Identification of a putative cell wall-hydrolyzing amidase involved in sporangiospore maturation in Actinoplanes missouriensis

ABSTRACT

Actinoplanes missouriensis is a filamentous bacterium that differentiates into terminal sporangia, each containing a few hundred spores. Previously, we reported that a cell wall-hydrolyzing N-acetylglucosaminidase, GsmA, is required for the maturation process of sporangiospores in A. missouriensis; sporangia of the gsmA null mutant (ΔgsmA) strain released chains of 2–20 spores under sporangium dehiscence-inducing conditions. In this study, we identified and characterized a putative cell wall hydrolase (AsmA) that is also involved in sporangiospore maturation. AsmA was predicted to have a signal peptide for the general secretion pathway and an N-acetylmuramoyl-l-alanine amidase domain. The transcript level of asmA increased during the early stages of sporangium formation. The asmA null mutant (ΔasmA) strain showed phenotypes similar to those of the wild-type strain, but sporangia of the ΔgsmAΔasmA double mutant released longer spore chains than those from the ΔgsmA sporangia. Furthermore, a weak interaction between AsmA and GsmA was detected in a bacterial two-hybrid assay using Escherichia coli as the host. Based on these results, we propose that AsmA is an enzyme that hydrolyzes peptidoglycan at septum-forming sites to separate adjacent spores during sporangiospore maturation in cooperation with GsmA in A. missouriensis.

IMPORTANCE

Actinoplanes missouriensis produces sporangiospores as dormant cells. The spores inside the sporangia are assumed to be formed from prespores generated by the compartmentalization of intrasporangium hyphae via septation. Previously, we identified GsmA as a cell wall hydrolase responsible for the separation of adjacent spores inside sporangia. However, we predicted that an additional cell wall hydrolase(s) is inevitably involved in the maturation process of sporangiospores because the sporangia of the gsmA null mutant strain released not only tandemly connected spore chains (2–20 spores) but also single spores. In this study, we successfully identified a putative cell wall hydrolase (AsmA) that is involved in sporangiospore maturation in A. missouriensis.

INTRODUCTION

Actinomycetes constitute a large group of gram-positive bacteria with a high G + C content, and typical filamentous actinomycetes undergo a complex life cycle comprising spore formation from special differentiating hyphae. In particular, the morphological development of members of the genus Streptomyces has been extensively characterized as a model microorganism (1–4). They form branched substrate mycelia during vegetative growth and subsequently produce aerial hyphae that grow above substrate mycelia. Finally, unigenomic spores are generated via septation along aerial hyphae. Filamentous actinomycetes, other than Streptomyces, also display a complex lifestyle. Actinoplanes missouriensis, the most characterized member of the genus Actinoplanes, is an actinomycete that forms branched substrate mycelia during vegetative growth and produces globose or subglobose terminal sporangia growing from substrate mycelia via short sporangiophores (5, 6). Each sporangium contains a few hundred spores, and the exterior space of the spores inside the sporangium is filled with an intrasporangial matrix called the sporangium matrix (7). In response to water, sporangia open to release spores into the external environment through a process called sporangium dehiscence (8). After release from sporangia, spores swim in aquatic environments using flagella as zoospores. Eventually, zoospores stop swimming and begin to germinate in niches suitable for vegetative growth (9–12). Because of its life cycle, including sporangium formation, A. missouriensis is an ideal bacterium for research on sporangiospores, which are assumed to be formed from prespores generated by the septation of hyphae inside a sporangium.

On sporangium-forming humic acid-trace element (HAT) agar, A. missouriensis produces small sporangium-like structures after 2- to 3-day cultivation. Then, mature sporangia that can release spores under dehiscence-inducing conditions are formed after incubation for 5–7 days (13, 14). In a previous study, we identified and characterized a cell wall hydrolase, GsmA, that plays a pivotal role in sporangiospore maturation (15). Under dehiscence-inducing conditions, the gsmA null mutant (ΔgsmA) sporangia released many spore chains, in which 2–20 spores were connected in tandem, while the wild-type sporangia released single spores. In silico searches showed that GsmA has a signal peptide for the twin arginine translocation (TAT) pathway, two bacterial SH3-like domains that could mediate protein-protein interactions, and an N-acetylglucosaminidase domain (see Fig. 1). Zymographic analysis using a recombinant protein containing the N-acetylglucosaminidase domain of GsmA demonstrated that GsmA hydrolyzes cell wall peptidoglycans extracted from A. missouriensis and Streptomyces griseus (15). Therefore, GsmA is a cell wall hydrolase required for the separation of adjacent spores by hydrolyzing peptidoglycan at septum-forming sites during sporangiospore maturation.

Fig 1.

Transcript levels of six genes encoding possible cell wall lytic enzymes. The number of reads per kilobase of coding sequence per million mapped reads (RPKM) of asmA, AMIS_7130, AMIS_73700, AMIS_52470, AMIS_81130, and gsmA in the wild-type strain from a previous RNA sequencing analysis is shown (13). RNA samples were prepared from substrate hyphae (day 1) or mixtures of substrate hyphae and sporangia (days 3 and 6) grown on HAT agar for 1, 3, or 6 days. The putative functions of the gene products are shown in parentheses. Schematics showing the domain structures of the gene products are shown above the graphs. SP, signal peptide; amidase, N-acetylmuramoyl-l-alanine amidase domain; M23, peptidase M23 domain; S9, peptidase S9 domain; PG-binding, peptidoglycan-binding domain; SH3-like, bacterial SH3-like domain; glucosaminidase, N-acetylglucosaminidase-like domain.

Although the identification of GsmA shed light on the largely unknown process of sporangiospore maturation, the process proceeded to some extent without GsmA, i.e., in the ΔgsmA strain, because not only spore chains but also single spores are released from the ΔgsmA sporangia (15). These results suggested that an unknown cell wall hydrolase(s), other than GsmA, is involved in the maturation of sporangiospores. This hydrolase(s) should partially compensate for the deficiency in spore separation caused by the lack of GsmA. In the present study, we searched for such a cell wall hydrolase(s) that plays an important role in sporangiospore maturation, and successfully identified AMIS_65730, which is a putative N-acetylmuramoyl-l-alanine amidase, as a member of the cell wall-degrading enzymes that are involved in sporangiospore formation. Hereafter, we named AMIS_65730 AsmA (cell wall amidase required for sporangiospore maturation) for its putative function. We genetically characterized the asmA gene and proposed that AsmA is the second cell wall hydrolase that plays an important role in sporangiospore maturation in cooperation with GsmA in A. missouriensis.

RESULTS

Selection of cell wall hydrolase-encoding genes

To identify the cell wall lytic enzyme(s) involved in sporangiospore maturation, we focused on our previous transcriptomic data using RNA sequencing (RNA-Seq) analysis of the wild-type strain (13). Among the genes highly activated during sporangium formation, we focused on three genes: asmA, AMIS_7130, and AMIS_73700, which encode proteins of 693, 312, and 620 amino acids, respectively (Fig. 1). A protein database search using InterPro ver. 86.0 (http://www.ebi.ac.uk/interpro/) (16) showed that AsmA has an N-acetylmuramoyl-l-alanine amidase domain in its central region (IPR002502; residues 229 to 380), which cleaves the linkage between the N-acetylmuramoyl moiety and the l-amino acid residue in the cell wall peptidoglycan. AMIS_7130 contains a peptidase M23 domain in its C-terminal region (IPR016047; residues 199–292). The peptidase M23 domain is one of the most common catalytic domains among cell wall hydrolases and specifically cleaves the interpeptide bridges of peptidoglycans (17–19). AMIS_73700 has a peptidase S9 domain in its C-terminal region (IPR001375; residues 452–562). Although the peptidase S9 domain has not been reported to be responsible for cell wall peptidase activity to date (20), we included AMIS_73700 as one of the target genes because the transcript level of AMIS_73700 in wild-type mycelia (with premature sporangia) grown on HAT agar for 3 days was 18.5-fold higher than that in mycelia grown on HAT agar for 1 day (Fig. 1). The SignalP 5.0 server (http://www.cbs.dtu.dk/services/SignalP/) (21) predicted that AsmA and AMIS_7130 have a signal peptide for the general secretion (Sec) pathway (cleavage sites are between residues 43 and 44 and residues 60 and 61, respectively). In contrast, no signal peptide was predicted for AMIS_73700. These results indicated that AsmA and AMIS_7130 are secreted proteins similar to GsmA and that AMIS_73700 is probably located in the cytoplasm. Considering the predicted cellular location of AMIS_73700, its involvement in the degradation of cell wall peptidoglycan is unlikely. Nevertheless, we examined the function of AMIS_73700 by gene disruption, as described below, because it may be involved in sporangium formation.

In addition to the three genes induced during sporangium formation, we focused on two genes, AMIS_52470 and AMIS_81130, which encode proteins of 776 and 383 amino acids, respectively, as candidate genes encoding cell wall hydrolases involved in sporangiospore maturation, because both gene products are predicted to function as N-acetylmuramoyl-l-alanine amidases (Fig. 1). A protein database search using InterPro showed that AMIS_52470 and AMIS_81130 have an N-acetylmuramoyl-l-alanine amidase domain (IPR002502 for residues 300–460 in AMIS_52470; IPR002508 for residues 174–355 in AMIS_81130). AMIS_81130 also contains two peptidoglycan-binding domains (IPR002477; residues 10–65 and 87–142). The peptidoglycan-binding domain is found in a variety of enzymes involved in cell wall degradation, such as CwlA, an N-acetylmuramoyl-l-alanine amidase in Bacillus subtilis (22). SignalP predicted that AMIS_52470 contains a signal peptide for the TAT pathway, in which the cleavage site is between residues 26 and 27. In contrast, no signal peptide was predicted for AMIS_81130. However, we assumed that AMIS_81130 is a secreted protein because it contains two peptidoglycan-binding domains. Considering the possibility that these gene products compensate for the lack of AsmA in the ΔasmA strain, we included these genes in the targets of gene disruption experiments as described below, although the transcription of AMIS_52470 and AMIS_81130 was not activated during sporangium formation (Fig. 1).

AsmA is involved in sporangiospore maturation

To examine the in vivo functions of AsmA, AMIS_7130, and AMIS_73700, null mutant strains in which one of the three genes was deleted (ΔasmA, ΔAMIS_7130, and ΔAMIS_73700 strains) were generated. No difference was observed between the wild-type and mutant strains by macroscopic observation of mycelia on nutrient-rich yeast extract-beef extract-NZ amine-maltose monohydrate (YBNM) and HAT agar (data not shown). We then cultivated the wild-type and mutant strains on HAT agar at 30°C for 7 days and observed the substrate mycelia and sporangia by scanning electron microscopy (SEM). The three mutant strains formed globose or subglobose sporangia, similar to the wild-type strain (Fig. S1A through D in the Supplementary Material). We also observed sporangium dehiscence using sporangia scraped from mycelia grown on HAT agar at 30°C for 7 days. Sporangium dehiscence can be induced by incubating sporangia in 25 mM histidine solution. Under the conditions tested in this study, sporangia of the three mutant strains opened and released single spores, similar to the wild-type sporangia (data not shown). We quantified the spores released from sporangia of the wild-type and mutant strains by counting the colonies formed on YBNM agar after incubation at 30°C for 2 days. Consequently, all sporangia of the ΔasmA, ΔAMIS_7130, and ΔAMIS_73700 strains formed on one HAT agar plate released almost the same number of spores as wild-type sporangia (Fig. S2). Therefore, we observed no phenotypic changes in these three mutant strains compared with the wild-type strain.

Considering the possibility that cell wall lytic enzymes involved in sporangiospore maturation have overlapping activities, we further generated mutant strains in which two genes were deleted from the chromosome. In this experiment, we used the ΔgsmA strain as a parental strain because it is possible that GsmA compensated for the lack of other cell wall hydrolases in the mutant strains owing to its pivotal role in sporangiospore maturation (15). We generated three double mutant strains (ΔgsmAΔasmA, ΔgsmAΔAMIS_7130, and ΔgsmAΔAMIS_73700). No difference was observed among the wild-type, ΔgsmA, and double mutant strains by macroscopic observation of mycelia on YBNM and HAT agar (data not shown). We then observed the mycelia and sporangia grown on HAT agar at 30°C for 7 days using SEM to examine sporangium formation in detail. However, all the strains produced normal sporangia (Fig. S1A and E through H).

Next, we observed sporangium dehiscence and spores released from sporangia using phase-contrast microscopy. In this experiment, sporangia produced on HAT agar were harvested and suspended in 25 mM histidine solution to induce sporangium dehiscence. The wild-type sporangia normally opened and released a large number of single spores (Fig. 2A through C). In contrast, ΔgsmA sporangia released spore chains, in addition to single spores (Fig. 2D through F), as described previously (15). Under the same conditions, ΔgsmAΔasmA sporangia also opened and released single spores and spore chains, but the released spore chains seemed to be longer than those from the ΔgsmA sporangia (Fig. 2G through I). In contrast, the ΔgsmAΔAMIS_7130 and ΔgsmAΔAMIS_73700 strains showed phenotypes similar to those of the ΔgsmA strain (data not shown). Thus, we quantified the state (free or connected) of zoospores via microscopic observation of more than 5,000 zoospores of the ΔgsmA and ΔgsmAΔasmA strains, both of which contained an empty vector, pTYM19-Apra, on the chromosome. Consistent with previous results (15), more than half (56%) of the analyzed spores released from the ΔgsmA sporangia were arranged in chains, although single zoospores (44%) were also observed (Fig. 3A). In the ΔgsmAΔasmA strain, the proportion of spores in chains in the analyzed spores was much higher (83%) than that in the ΔgsmA strain, and only a small portion (17%) was observed as single zoospores (Fig. 3B). In a gene complementation test, introduction of the asmA gene with its own promoter into the ΔgsmAΔasmA strain resulted in a higher proportion of single zoospores (35%) than in the ΔgsmAΔasmA strain, confirming that AsmA is involved in sporangiospore maturation (Fig. 3C).

Fig 2.

Observation of sporangia and zoospores by phase-contrast microscopy. Sporangia produced on HAT agar were harvested and suspended in 25 mM histidine solution to induce sporangium dehiscence. Microscopic images of the wild-type (A–C), ΔgsmA (D–F), and ΔgsmAΔasmA (G–I) strains are shown. Panels A, D, and G were taken immediately after the suspension. Panels B, E, and H were taken 15 min after the suspension. Panels C, F, and I were taken 30 min after the suspension. Immediately after suspension, the sporangia appeared phase-bright (A, D, and G). The sporangium membrane gradually became transparent before the spore release (B, E, and H). Sporangia (including transparent ones) are indicated by arrows. The released spores and spore chains are indicated by arrowheads. Bars, 10 µm.

Fig 3.

Proportions of single and tandemly connected zoospores in the ΔgsmA (A) and ΔgsmAΔasmA (B) strains, both of which contained pTYM19-Apra, and the ΔgsmAΔasmA strain harboring the asmA complementation plasmid (C). The total number (n) of observed zoospores of each strain is shown below each chart. Zoospores were observed as single zoospores or components of spore chains. We believe that multiple hyphae occur by branching in a sporangium, but we have not been able to definitively demonstrate this. Furthermore, we do not know exactly how many spores are formed from a single intrasporangium hypha. Because, in our previous study (15), the longest spore chain in the ΔgsmA strain consisted of 20 spores, we believe that 10–20 spores can be formed from a single intrasporangium hypha.

Conditional sporangium dehiscence deficiency was observed in the ΔgsmA and ΔgsmAΔasmA strains

We then quantified spores and spore chains released from the sporangia of the ΔgsmA, ΔgsmAΔasmA, ΔgsmAΔAMIS_7130, and ΔgsmAΔAMIS_73700 strains by cultivating spores and spore chains on YBNM agar at 30°C for 2 days. In this experiment, we induced sporangium dehiscence by pouring 25 mM NH₄HCO₃ solution on the sporangia produced on HAT agar. Because not only a free spore but also a spore chain can form a single colony after cultivation on YBNM agar, we expected that the number of colonies formed by a strain that produced spore chains would be lower than that formed by the wild-type strain. Considering the proportions of single and tandemly connected zoospores (Fig. 3), 33% and 57% decrease was expected in the numbers of colonies in the ΔgsmA and ΔgsmAΔasmA strains, respectively, compared to the wild-type strain. However, the number of colonies of the ΔgsmA and ΔgsmAΔasmA strains was only 0.24% and 0.03%, respectively, of the number of colonies formed by the wild-type strain (Fig. S2). Using phase-contrast microscopy, we roughly confirmed that decreased amounts of single spores and spore chains were released from these two mutants (Fig. S3). We also confirmed that the zoospores of these two mutants could germinate normally under nutrient-rich conditions (Fig. S4). Therefore, the dramatically decreased colony-forming unit (CFU) numbers in the ΔgsmA and ΔgsmAΔasmA strains were ascribed to the deficiency of sporangium dehiscence. Because the total number of spores (single spores plus spores in spore chains) did not seem to be very different among the wild-type, ΔgsmA, and ΔgsmAΔasmA strains in the phase-contrast microscopic observation when the sporangia were suspended in 25 mM histidine solution (data not shown), the deficiency of sporangium dehiscence in the ΔgsmA and ΔgsmAΔasmA strains was observed only when 25 mM NH₄HCO₃ solution was poured onto the sporangia produced on HAT agar. This was unexpected because we sometimes observed that sporangium dehiscence of some mutant strains occurred more efficiently in the 25 mM NH₄HCO₃ solution-pouring procedure than in the sporangium suspension procedure using 25 mM histidine solution. The reason for this conditional sporangium dehiscence deficiency in the ΔgsmA and ΔgsmAΔasmA strains was unclear (see Discussion).

Sporangia of the ΔgsmAΔAMIS_7130 and ΔgsmAΔAMIS_73700 strains formed on one HAT agar plate released almost the same number of spores as ΔgsmA sporangia in the 25 mM NH₄HCO₃ solution-pouring procedure (Fig. S2). In contrast, the number of colonies of the ΔgsmAΔasmA strain was significantly lower than that of the ΔgsmA strain (Fig. S2). This result showed that deletion of asmA, but not AMIS_7130 or AMIS_73700, had an effect on the phenotypes of the ΔgsmA strain. As described above, the reason for the decreased efficiency of sporangium dehiscence is unclear; however, the deletion of asmA in the ΔgsmA strain appeared to enhance the phenotypic changes caused by the deletion of gsmA. Thus, this result supports the functional relationship between GsmA and AsmA.

Disruption of other putative N-acetylmuramoyl-l-alanine amidase genes

Although longer spore chains were released from ΔgsmAΔasmA sporangia than ΔgsmA sporangia (Fig. 2 and 3), no phenotypic changes were observed between the wild-type and ΔasmA strains (Fig. S1 and S2). As a cause of this result, we hypothesized that an unidentified amidase(s) compensated for the loss of AsmA in the ΔasmA strain. Considering the possibility that GsmA, which has two bacterial SH3-like domains presumably for protein-protein interactions, functions as a core component of a possible cell wall-modifying protein complex, we also hypothesized that this unidentified amidase(s) functions only cooperatively with GsmA, which can explain why it could not compensate for the loss of AsmA in the ΔgsmAΔasmA strain. As described above, AMIS_52470 and AMIS_81130, in addition to asmA, encode N-acetylmuramoyl-l-alanine amidases in the A. missouriensis genome. Therefore, these gene products are candidates for the amidase involved in sporangiospore maturation in cooperation with GsmA. To examine whether AMIS_52470 and AMIS_81130 are involved in sporangium formation and sporangiospore maturation, we generated two double (ΔasmAΔAMIS_52470 and ΔasmAΔAMIS_81130) and one triple (ΔasmAΔAMIS_52470ΔAMIS_81130) mutant strains using the ΔasmA strain as the parental strain. Observation of mycelia and sporangia grown on HAT agar at 30°C for 7 days using SEM revealed that all the mutant strains formed normal sporangia (Fig. S1I through K). Furthermore, these mutant sporangia normally opened and released a similar number of spores as the wild-type strain under sporangium dehiscence-inducing conditions (Fig. S2).

Considering the pivotal role of GsmA in sporangiospore maturation (15), we also generated two triple (ΔgsmAΔasmAΔAMIS_52470 and ΔgsmAΔasmAΔAMIS_81130) and one quadruple (ΔgsmAΔasmAΔAMIS_52470ΔAMIS_81130) mutant strains, using the ΔgsmAΔasmA strain as the parental strain. Although we initially hypothesized that the unidentified amidase would function only cooperatively with GsmA, we thought that some negative effects of gene disruption on sporangiospore maturation might be detected in the ΔgsmAΔasmA genetic background. However, all the mutant strains formed normal sporangia (Fig. S1L through N), which released a similar number of spores/spore chains as the ΔgsmAΔasmA sporangia under sporangium dehiscence-inducing conditions (Fig. S2). Based on these results, we concluded that AMIS_52470 and AMIS_81130 are not involved in sporangium formation and sporangiospore maturation in A. missouriensis.

Possible interactions between AsmA and GsmA

In B. subtilis, the N-acetylglucosaminidase LytD was suggested to interact with other cell wall lytic enzymes or cell surface proteins via its SH3-like domain (23). Because GsmA has two SH3-like domains (15) and both AsmA and GsmA are involved in sporangiospore maturation, we hypothesized that AsmA and GsmA interact with each other to form a cell wall-degrading protein complex. To test this possibility, we performed a bacterial adenylate cyclase-based two-hybrid (BACTH) assay using Escherichia coli as the host. In this assay, we quantified protein-protein interactions using a β-galactosidase assay in liquid culture. Consequently, we detected significantly higher β-galactosidase activity in the E. coli cells co-transformed with the asmA- and gsmA-expressing plasmids than in the control cells with empty vectors, indicating that AsmA and GsmA interact with each other (Fig. 4). However, we speculate that the interaction may be weak because the β-galactosidase activity was not very high. Since the negative effect of asmA inactivation on sporangiospore maturation was clearly detected in the ΔgsmA strain, AsmA seems to function in the absence of GsmA (i.e., as a free form), which also supports the notion that the interaction between GsmA and AsmA is not very strong.

Fig 4.

BACTH assay for AsmA and GsmA. β-Galactosidase activity (Miller unit) of E. coli BTH101 strains co-transformed with pUT18C (or its derivative) and pKT25 (or its derivative) was measured. pUT18C-asmA and pKT25-gsmA encode AsmA fused with T18 (T18-AsmA) and GsmA fused with T25 (T25-GsmA), respectively. Values represent the mean ± standard error of three biological replicates tested for each interaction. Statistically significant differences between the two transformants (P-value <0.05) are marked with asterisks.

DISCUSSION

In the present study, we revealed that AsmA is involved in sporangiospore maturation in A. missouriensis. AsmA is predominantly produced during sporangium formation and is presumably secreted into the intrasporangial matrix space through the Sec pathway, where it presumably functions as a cell wall hydrolase to separate sporangiospores along the spore-forming hyphae. A database search using InterPro showed that AsmA has an N-acetylmuramoyl-l-alanine amidase domain (IPR002502). The most extensively analyzed cell wall hydrolases with this domain are the AmpD proteins in E. coli and Pseudomonas aeruginosa, which rapidly cleave the linkage between 1,6-anhydro-N-acetylmuramic acid (MurNAc) and l-alanine in MurNAc-tri and -tetrapeptides (24, 25). The catalysis is zinc-dependent and the structure of the catalytic domain has been well investigated (26–29). In actinomycetes, peptidoglycan structures of several species in the genera Micromonospora, Catenuloplanes, and Couchioplanes, all of which belong to the family Micromonosporaceae, have been analyzed (30). The peptide stems consist of Gly-d-Glu-meso-DAP-d-Ala in Micromonospora aurantiaca JCM 3232 and Gly-d-Glu-l-Lys-d-Ala in Catenuloplanes japonicus NBRC 14176 and Couchioplanes caeruleus subsp. caeruleus NBRC 13939. Because the genus Actinoplanes belongs to the family Micromonosporaceae and its peptidoglycan was suggested to have a similar structure to that of the genus Micromonospora (31), we assumed that AsmA hydrolyzes the linkage between N-glycolylmuramic acid and the glycine residue, which separates the stem peptide of peptidoglycan from the glycan strand, whereas the N-acetylglucosaminidase GsmA hydrolyzes the linkage between N-acetylglucosamine and N-glycolylmuramic acid (Fig. 5).

Fig 5.

Schematic representation of the working hypothesis of major cell wall hydrolases during sporangiospore maturation. The peptidoglycan structure of A. missouriensis is based on the suggestion that it is similar to that of the genus Micromonospora (31). The non-crosslinked d-Ala residue at the C-terminus of the tetrapeptide may be lost in peptidoglycan. AsmA and GsmA are predicted to hydrolyze the linkages between N-glycolylmuramic acid and the glycine residue and between N-acetylglucosamine and N-glycolylmuramic acid, respectively, as indicated by arrows. AsmA and GsmA can form a complex, but complex formation does not seem to be essential for their enzymatic activity. Another cell wall degrading enzyme(s) (indicated by “X”) may also be involved in sporangiospore maturation. GlcNAc, N-acetylglucosamine; MurNGlyc, N-glycolylmuramic acid; mDAP, mesodiaminopimelic acid.

In B. subtilis, LytC and LytD, which function as N-acetylmuramoyl-l-alanine amidase and N-acetylglucosaminidase, respectively, are involved in separation of vegetative cells (32). In single gene mutants of lytC and lytD, cell separation was prevented by the loss of hydrolysis of nascent septa, leading to an increase in the length of the growing cell chains. Furthermore, longer cell chains were generated in a double mutant lacking lytC and lytD, indicating that these cell wall hydrolases mutually compensated for the lack of each other (32). According to this report, it is possible that GsmA compensated for the lack of AsmA in the ΔasmA strain, leading to no phenotypic change between the wild-type and ΔasmA strains and to a significant difference in the length of spore chains between the ΔgsmA and ΔgsmAΔasmA strains (Fig. 2 and 3). Meanwhile, the importance of cell wall remodeling during the life cycle of Streptomyces coelicolor A3(2) has been indicated in several studies (33–35). However, no cell wall hydrolases specifically involved in spore separation have been identified, although the physiological functions of several cell wall hydrolases, including RpfA (lysozyme), SwlA (endopeptidase/amidase), SwlB (lytic transglycosylase), SwlC (endopeptidase), and SCO4439 (d-alanyl-d-alanine carboxypeptidase), have been analyzed (33, 34).

Unexpectedly, sporangium dehiscence of the ΔgsmA and ΔgsmAΔasmA strains was severely repressed when 25 mM NH₄HCO₃ solution was poured onto the sporangia formed on HAT agar (Fig. S2 and S3), but not when sporangia were suspended in 25 mM histidine solution. This result indicated that GsmA and AsmA are conditionally involved in sporangium dehiscence, probably in an indirect manner, although the functional mechanisms remain to be elucidated. One possibility is that premature spores (i.e., spores in spore chains) in the ΔgsmA and ΔgsmAΔasmA sporangium cannot produce enzymes or proteins that promote sporangium dehiscence in the 25 mM NH₄HCO₃ solution-pouring procedure.

We attempted to examine the cell wall-hydrolyzing activity of AsmA by zymography because this technique has been utilized to detect the cell wall-hydrolyzing activity of the N-acetylmuramoyl-l-alanine amidase CwlA in B. subtilis (36). We failed to produce a soluble AsmA protein in its full-length form (except for the N-terminal signal sequence for the Sec pathway) with an N-terminal polyhistidine tag (His-AsmA) in E. coli. Thus, the insoluble form of His-AsmA was solubilized with a denaturing reagent of 2 M guanidine hydrochloride and purified. In addition, a shorter protein (residues 214–417) containing the N-acetylmuramoyl-l-alanine amidase domain with an N-terminal polyhistidine tag (His-AsmAs) was produced in E. coli. Because His-AsmAs was also produced in an insoluble form, we purified His-AsmAs after solubilization with the denaturing reagent. Although zymographic analysis was performed using the cell wall extracted from A. missouriensis vegetative hyphae, we failed to detect reproducible cell wall-hydrolyzing activity of recombinant His-AsmA or His-AsmAs (data not shown). In this experiment, we added zinc chloride to renaturing buffer solution (50 mM Tris-HCl, pH 7.8) at final concentrations of 0, 2, 50, or 5,000 µM. We assumed that the recombinant proteins were not folded properly during the renaturing process owing to their structural instability. The coordination of Zn²⁺ to three protein ligands and a water molecule is universally observed in enzymes in which Zn²⁺ has a catalytic role (37). In AsmA, zinc-binding residues (His-235, His-369, and Cys-377) and a residue (Tyr-270) that binds to a water molecule involved in Zn²⁺ coordination are conserved (Fig. S5). Therefore, we believe that AsmA has amidase activity, although we could not demonstrate this experimentally in this study. In the BACTH assay, we detected the interaction between AsmA and GsmA in E. coli. The N-terminal portion of mature GsmA harbors two bacterial SH3-like domains, which have been suggested to mediate protein-protein interactions by binding to the Pro-X-X-Pro motif (where X represents any amino acid) (23). Because AsmA contains four copies of the Pro-X-X-Pro motif (Fig. S5), AsmA and GsmA may interact via binding between the SH3-like domains of GsmA and the Pro-X-X-Pro motifs of AsmA.

In A. missouriensis, the transcriptional regulator TcrA globally activates genes involved in sporangium formation, spore dormancy, and sporangium dehiscence, including the flagellar and pilus gene clusters (7, 38). Although the transcription of gsmA is activated by TcrA in the late stages of sporangium formation (15), the transcript level of asmA was not significantly different between the wild-type and tcrA null mutant (ΔtcrA) strains (1.2-fold higher in the wild-type strain than in the ΔtcrA strain after the 6-day cultivation on HAT agar), according to our previous RNA-Seq analysis (7). Therefore, we assumed that the transcription of asmA is activated in a manner different from that of gsmA during sporangium formation. According to our Cappable-Seq analysis (T.T. and Y.O., unpublished data), asmA seems to be expressed as a monocistronic transcript. Furthermore, there are no genes that seem to be functionally related to asmA around the asmA gene in the A. missouriensis genome (Fig. S6). GsmA homologs are highly conserved among Actinoplanes, but not in Streptomyces (15). In contrast, AsmA homologs are widely distributed among actinobacteria, including Actinoplanes and Streptomyces (approximately 40% sequence identity in Streptomyces), although functional analyses have not been performed. Therefore, AsmA homologs may function as a prevalent cell wall-modifying protein in spore-forming actinomycetes, while GsmA homologs have evolved as an original component of the cell wall-modifying protein complex in sporangiospore-forming species. In this respect, it should be noted that the surrounding environments are completely different between the intrasporangium hypha of Actinoplanes and the aerial hypha of Streptomyces; the former is enclosed by the aqueous sporangium matrix and the latter is enclosed by air. Although septation is required for compartmentalization to produce spores in both hyphae, different strategies to separate adjacent cells should have been evolutionarily developed. Since the separation of cells in spore chains occurred to some extent in the absence of both GsmA and AsmA, there seems to be another cell wall hydrolase(s) involved in sporangiospore maturation. However, the identification of AsmA as the second enzyme required for sporangiospore maturation is important to understanding the largely unexplored process of spore maturation inside a sporangium.

MATERIALS AND METHODS

General methods

The bacterial strains, plasmid vectors, and media used in this study have been described previously (8, 14, 38). The primers used in this study are listed in Table S1 in the Supplementary Material. A. missouriensis cells were prepared as described previously (7). SEM was performed using an S-4800 scanning electron microscope (Hitachi, Tokyo, Japan) as described previously (39).

Construction of gene deletion mutant strains

To construct gene deletion mutants, the upstream and downstream regions of the target gene were amplified using PCR. The amplified DNA fragments were digested with EcoRI and XbaI (for upstream regions) or XbaI and HindIII (for downstream regions), and cloned into pUC19 digested with the same restriction enzymes. To disrupt AMIS_81130, XbaI and Sse8387I (upstream region) and Sse8387I and HindIII (downstream region) were used. The generated plasmids were sequenced to confirm that no PCR-derived errors were present. The cloned fragments were digested with EcoRI and XbaI (for upstream regions) and XbaI and HindIII (for downstream regions) and cloned together into the EcoRI and HindIII sites of pK19mobsacB (40), whose kanamycin resistance gene had been replaced with the apramycin resistance gene aac(3)IV (14). To disrupt AMIS_81130, XbaI and Sse8387I (upstream region) and Sse8387I and HindIII (downstream region) were used for the cloning, and pK19mobsacB with aac(3)IV was digested with XbaI and HindIII. The generated plasmids were introduced into A. missouriensis by conjugation as described previously (13). Apramycin-resistant colonies resulting from single crossover recombination were isolated. One of them was cultivated in peptone-yeast extract-magnesium liquid medium at 30°C for 36 h, and the mycelia suspended in 0.75% NaCl solution were spread onto Czapek-Dox Broth agar medium (BD, NJ, USA) containing extra sucrose (final concentration 5%). After incubation at 30°C for 5 days, sucrose-resistant colonies were inoculated onto YBNM agar with or without apramycin to confirm that they were sensitive to apramycin. The apramycin-sensitive and sucrose-resistant colonies resulting from the second crossover recombination were isolated as candidates for gene deletion mutants. Disruption of the target genes was confirmed by PCR (data not shown).

Construction of the recombinant strain for the complementation test

A 2.6-kbp DNA fragment containing the promoter and coding sequences of asmA was amplified by PCR. The amplified fragment was digested with EcoRI and HindIII, and cloned into pTYM19-Apra (7, 41) digested with the same restriction enzymes, resulting in pTYM19-Apra-asmA. Plasmid pTYM19-Apra-asmA was sequenced to confirm that no PCR-derived error was introduced and introduced into the ΔgsmAΔasmA strain by conjugation as described previously (13). Plasmid pTYM19-Apra was also introduced into the ΔgsmA and ΔgsmAΔasmA strains for the vector control strains. Apramycin-resistant colonies were obtained.

Microscopic observation of sporangium dehiscence

The wild-type, ΔgsmA, and ΔgsmAΔasmA strains were grown on HAT agar at 30°C for 7 days to prepare sporangia. Sporangia and substrate hyphae formed on HAT agar were harvested with a spatula and mixed with 25 mM histidine solution in a 2.0 mL plastic tube. Immediately after mixing or following incubation with rotation for 15 or 30 min, the suspension was observed using a BH-2 phase-contrast microscope (Olympus, Tokyo, Japan).

Quantification of free spores and spore chains

The ΔgsmA and ΔgsmAΔasmA strains, both of which contained pTYM19-Apra on the chromosome, and the ΔgsmAΔasmA strain harboring the asmA complementation plasmid were grown on HAT agar at 30°C for 7 days. Sporangia and substrate hyphae formed on HAT agar were harvested and mixed with 25 mM histidine solution. The suspension was incubated with rotation at room temperature for 1 h to induce sporangium dehiscence. Then, glycerol (final concentration 50%) was added to the suspension to repress zoospore motility. The micrographs of zoospores were recorded using a BH-2 phase-contrast microscope, and the numbers of free zoospores, spore chains, and component zoospores in spore chains were counted by visual inspection.

Counting of spores released from sporangia

The tested strains were cultivated in peptone-yeast extract-magnesium (PYM) liquid broth with shaking at 30°C for 2 days. A portion of the mycelium (approximately 1 mL of preculture) was inoculated onto HAT agar plates, and the plates were incubated at 30°C for 7 days to produce sporangia. To release zoospores from sporangia, 10 mL of 25 mM NH₄HCO₃ solution was poured onto one HAT agar plate, and the plate was incubated at room temperature for 1 h. After collection from the plate, the zoospore suspension was filtered through a 5 µm Acrodisc membrane filter (Pall Corporation, NY, USA) to eliminate hyphae and sporangia. A portion of the filtrate was inoculated onto YBNM agar plates, and the plates were incubated at 30°C for 2 days. The number of zoospores released from the sporangia was estimated by counting the colonies formed on YBNM agar.

Microscopic observation of germination and outgrowth in zoospores

Zoospore suspensions of the wild-type, ΔgsmA, and ΔgsmAΔasmA strains were prepared by pouring 25 mM NH₄HCO₃ solution onto sporangium-forming HAT agar plates as described above. After collection from the plates, the zoospore suspensions were filtered through a 5 µm Acrodisc membrane filter (Pall Corporation) to eliminate hyphae and sporangia. A portion of the filtrate was mixed with an equal volume of 2 × PYM liquid broth [1% Bacto peptone (Difco), 0.6% yeast extract (Difco), and 0.2% MgSO₄·7H₂O, pH 7.0] and cultivated with shaking at 30°C for 4 h to induce germination and outgrowth. Following cultivation for 2 or 4 h, the suspensions were observed under a BH-2 phase-contrast microscope (Olympus).

BACTH assay

The bacterial two-hybrid assay was conducted using a BACTH system kit (Euromedex, Strasbourg, France), according to the manufacturer’s instructions. For the construction of the T18 or T25 domain fusion plasmids, the coding sequences of asmA and gsmA were amplified by PCR. The DNA fragments were digested with PstI and EcoRI, and cloned into pUC19 digested with the same restriction enzymes. The generated plasmids were sequenced to confirm that no PCR-derived errors were present. Then, the cloned fragments were digested with PstI and EcoRI and cloned into the vectors pUT18C (asmA) and pKT25 (gsmA), digested with the same restriction enzymes. E. coli BTH101 cells were co-transformed with the T18 and T25 domain fusion plasmids, and transformants were selected on Luria-Bertani (LB) agar containing ampicillin and kanamycin. At least three individual colonies per assay were grown overnight at 30°C in LB broth containing ampicillin and kanamycin. The cultures were inoculated into LB broth containing ampicillin, kanamycin, and isopropyl-β-d-1-thiogalactopyranoside (IPTG) and cultivated at 30°C for 48 h. β-Galactosidase activity was quantified as previously described (42).

Production and purification of recombinant proteins

The 1.9- and 0.6-kbp DNA fragments covering the full-length (except for the N-terminal signal peptide) and N-acetylmuramoyl-l-alanine amidase domain of asmA, respectively, were amplified by PCR. The fragments were digested with EcoRI and HindIII, and cloned into pUC19 digested with the same restriction enzymes to generate pUC19-asmA and pUC19-asmAs. Plasmids pUC19-asmA and pUC19-asmAs were sequenced to confirm that PCR-derived errors were not introduced. The cloned fragments were digested with EcoRI and HindIII, and cloned into pColdII digested with the same restriction enzymes to generate pColdII-asmA and pColdII-asmAs. Plasmids pColdII-asmA and pColdII-asmAs were introduced into E. coli BL21(DE3). The transformants were cultivated in LB broth (50 mL) at 37°C for 2.5 h and at 15°C for 30 min. Then, IPTG was added to the culture to a final concentration of 1 mM. After further cultivation at 15°C for 24 h, cells were collected by centrifugation at 3,000 × g for 10 min, suspended in 5 mL of lysis buffer (50 mM NaH₂PO₄, 300 mM NaCl, 10 mM imidazole, and 10% glycerol, pH 8.0), and disrupted by sonication. After centrifugation at 10,000 × g for 30 min, the cell debris containing insoluble proteins was solubilized by suspension in denaturing buffer (50 mM NaH₂PO₄, 300 mM NaCl, 10 mM imidazole, 2 M guanidine hydrochloride, and 10% glycerol, pH 8.0). The N-terminally polyhistidine-tagged full-length AsmA (His-AsmA) and N-acetylmuramoyl-l-alanine amidase domain (His-AsmAs) proteins were purified from the solubilized proteins using Ni-nitrilotriacetic acid Superflow resin (Qiagen, Tokyo, Japan) according to the manufacturer’s instructions. His-AsmA and His-AsmAs were eluted with elution buffer (50 mM NaH₂PO₄, 300 mM NaCl, 400 mM imidazole, 2 M guanidine hydrochloride, and 10% glycerol, pH 8.0). Then, trichloroacetic acid was added to the purified proteins to a final concentration of 10% (wt/vol) and the proteins were precipitated by centrifugation at 20,000 × g for 30 min. After washing with acetone, the quality of the purified proteins was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00456-23.

Supplemental material. jb.00456-23-s0001.pdf.

Fig. S1 to S6 and Table S1.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.