cells but not in BoNT/C E230Q -LC-expressing cells, the protein levels of Spo20/SNAP25(C)-3HA were decreased, presumably due to digestion by the neurotoxin (Figure 2d). The levels of Spo20-3HA were not altered by the expression of BoNT/C-LC (Figure 2e).  The intensity of the Spo20-3HA band detected in the cells harboring pRS424-TEFpr was defined as 1. Data are presented as the mean ± SEM. Statistical significance was determined by two-tailed unpaired Student's t-tests; n = 3 (a-c), n = 4 (d,e); **** p < 0.0001, ns: not significant (p ≥ 0.05).

Yeast Cells Harboring the Spo20/SNAP25 Chimera Can Be Used to Assay BoNT-LCs
Spores contain a fluorescent molecule, bisformyldityrosine, in the spore wall [32]. We found that the quantities of spores were correlated with the fluorescent levels of dityrosine ( Figure S5a). While spores were counted under a microscope in the experiments described above, the obtained result suggests that sporulation efficiency can be quantified more readily and unbiasedly by the colorimetric measurement. To test this possibility, the sporulation efficiencies of the cells described in Figure 2a,c were measured by colorimetric assay. Figure S5b shows that the microscopy and colorimetric assays exhibited similar results (compare Figures 2a,c and S5b). Thus, hereafter, the colorimetric assay was employed to measure sporulation efficiencies.
To further examine whether Spo20/SNAP25 chimeras can be used to analyze BoNT-LCs, BoNT/A-LC, which is another SNAP25 targeting BoNT-LC, was expressed in spo20∆ cells harboring HA-Spo20/SNAP25(C) from the TEF1 promoter (Figure 3a). The expression of BoNT/A-LC caused a 65% decrease in sporulation efficiency in the HA-Spo20/SNAP25(C)-harboring cells (Figure 3a). A severe sporulation defect was not observed when a catalytically inactive BoNT/A-LC mutant, BoNT/A E224Q -LC, was expressed in the HA-Spo20/SNAP25(C)-harboring cells, although its expression caused a slight decrease (by 15%) in sporulation efficiency (Figure 3a). Notably, expression levels of the wildtype and mutant BoNT/A-LC were comparable (Figures 3a and S3b). Levels of Spo20/SNAP25(C)-3HA were decreased by expression of BoNT/A-LC but not by BoNT/A E224Q -LC (Figure 3b). In spo20∆ cells harboring HA-Spo20, sporulation efficiencies were not altered by the expression of BoNT/A-LC (Figure 3c). The expression of BoNT/A-LC did not affect the levels of Spo20-3HA detected in spo20∆ cells (Figure 3d).

Use of Yeast Cells Harboring Spo20/SNAP25 Chimeras to Analyze BoNT/E Family Members
BoNT/E1 is another botulinum toxin that cleaves SNARE(C) in SNAP25 [18]. When BoNT/E1-LC was expressed in wildtype cells from the TEF1 promoter, we found that growth of the yeast cells became slower compared to those harboring the empty vector ( Figure S7a,b). Thus, BoNT/E1-LC was expressed from the SPO20 promoter so that the expression of BoNT-LC was restricted in sporulating cells (Figure 4a). The spo20∆ cells harboring SPO20 promoter-driven BoNT/E1-LC grew normally ( Figure S7b). Importantly, sporulation of spo20∆ cells harboring HA-Spo20 cells was not inhibited by expression of BoNT/E1-LC ( Figure S7c). However, in spo20∆ cells harboring HA-Spo20/SNAP25(C), sporulation efficiency was decreased by 67% by expression of BoNT/E1-LC (Figure 4b). Western blotting results showed that the levels of Spo20/SNAP25(C)-3HA were decreased by the expression of BoNT/E1-LC (Figure 4c). By expression of a catalytically inactive BoNT/E1-LC mutant, BoNT/E1 E213Q -LC, sporulation efficiency and levels of the chimeric SNARE were not altered (Figure 4c,d), although expression levels of BoNT/E1 E213Q -LC were decreased by 46% compared to those of wildtype ( Figure S3c). BoNT/E1-LC does not cleave SNARE(N) in SNAP25. Consistent with this specificity, sporulation of HA-Spo20/SNAP25(N)-harboring spo20∆ cells was not inhibited by expression of BoNT/E1-LC (Figure 4e).

Analysis of BoNT/En-LC, Which Cleaves SNARE(N) in SNAP25
BoNT/En is a BoNT-like protein found in Enterococcus spp. [4]. BoNT/En-LC is known to cleave SNARE(N) in SNAP25. Thus, to examine whether HA-Spo20/SNAP25(N) could be targeted by BoNT-LCs, we expressed BoNT/En-LC from the TEF1 promoter in wildtype cells. However, like BoNT/E1-LC and/E12-LC, the growth of the yeast cells also became slower than that of those harboring the empty vector ( Figure S7a,b). Thus, BoNT/En-LC was expressed from the SPO20 promoter, which did not interfere with the vegetative growth ( Figure S7b). In spo20∆ cells harboring HA-Spo20, sporulation was not compromised by the SPO20 promoter-driven BoNT/En-LC ( Figure S7c). However, in spo20∆ cells harboring HA-Spo20/SNAP25(N), sporulation efficiency was decreased by expression of BoNT/En-LC by 81% (Figure 5a). The sporulation efficiency of the HA-Spo20/SNAP25(C)-harboring cells was not decreased by BoNT/En-LC expression ( Figure 5b). These results are consistent with the notion that BoNT/En-LC targets SNAP25(N) but not SNAP25(C). Additionally, the levels of Spo20/SNAP25(N)-3HA, but not Spo20/SNAP25(C)-3HA, were decreased by the expression of BoNT/En-LC (Figure 5c,d).
We constructed a putative catalytically inactive BoNT/En-LC mutant by mutating the conserved Gln226 residue to Gln. Expression of BoNT/En E226Q -LC caused a slight decrease in sporulation efficiency (by 10.4%) in the HA-Spo20/SNAP25(C)-harboring cells; however, the effect was relatively mild compared to that of BoNT/En-LC expression (Figure 5a). Protein levels of Spo20/SNAP25(N) were not altered by BoNT/En E226Q -LC expression (Figure 5c). Expression levels of BoNT/En E226Q -LC were decreased by 68% compared to those of wildtype ( Figure S3e). because vesicle fusion at the plasma membrane requires more energy to generate the prospore membrane [34]. Thus, the SNARE domain in Sec9 is unable to be replaced with that of SNAP25, presumably because the SNARE complex containing a SNAP25 SNARE domain is not sufficient to induce vesicle fusion at the plasma membrane. Regulation of synaptic SNARE proteins is more complex compared to that of yeast exocytic SNARE proteins. The previous study using Sso1/syntaxin 1A revealed that syntaxin-1A and Sso1 exhibit distinctive properties [35]. Thus, further analysis of yeast/human chimeric SNAREs, including Spo20/SNAP25, would provide further insight into the regulatory mechanism of synaptic SNARE proteins. We examined various BoNT-LCs in our assay system; generally, the results are consistent with the previously reported specificities of BoNT-LCs. Of note, the expression of catalytically inactive BoNT/A E224Q -LC and/En E226Q -LC caused a slight decrease in spo20∆ cells harboring Spo20/SNAP25 chimeras. Levels of Spo20/SNAP25 chimeras were not altered by their expression. While the reason why the mutant BoNT-LCs exhibit the inhibitory effect is not clear, a similar phenomenon was reported in vivo; mice treated with a high dose of a catalytically inactive BoNT/C exhibited neuromuscular impairment [36]. Thus, BoNT-LCs may be able to compromise SNARE function even without cleaving them; for example, their binding to the target may inhibit SNARE assembly. Through mutational analysis, we showed that the catalytic activity is required to exhibit toxicity in yeast cells for BoNT/A-LC. However, for other BoNT-LCs, expression levels of catalytically inactive mutants were lower compared to those of wildtype ones. Nevertheless, our results overall showed that the BoNT-LCs target appropriate chimeras in yeast cells.
Expression of BoNT/En-LC from a constitutive promoter causes growth defects in yeast cells. Since BoNT/En-LC is known to target multiple SNAREs, including SNAP25 and VAMP2 [4], yeast SNARE(s) may be targeted by the BoNT-LC, which is a probable reason for the growth defect. BoNT/E12-LC is highly identical to BoNT/E1-LC, and our results showed that their LCs have a similar substrate specificity to that of BoNT/E1-LC. Like BoNT/En-LC, the expression of BoNT/E1 and/E12-LCs induces growth defects in yeast cells. Thus, while BoNT/E1-LC is known to cleave only SNAP25 in mammalian cells [37], BoNT/E family members may target SNAREs in other organisms. Notably, when these BoNT-LCs were expressed from the SPO20 promoter, cells harboring wildtype Spo20 showed neither growth nor sporulation defects. Thus, the expression of BoNT-LCs from SPO20 or other sporulation-specific promoters is likely a good option to improve the assay system. Sso1/Syntain 1A and Snc2/VAMP2 chimeras were previously produced [25,26]. However, in the Snc2/VAMP2 chimera, only a part of the VAMP2 SNARE domain was introduced in the yeast ortholog [25]. Thus, if a functional Snc2/VAMP2 chimera containing the entire VAMP2 SNARE domain is constructed, a yeast cell-based comprehensive BoNT-LC assay system could be established. Recent advances in genome mining have revealed novel BoNTs and BoNT-like genes. The yeast cell-based assay system will be useful for characterizing novel neurotoxins. Furthermore, the assay system could be used to screen BoNT-LCs with modified activities.