Identification and Characterization of a Spore-Like Morphotype in Chronically Starved Mycobacterium avium Subsp. Paratuberculosis Cultures

Mycobacteria are able to enter into a state of non-replication or dormancy, which may result in their chronic persistence in soil, aquatic environments, and permissive hosts. Stresses such as nutrient deprivation and hypoxia provide environmental cues to enter a persistent state; however, a clear definition of the mechanism that mycobacteria employ to achieve this remains elusive. While the concept of sporulation in mycobacteria is not novel, it continues to spark controversy and challenges our perceptions of a non-replication. We investigated the potential role of sporulation in one-year old broth cultures of Mycobacterium subsp. paratuberculosis (MAP). We show that dormant cultures of MAP contain a mix of vegetative cells and a previously unknown morphotype resembling a spore. These spore-like structures can be enriched for using sporulating media. Furthermore, purified MAP spore forms survive exposure to heat, lysozyme and proteinase K. Heat-treated spores are positive for MAP 16SrRNA and IS900. MAP spores display enhanced infectivity as well as maintain acid-fast characteristics upon germination in a well-established bovine macrophage model. This is the first study to demonstrate a new MAP morphotype possessing spore-like qualities. Data suggest that sporulation may be a viable mechanism by which MAP accomplishes persistence in the host and/or environment. Thus, our current understanding of mycobacterial persistence, pathogenesis, epidemiology and rational drug and vaccine design may need to be reevaluated.


Introduction
Mycobacteria represent a group of highly successful organisms that range from free-living saprophytes to those that have adapted full dependence on a living host [1,2]. During their life cycle, mycobacterial species may encounter a number of stresses including nutrient deprivation, hypoxia, acidic pH, and even competition with other organisms for limited resources and occupation of a specific niche, such as soil and water [3,4,5,6]. In order to survive in such unfavorable conditions, mycobacteria have developed mechanisms to achieve dormancy, latency and persistence [7,8,9]. While several studies have investigated persistence in mycobacteria, the definition remains loosely explained and the mechanisms that lead to and sustain this state of non-replication are poorly understood. A recent study by Ghosh et al. stated the formation of endospores in two month old cultures of M. marinum and M. bovis, which may serve as an unprecedented method employed by mycobacteria to withstand harsh conditions [10]. The concept of sporulation in mycobacteria continues to spark controversy and challenges our current perceptions of the facets involved in mycobacterial persistence. Follow-up studies conducted by Traag et al. could not reproduce endospore formation in 4 week to ,8.5 month liquid cultures of M. marinum, which questioned the purity of cultures used in ultrastructural characterization by Ghosh et al. [11]. The current research trend focusing on sporulation in mycobacteria remains to reproduce findings by Ghosh et al. using identical isolation methods; however, we investigated whether the potential for sporulation was limited to M. marinum or may encompass another saprophyte and animal pathogen, Mycobacterium avium subsp. paratuberculosis (MAP).
MAP, the causative agent of Johne's disease (JD) in ruminants, is one of the most prevalent and well-documented infections of dairy cattle worldwide [12]. To date, JD eradication remains implausible and control efforts are hampered due to MAP's persistence within soil and water as well as shedding by subclinical and clinical cattle [12,13,14,15]. Therefore, it is critical to augment our knowledge of the events that precede non-replication as well as the various mechanisms used to attain it. Our data showed that one year old broth cultures of MAP strains K-10, 7565, Linda and Ben have the potential to produce a previously undocumented morphotype possessing a spore-like structure given optimal sporulation conditions. All isolated MAP spore-like morphotypes shared appropriate spore ultrastructures, presence of dipicolinic acid and the MAP specific insertion sequence, IS900, and heat resistance. More importantly, heat treated MAP spore structures retained macrophage infectivity as well as acid-fast characteristics upon germination. These data suggest that sporulation may be a viable route by which MAP accomplishes persistence in the environment.

Results and Discussion
MAP produces a new morphology when grown on Arret-Kirshbaum sporulation agar under physiologically relevant temperature One year old Middlebrook 7H9 (MB7H9) broth cultures (herein referred to as dormant cultures) of MAP strains K-10 (cattle isolate) and 7565 (sheep isolate) were used to examine sporulation activity and isolate spores. It is important to note that the culture medium was never changed and agitation was not supplied for ,6 months (mo.); therefore, MAP strains were assumed to be subject to nutrient starvation and hypoxia. These methods differ from those utilized by Ghosh et al. in which M. marinum and M. bovis were grown on 7H10 agar plates for 2 and 12 weeks, respectively [10]. Methods also varied from those described by Traag et al. since MAP broth cultures were kept at a constant 37uC as opposed to intermittent incubations at 4uC and 33uC. Furthermore, we included a spore enrichment process in which MAP broth cultures were separately plated on Arret-Kirshbaum (A-K) sporulating agar and grown at 37uC and 39uC. Since A-K sporulating agar is used for induction of sporulation, particularly from Bacillus spp., we hypothesized that MAP may also produce spores upon nutrient exhaustion in this medium. We showed that MAP is capable of growth and sporulation on A-K agar after 72 h of incubation at 39uC, which was validated by differential staining for spores using malachite green and safranin ( Figure 1A and B). These results are in stark contrast to E. coli K-12 A-K growth, which is negative for malachite green staining ( Figure 1B). It is interesting to note that successful MAP growth on A-K medium occurred at 39uC and not at 37uC, the physiological body temperature of cattle in contrast to the B. subtilis control which grew at 37uC and 39uC ( Figure 1A). Previous studies from our laboratory demonstrate that physiologically relevant temperatures greatly influence MAP gene expression profiles and speed of macrophage invasion [16]. It is well recognized that mycobacteria are sensitive to changes in temperature, which influence growth, cell morphology and pathogenesis [16,17,18,19]. Temperature also impacts sporulation efficiency [20,21,22]. As previously mentioned, dormant cultures of M. marinum used by Traag et al. were stored at 4uC for an unspecified period of time, which may provide one explanation to the lack of spore detection [11]. Thus, additional pressures like host related temperature may be one of several contributing stressors capable of inducing differential rates of sporulation.
Due to the age of MAP cultures, a valid concern arose that isolated morphotypes were not of a MAP origin but a known endospore contaminant. In order to confirm MAP purity, dormant MAP K-10 cultures used for spore enrichment were streaked on Brain-Heart Infusion (BHI) agar. While B. subtilis produced visible colonies, MAP K-10 showed no growth and was determined to be Figure 1. MAP morphotype induction is dependent upon temperature. One year old MB7H9 MAP broth cultures were inoculated on A) A-K agar and C) BHI agar for 72 h at 37uC and 39uC. MAP showed growth only at 39uC compared with B. subtilis and E. coli K-12 controls, which had substantial growth at both temperatures (A). Spore enrichment was determined by malachite green staining (B). In order to confirm purity of MAP culture, the year old MAP culture was grown on BHI agar and determined to be free of any contaminating organisms (C). doi:10.1371/journal.pone.0030648.g001 devoid of any contaminating organisms ( Figure 1C). Also, MAP K-10 was revived on MB7H9 agar indicating that dormant cultures remained viable. Therefore, MAP K-10 has the potential to produce spore-like structures given sufficient and appropriate cues. Environmental cues that are involve nutrient and moisture limitation, temperature, hypoxia, and competing microbes are hypothesized to be sufficient to cause mycobacterial sporulation but may not all be necessary at the same time.
Transmission Electron Microscopy (TEM) shows presence of spore-like forms Since malachite green primarily functions by permeating the mother cell through heat and is retained within the cell coat layers due to their thickness, the possibility remained that malachite green may be bound to the complex, waxy mycobacterial cell wall. In fact, malachite green has been used to stain flagella and leprosy bacilli in addition to endospores [23,24]. To validate A-K agar and malachite green staining results and characterize the MAP spore-like structure, we examined the spore forms during different phases of growth: log, dormant, and induced (A-K isolated) spore phases ( Figure 2A). Dormant MAP K-10 cultures displayed a mix of bacilli and spores, while the induced phase contained only spores ( Figure 2A). Next, we increased the stringency of our testing for possible contamination and conducted duplex and normal polymerase chain reactions (PCRs) to demonstrate the presence and/or absence of IS900 integration sites in MAP, Bacillus spoIVA gene and Clostridium 16SrDNA gene in all MAP and control samples ( Figure 2B). MAP dormant, log phase, germinated and A-K isolated spore cultures showed the presence of IS900 integration sites but did not display amplification of specific Bacillus and Clostridium gene elements, which suggested the unlikelihood of isolated MAP morphotypes containing Bacillus and Clostridium species ( Figure 2B). As expected, control spore samples from B. subtilis and C. perfringens did not amplify the MAP specific IS900 integration sites ( Figure 2B).
We tested if MAP dormant and A-K spore cultures represented true spores and not a simple thickening of the cell wall as has been reported by Cunningham et al. by determining the presence of dipicolinic acid (DPA; pyridine-2,6 dicarboxylic acid), a chemical commonly found in spores [25]. Calcium bound DPA is found within the spore core and contributes to spore DNA resistance to wet and dry heat, desiccation and hydrogen peroxide [26,27,28]. In order to detect DPA, we used a colorimetric assay that releases DPA after autoclaving and acetic acid treatment and is subsequently available to bind to iron under an acidic pH [29]. Since mycobacteria secrete, contain, and are supplemented with siderophores, which form complexes with iron, we included a mycobactin J control to ensure abrogation of potential reactions with iron involved in DPA detection ( Figure 2C) [30]. Elevated levels of DPA were found in MAP dormant and spore cultures and B. subtilis spores (matched for wet pellet weight) ( Figure 2C). Not surprisingly, isolated MAP spores contained increased amounts of DPA (470.0 mg/mL) compared to any other MAP culture, which was not due to reactive mycobactin J ( Figure 2C). It has been noted that DPA synthesis in addition to endospore formation occurs with low guanine (G) and cytosine (C) % content among Firmicutes, which conflicts with the high percentage of G+C found in mycobacteria [11]. Although rare, members of the Streptomyces (S. globisporus and S. avermitilis) are capable of endospore production given conditions conducive to endsporulation such as incubation in submerged cultures and phosphate limitation [31,32]. Also, the presence of DPA was found by Stastná et al. despite the lack of the spoVF locus, which encodes a DPA synthetase [32]. Orsburn et al. has described that the spoVF locus is also absent in certain clostridia and that an electron transfer flavoprotein may compensate for DPA generation [33]. Therefore, spoVF may not be a strict requirement for DPA production and endosporulation may infrequently encompass genera from Actionbacteria and may not be restricted solely to low G+C% bacteria. Future studies should involve mechanistic based experiments to determine if DPA production in MAP depends upon an electron transfer flavoprotein. It is possible that novel morphotypes of MAP may not be strict endspores and are more reminiscent of Streptomyces spp. spore structures.

Sporulation in MAP is reproducible and consistent in a different culture, medium and multiple strains
Our enrichment technique using A-K agar was validated by another laboratory (National Animal Disease Center (NADC), USDA) using a separate MB7H9 liquid culture of MAP K-10 ( Figure 3A). TEM images of MAP K-10 showed characteristic spore features (e.g. condensed core and coat layers) ( Figure 3A). Since A-K agar is not a typical method used to induce sporulation from bacterial cultures, we included spore enrichment using potato extract agar (PEA) using a described method ( Figure 3B) [34].
Again, year old liquid culture of MAP K-10 was capable of producing spore-like morphotypes and displayed identifiable spore structures ( Figure 3B). B. subtilis and E. coli K-12 were included as positive and negative controls, respectively ( Figure 3B). MAP K-10 PEA spore-like structures were free of contamination as determined by the absence of growth on blood agar (Figure 3d). Furthermore, we sought to determine if other MAP strains were capable of sporulation. A-K enrichment of and TEM visualization showed that MAP 7565 (sheep strain), Ben (human strain) and Linda (human strain) sporulated ( Figure 4). Both Ben and Linda strains were isolated from patients with Crohn's disease [35,36]. Although controversial, the presence of MAP (either in blood or intestinal tissue) has been shown to be associated with Crohn's disease but an etiological link between the two remains to be established [37,38,39].
MAP spore-like forms survive heat treatment and are positive for MAP 16SrRNA and IS900 To be considered a true spore former, isolated novel MAP morphotypes must survive exposure to heat and be capable to germinate given a nutrient source [26]. We heat treated MAP K-10 spores at 70uC and 90uC for 30 min in addition to 2% lysozyme, proteinase K (PK), kanamycin and anaerobic exposure. Heat treatment served two purposes to: 1) determine temperature threshold for survival and 2) eliminate any remaining vegetative cells such that the re-grown culture only originated from spores. Both lysozyme and PK are typically used as a standard DNA extraction protocol that functions by damaging the cell wall of vegetative MAP cells, which causes bacterial lysis. MAP K-10 spores survived exposure to 70uC but not at 90uC (Figure 5a). Heat exposed spores treated in combination with either lysozyme, PK or kanamycin were capable of re-growth due to the coat layer resistance to these enzymes (Figure 5a). Incubation under anaerobic conditions was included to rule out Clostridium spp. contamination. Exposure to anaerobic conditions post heat treatment failed to produce any visible growth (Figure 5a). Contamination during heat treatment was also eliminated as can be observed by the absence of growth on blood agar plates (Figure 5b). Ten colonies from each heat treated MAP plate (70uC alone/and+lysozyme, +PK, +kanamycin) were selected and submitted to 16SrRNA sequencing. Sequencing results showed that all colonies were positive for MAP 16SrRNA (Figure 5c). These colonies were also positive for MAP-specific IS900 (Figure 5d). MAP spore survival post exposure to 70uC may in fact not be extremely surprising since many studies have shown the presence of MAP as a food contaminant in pasteurized (also treated at 70uC) milk, cheeses and yogurt [40,41,42,43]. It is currently unknown if the MAP found in these dairy products may exist in a spore or spore-like state. It is important to note that MAP is hypothesized to be one potential trigger or causative agent for Crohn's disease (CD) onset [37,38,39,44]. Several studies indicate that the gross pathology of JD and CD are similar, such as the thickened intestinal mucosa and transmural inflammation [36,45,46]. It is proposed that MAP survival in pasteurized dairy products may serve as a vehicle for MAP infection in a subset of CD patients [41,47,48]. If MAP does survive pasteurization as a spore, this may result in an important finding and further understanding of MAP's potential role in public health.

Dormant MAP cultures express elevated transcript levels of sporulation related genes
In addition to spore visualization, we have identified a number of mycobacterial candidate genes corresponding to those in the sporulation pathway of several Bacillus and Streptomyces species (Figure 6a). We show that MAP 1002c, which has a 57% similarity to the sporulation response regulator spo0A in Bacillus spp., has a 40-fold increase expression in dormant MAP K-10 compared to respective log-phase culture (Figure 6b). Other studies have also noted the presence of spore related genes in M. tuberculosis and other mycobacterial species [49].
Research investigating Bacillus spp. and S. coelicolor A3(2) morphology indicate that the stringent response is essential to robust spore production [50,51]. A recent study by Stallings et al. showed that the mycobacterial gene, carD, is an essential regulator of the stringent response that is also found in a number of sporulating bacteria that downregulates rRNA by binding onto the b-subunit of RNA polymerase (RNAP) in response to nutrient starvation and oxidative stress (Figure 6c) [52]. Transcript levels in dormant MAP culture show a 15 and 2-fold upregulation of MAP 0475 and MAP 0987(carD orthologues), respectively (Fig. 6b). The presence and absence of orthologous genes alone is unlikely to shed any deeper understanding of the sporulation process in mycobacteria but will require the addition of an evolutionary systems biology approach [53]. In other words, functional assays on identified genes are necessary to assess their impact on the generation of this new MAP morphotype. Evolutionary systems biology, which focuses on the changing relationships between genes and gene products, will reveal developmental networks that may include regulatory molecules and feed-forward networks. Ongoing studies in our laboratory are seeking to create knock-out and knock-in mutations of genes identified in Table 1 to determine what role if any they may have in formation of the MAP spore-like morphotype.

MAP spores retain infectivity and germinate into acid-fast bacilli during exposure to bovine monocyte derived macrophages (MDMs)
Previous studies have reported that MAP can readily be isolated in soil and aquatic environments, which may come into contact with animals and serve as transmission routes [15,54,55,56,57,58]. For example, livestock manure stored in liquid lagoons is often applied to agricultural land and MAP may persist within this environment upwards of 175 days [59]. Therefore, we asked the question if MAP spore-like forms could transmit and maintain  (Figure 7). Spore germination into developed bacilli show strong acid-fast staining, which is a major diagnostic feature of mycobacteria. Furthermore, the progression of infection with MAP spores was enhanced when compared to the MAP log-phase control and MDMs lysed at 48 h p.i. (Figure 7). These data combined with heat resistance suggest that sporulation in MAP may aid and impact the rate of transmission and consequently establishment of infection in host species.

Conclusions
This is the first study to identify and describe a new spore-like morphotype in MAP associated with nutrient starvation. We propose that MAP may utilize sporulation as a mechanism to persist in unfavorable conditions such as those encountered in soil and aquatic environments. MAP may also commit itself to a sporelike state to survive the pressures applied by pasteurization and thereby provide one explanation for MAP detection in commercialized dairy products. Although significant strides have been made, especially within the last ten years, in understanding mycobacterial persistence, it continues to be fraught with ambiguities and dissension. The findings by Ghosh et al., which identified spores in M. marinum, and those presented in this paper for MAP are difficult to grapple with as they defy key concepts and change our perceptions of persistence, dormancy and transmission for MAP. This new MAP morphotype or spore readily invaded bovine MDMs, germinated and developed into acid-fast bacilli. More importantly, enrichment and isolation of this new morphotype was independently conducted by a second laboratory (NADC) using a separate MAP culture grown at that facility. Concerns are raised due to the similarities of spore formation in MAP and Bacillus given widely divergent genera. However, 1) certain species of Streptomyces, another genus of the Actinobacteria,  are capable of endospore formation at suitable conditions and 2) DPA is also present in Streptomyces despite the lack of spoVF operon. Identification of unique MAP spore coat proteins as well as the cues leading to sporulation may aid in future diagnostics for food and environmental safety. Further studies are needed to examine the role of this newly described MAP morphotype in soil and aquatic environments as well as post pasteurization in dairy products utilizing the above aspects to assess impact in transmission and persistence.

Bacterial and MDM Cell Culture
Dormant cultures of MAP strains K-10, 7565, Ben and Linda were grown at 37uC in MB7H9 broth supplemented with oleic acid, albumin, dextrose and catalase (OADC) enrichment and mycobactin J (2.0 mg/L) (Allied Monitor, Fayette, MO) for one year without agitation. MAP cultures were determined to be free of contamination by absence of growth on BHI agar at 37uC and 39uC. Bacillus subtilis and Escherichia coli K-12 were cultured in Luria-Bertani (LB) broth at 37uC with shaking at 150 rpm. Clostridium perfringes was grown on blood agar at 37uC in an anaerobic chamber. Mitsubishi anaeropacks (Fisher Scientific, Pittsburgh, PA) used for the anaerobic chamber were changed every 24 h.
Peripheral blood was collected from the jugular veins of two JD free cattle (542 and 2170) at the University of Minnesota's dairy barn and teaching facility. Monocyte derived macrophages (MDMs) were elutriated using an established protocol [60]. All cattle work performed was in concordance with the institutional guidelines and approved animal care and use protocols at the University of Minnesota. A time point of 72 h for MAP cultures were selected based on the time required to observe MAP growth on A-K plates. MAP cultures were also grown on Potato Extract Agar (PEA) at 39uC until growth was achieved (2 weeks) [34]. Upon completion of incubation times, A-K and PEA agar plates were allowed to rest at

DNA Extraction and Polymerase Chain Reaction (PCR)
DNA was extracted from 500.0 mg (wet weight) MAP K-10 logphase (O.D. 600 = 0.5), dormant cultures and spores by incubating samples for 10 min at 37uC with 10% sodium dodecyl sulfate (SDS) followed by homogenization in a mini-bead beater (Roche) using 0.2 mL of 0.1 mm sterile RNase free zirconium beads (Biospec, Bartlesville, OK) for 4 min. DNA samples were cleaned using PE buffer and spin columns (Qiagen, Valencia, CA). B. subtilis DNA was isolated from subcultures grown for 16 h at 37uC using the QIAamp DNA mini kit (Qiagen, Valencia, CA) per manufacturer's instructions. Clostridium perfringens DNA was generously provided by Arpita Ghosh (University of Minnesota). All DNA samples were checked for purity and concentration using the 260/280 ratio provided by the NanoDrop sample retention system (Thermoscientific, Wilmington, DE). All primers with the exception of Clostridium 16SrDNA [61] were designed using Primer3 software (http://frodo.wi.mit.edu/primer3/) (

Transmission Electron Microscopy (TEM)
TEM ultrastructural imaging was conducted on MAP logphase, dormant and isolated spore cultures using the methods described by Ghosh et al. [10]. Dormant cultures and A-K and PEA biomasses of MAP strains K-10, 7565, Linda and Ben were centrifuged at 3,000 rpm and supernatants were discarded. Pellets were fixed with 2.5% glutaraldhyde in 0.1 M sodium cacodylate buffer overnight at 4uC. Samples were washed 36 using 0.1 M sodium cacodylate buffer and post fixed in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer reduced with ferroyanide and washed 36 in autoclaved water. After a series of acetone dehydration, all samples were infiltrated with 1:2, 2:1 resin:acetone mixtures and 3 100% resin. Samples were embedded and cured at 60uC for 48 h and later visualized using TEM.

DPA Assay
DPA was detected using a previously reported colorimetric assay [29]. Wet pellet weight of 250.0 mg was used for each bacterial sample. Autoclaved and ''intact'' mycobactin J (250 mg/ mL) were used as controls. DPA concentrations were calculated based on a DPA (MP Biomedicals, Riverside CA) protein concentration curve. DPA assay was performed six times and all  samples were conducted in triplicate with duplicate technical replicates. Figure 2C shows a representative experiment.

Heat Treatment
MAP K-10 spores, B. subtilis spores and C. perfringes spores were heat treated at 70uC and 90uC for 30 min. Additional treatments post heating included exposure to 2% lysozyme at 37uC for 10 min, PK (100 mL of 2 mg/mL) digestion at 37uC for 10 min, kanamycin (50 mg/mL) at 37uC for 2 h and anaerobic exposure at 37uC for 3 weeks. To ensure that MAP cultures were devoid of contaminants post heat treatment, all samples were grown on blood agar at 37uC for 4 weeks. Ten colonies were selected from each MAP heat treatment plate and re-grown in MB7H9 broth at 37uC for 10 weeks. DNA was extracted from re-grown cultures and sequenced for 16SrRNA ( Table 1). The universal 16SrRNA cycling program is as follows: 94uC for 15 min., 94uC for 15 s, 54uC for 15 s, 72uC for 30 s and 72uC for 1 min for 30 cycles. 16SrRNA sequences were analyzed using Sequencher (Gene Codes Corporation, Ann Arbor, MI) and aligned using MEGA software (http://www.megasoftware.net/) [62]. Colonies were further submitted to IS900 PCR. All samples were plated twice and heat treatment experiments were conducted three separate times.

RNA Extraction and Qt-RT-PCR
1.0 mL of dormant and log-phase MAP K-10 cultures were centrifuged separately in 1.5 mL eppendorf tubes for 10 min. at 13,000 rpm. Supernatants were decanted and pellets were washed 36 in 16 PBS. 1.0 mL of TRIzol reagent (Invitrogen, Carlsbad, CA) was added to each sample and allowed to incubate at room temperature for 5 min. MAP K-10 samples mixed with TRIZol were homogenized using 0.3 ml of 0.1 mm sterile RNase-free zirconium beads for 4 min. in the MagNa Lyser system (Roche, Basel, Switzerland). RNA was extracted following TRIzol protocol. RNA was subsequently treated with TurboDNase (Ambion, Austin, TX) for 30 min at 37uC. All samples used had a 260/280 ratio of at least 1.9 as measured by NanoDrop sample retention system (Thermoscientific, Wilmington, DE). Primers used for Qt-RT-PCR were designed using Primer3 software (http://frodo.wi.mit.edu/primer3/) ( Table 1). Qt-RT-PCR analysis was conducted on 50.0 mg of purified dormant or log-phase MAP K-10 culture combined with the Quantifast one-step RT-PCR reagents (Qiagen, Valencia, CA) using the Lightcycler 480 II (Roche, Basel, Switzerland) programmed for the following: 50uC for 10 min, 95uC for 5 min, 95uC for 10 s, 60uC for 30 s for 40 cycles. Fold change was calculated using the 2 2DDCt method. All samples were conducted in triplicate.

Spore Invasion Assay
Bovine MDMs in RPMI containing 2% autologous serum were seeded at 2.0610 4 cells/mL in a 24 well plate containing 1.0 No. 1.5 thickness glass coverslips and allowed to adhere for 2 h at 37uC in humidified incubator containing 5% CO 2 . Following incubation, MDMs were washed 36 in 16 Dulbecco's phosphate buffer saline (D-PBS) to remove non-adherent cells and medium was replaced with fresh serum-free RPMI prior to infection. MAP K-10 subculture and spores and B. subtilis spores were pelleted at 13,000 rpm for 5 min. or 10 min., respectively, washed 26 with warm 16 D-PBS and resuspended in 37uC warmed serum-free RPMI such that a MOI (multiplicity of infection) ratio of 10:1 was achieved. Serum free RPMI was used to prevent spore germination outside of MDMs. Spores were heat treated at 70uC for 30 min. Subsequently, cultures were vigorously vortexed and allowed to rest for 5 min. at 37uC so that potential clumps would settle to the bottom of the tube. An 18.5 gauge syringe needle was used to repeatedly draw the upper three-fourths of the newly suspended MAP cultures to disperse any remaining clumps. MDMs were separately infected with upper three-fourths of RPMI-MAP/spore cultures for 2 h at 37uC in humidified incubator containing 5% CO 2 , rinsed 36 with 16 D-PBS and resuspended in RPMI containing 2% autologous serum for the following post infection (p.i.) time points: 0, 0.5, 6, 2, 24, and 48 h. Upon completion of post infection time points, MDMs were rinsed 36 in D-PBS and fluorescently or acid-fast stained for visualization. All time points were conducted in triplicate.

Cell Staining
Fluorescent staining was conducted as stipulated by Lamont et al. [16] with the exception that log-phase and spore cultures were pre-stained for 30 min in 2.0 mg/mL of 5-carboxyfluorescein diacetate (CFDA) (Sigma-Aldrich, St. Louis, MO) at 37uC. Fluorescent images were visualized and collected as a Z-series (step size: 1.0 mm) using DAPI, FITC and Cy5 lasers on an Olympus Fluoview 1000 upright confocal microscope (Olympus, Center Valley, PA). Joint acid-fast images were stained using a modified Zeil-Neelsen protocol (Trend Laboratories Inc., Atlanta, GA) and imaged on an Olympus IX70 inverted fluorescence microscope.