Alternate response of Bacillus anthracis atxA and acpA to serum, HCO3- and CO2

Bacillus anthracis overcome host immune response by producing capsule and secreting toxins. Production of these virulence factors in response to host environment was shown to be regulated by atxA, the major virulence regulator, that was shown to be activated by HCO3− and CO2. While toxin production is regulated directly by atxA, capsule production is mediated by either one of two regulators; acpA and acpB. In addition, it was demonstrated that acpA gene have at least two promotors were one of them was linked to atxA. We used a genetic approach to study capsule and toxin production under different conditions. Unlike previous works that used NBY-HCO3− medium in a CO2 enriched atmosphere, we used a sDMEM based medium in which toxins and capsule production can be induced in B. anthracis in ambient or CO2 enriched atmosphere. Using this system, we could differentiate between induction by 10% NRS, 10% CO2 or 0.75% HCO3−. In response to high CO2, capsule production is induced by acpA in an atxA independent manner, while little or no toxins (protective antigen PA) is produced. atxA is activated in response to serum in CO2 independent manner and induce toxins production and capsule production in an acpA or acpB dependent manner. HCO3− activates atxA as well, however in non-physiological concentrations. Our findings might be relevant to the first stages of inhalational infection were in the dendritic cell the germinating spore must be protected by the capsule without effecting cell migration to the draining lymph-node by toxin secretion.


Introduction 46
Bacillus anthracis, the causative agent of anthrax, is a gram positive, spore forming bacterium that infect humans via three major routes; lung (inhalation), cutaneous and the gut (digestion) [1,2]. The infectious Protective antigen (PA) concentration was determine by capture ELISA using the combination of a 111 polyclonal and monoclonal PA antibodies as previously described [23] 112

129
The effect of supplements and growth condition on capsule production in sDMEM medium. production could be detected in 100l sDMEM (Figure 1) but increasing the culture volume to 300l 136 results in capsule production were in 200l only part of the bacteria were encapsulated (Figure 1). This 137 effect could result from dissolved CO 2 generated by bacterial growth and limited gas exchange in the 138 case of low surface to liquid volume. This is supported by the finding that growth in 10% CO 2 atmosphere 139 induces capsule accumulation even in medium volume of 100l sDMEM (Figure 1)

150
The role of atxA on capsule production and toxin secretion under different growth conditions. To 151 test the role of atxA on capsule and toxins production following growth in sDMEM under the different 152 growth conditions we compared the wild type Vollum strain to the pXO1 or atxA mutants. None of the 153 strains produced capsule under ambient atmosphere in the un-supplemented sDMEM medium (Figure 2).

154
Addition of 10% NRS resulted in capsule production by the wild type Vollum strain but not by the atxA 155 null mutants (pXO1 or atxA Figure 2). In a 10% CO 2 atmosphere all the strains were encapsulated 156 regardless of the presence or absence of atxA. Hence, the effect of NRS in the presence of CO 2 on 157 capsule production under these conditions could not be resolved since capsule production is induced in 158 response to the presence of CO 2 or NRS. Since it appeared that the response to NRS is atxA dependent 159 we can assume that capsule production in the atxA null mutants derived form only the CO 2 (Figure 2).
160 161 Figure 2. Capsule production of the Vollum wild type, pXO1 or xtXA mutants under different growth 162 conditions.  conditions. Growth of the Vollum strain in ambient condition did not result in any capsule accumulation or 220 toxin secretion following 24h incubation (Figure 2, Table 1) or a short 5h incubation (Figure 5, Table 3).

221
Supplementing the media with 10% NRS induced capsule production and PA secretion flowing 24h 222 incubation in ambient atmosphere (Figure 2, Table 1). Examination these parameters following a shorter, 223 5h incubation, demonstrating PA accumulation ( Table 3) but no capsule accumulation ( Figure 5). This PA 224 accumulation is AtxA dependent as deletion of the atxA gene resulted in no PA accumulation following 5h 225 ( Table 3) or 24h incubation ( Table 1). Incubating the bacteria in 10% CO 2 atmosphere for 5h result in 226 capsule accumulation ( Figure 5) the same as was observed following a 24h incubation (Figures 1-3). This 227 capsule accumulation was atxA independent and was not dramatically affected by the addition of 10%NRS 228 or HCO 3 - (Figure 5). PA secretion was not induced by 10%CO 2 per se buy required addition of NRS or 229 HCO 3 - (Table 3), however unlike the 24h incubation ( Regulation of acpA and acpB in response to different growth conditions. As toxin production is induced 245 in an atxA dependent manner in response to HCO 3 -or NRS, capsule production is also induced by CO 2 246 enriched (10%) atmosphere. Since capsule production is regulated by two regulatory proteins, AcpA and  (Figure 7).

262
As we demonstrated (Figure 2), capsule accumulation could be induced in ambient atmosphere by adding 263 10% NRS to the growth media. This induction is AtxA dependent since no capsule accumulation was 264 detected in the VollumatxA mutant under these conditions (Figure 2). Since AcpA dependent capsule 265 accumulation in 10% CO 2 atmosphere was AtxA independent, we tested the role of AtxA on AcpA 266 dependent capsule accumulation in response to 10% NRS in ambient atmosphere. As 10% NRS induces 267 capsule accumulation of VollumacpB in ambient atmosphere (Figure 6, Figure 8) we tested the effect 268 of atxA deletion on capsule accumulation under these conditions. Unlike the CO 2 induction, in ambient 269 atmosphere, AcpA dependent capsule accumulation in response to NRS is AtxA dependent (Figure 8). For successful invasion the pathogen mast regulates its virulence factors in a way that will maximize their 291 effect on the host defense mechanisms. The trigger of such activation is usually host derived and can be 292 biological (such as proteins) of physical (pH or temperature for example). B. anthracis naturally infect 293 humans following spore inhalation, contact with broken skin or ingestion of undercook contaminated meat, 294 in each of them the bacteria faces different environmental conditions [2]. It was previously demonstrated 295 that toxins secretion and capsule accumulation could be induced by growing the bacteria in culture media 296 supplemented with HCO 3 -or serum (10-50%) in a CO 2 enriched (5-15%) atmosphere [5,16,18,19,[26][27][28].
297 HCO 3 -/CO 2 condition were used to study in most studies of atxA, acpA and acpB regulation and their 298 effect on toxins and capsule biosynthetic genes [16,18,29]. Since these conditions always included these 299 two components it was concluded that atxA was induced in response to CO 2 and regulate the induction of 300 acpA and acpB. Although in some reports, capsule accumulation was shown to be atxA dependent [16,30], 301 the fact that pXO1 variants are encapsulated contradict this finding, indicating addition atxA 302 independent regulation of the process [19,26]. The use of sDMEM as a growth media enabled the 303 examination of the effect of CO 2 , HCO 3 -and serum on these processes.

304
The parameter of soluble CO 2 is influenced by multiple parameters such as surface area to volume ratio 305 and aerobic bacterial growth. Therefore, normal growth conditions were determined as 100l media/well 306 of a 96 well tissue culture plate for 24h at 37 o C in ambient condition (Figure 1). This baseline enabled 307 testing of the different supplements and/or growth conditions on capsule (Figure 1) or toxin (Figure 2, 308 Table 1). Capsule production is induced by the addition of 10% NRS or growth in a 10% CO 2 atmosphere 309 (Figure 1). The serum capsule induction (in ambient atmosphere) is atxA dependent (Figure 2) since there 310 was no significant capsule accumulation it the mutants that did not express AtxA (VollumatxA and 311 VollumpXO1). Alternatively, capsule accumulation in response to CO 2 enriched atmosphere is atxA 312 independent, as there is no significant difference in capsule accumulation under these conditions between 313 AtxA expressing and atxA null mutants (Figures 2, 3). Toxins secretion, as determined by PA media 314 concentration is serum dependent (Tables 1, 2), as PA could be detected only in NRS supplemented 315 sDMEM regardless of the CO 2 enriched atmosphere. HCO 3 -as serum induced toxin secretion in an atxA 316 dependent manner (Figure 3,  (Table 3). Hence, Serum seemed more 329 effective then HCO 3 -in inducing toxin secretion, two process that are atxA dependent. The atxA 330 independent 10% CO 2 atmosphere appeared more effective then atxA dependent serum for capsule 331 accumulation ( Figure 5).

332
Two major regulators; AcpA and AcpB controle capsule biosynthetic by promoting transcription of 333 acpB,C,A,D,E operon. acpA was shown to be regulated by atxA (assumed by CO 2 and HCO 3 -) and to possess 334 an additional atxA independent promotor. We found that by deleting acpA the mutant accumulates 335 significantly less capsule in response to CO 2 but maintain its ability to respond to NRS or HCO 3 -( Figure   336 6). Deletion of acpB did not have any effect on capsule accumulation under all tested conditions ( Figure   337 6) which support our previous in vivo data. Deletion of both atxA and acpA or acpB revealed that AcpB 338 activity is strictly AtxA dependent under all the conditions tested (Figure7). AcpA activity is not 339 affected by the absence of atxA in the presence of CO 2 , (Figure7) but is nulled in response to NRS in an 340 ambient environment (Figure 8).

341
Our findings support the following regulation cascade; CO 2 is inducing capsule accumulation via the 342 activation of acpA in an AtxA independent manner. Serum activate the AtxA dependent cascade that 343 induce toxins secretion and eventually capsule accumulation by activating acpA and acpB (Figure 9) since 344 there was no capsule accumulation following 5h growth in NRS containing in ambient atmosphere ( Figure   345  5). HCO 3 -induces toxin secretion by AtxA cascade, in a less efficient manner (relative to NRS, Table 3).

346
Direct activation of capsule accumulation by HCO 3 -in AtxA independent manner could not be eliminate 347 since even in ambient atmosphere, it modifies the levels of soluble CO 2 (PCO 2 ) and possibly induces capsule 348 production via acpA. Theoretically, this differential regulation of toxins and capsule have logical role in 349 B. anthracis pathogenicity. Inhalational and cutaneous infection involve spore engulfing and migration to 350 a draining lymph node. Within the phagocytic (dendritic) cell the spore germinates and produce capsule 351 that protect the bacteria form pagolysis. Toxin production in this stage in counterproductive as it could 352 result in cell arrest that might interfere with the pathogenic pathway. Toxin production may enhance 353 once the bacteria is released from the cell in the lymph node. The serum and CO 2 sensing pathway is still 354 to be determined and might be common to other pathogens and therefore be a target for therapeutics.
355 Figure 9. Proposed regulation of CO 2 , Serum and HCO 3 -on capsule production and toxin secretion.