Novel Cell Wall Hydrolase CwlC from Bacillus thuringiensis Is Essential for Mother Cell Lysis

ABSTRACT

In this study, a sporulation-specific gene (tentatively named cwlC) involved in mother cell lysis in Bacillus thuringiensis was characterized. The encoded CwlC protein consists of an N-terminal N-acetylmuramoyl-l-alanine amidase (MurNAc-LAA) domain and a C-terminal amidase02 domain. The recombinant histidine-tagged CwlC proteins purified from Escherichia coli were able to directly bind to and digest the B. thuringiensis cell wall. The CwlC point mutations at the two conserved glutamic acid residues (Glu-24 and Glu-140) shown to be critical for the catalytic activity in homologous amidases resulted in a complete loss of cell wall lytic activity, suggesting that CwlC is an N-acetylmuramoyl-l-alanine amidase. Results of transcriptional analyses indicated that cwlC is transcribed as a monocistronic unit and that its expression is dependent on sporulation sigma factor K (σ^K). Deletion of cwlC completely blocked mother cell lysis during sporulation without impacting the sporulation frequency, Cry1Ac protein production, and insecticidal activity. Taken together, our data suggest that CwlC is an essential cell wall hydrolase for B. thuringiensis mother cell lysis during sporulation. Engineered B. thuringiensis strains targeting cwlC, which allows the crystal inclusion to remain encapsulated in the mother cell at the end of sporulation, may have the potential to become more effective biological control agents in agricultural applications since the crystal inclusion remains encapsulated in the mother cell at the end of sporulation.

IMPORTANCE Mother cell lysis has been well studied in Bacillus subtilis, which involves three distinct yet functionally complementary cell wall hydrolases. In this study, a novel cell wall hydrolase, CwlC, was investigated and found to be essential for mother cell lysis in Bacillus thuringiensis. CwlC of B. thuringiensis only shows 9 and 21% sequence identity with known B. subtilis mother cell hydrolases CwlB and CwlC, respectively, suggesting that mechanisms of mother cell lysis may differ between B. subtilis and B. thuringiensis. The cwlC gene deletion completely blocked the release of spores and crystals from the mother cell without affecting insecticidal activity. This may provide a new effective strategy for crystal encapsulation against UV light inactivation.

INTRODUCTION

Bacillus thuringiensis (Bt) is a Gram-positive bacterium that forms dormant spores highly resistant to various environmental stresses. The production of one or more parasporal crystal (Cry) proteins during sporulation is a hallmark of B. thuringiensis, distinguishing it from the closely related species Bacillus anthracis and Bacillus cereus (1). The cry and cyt genes encode the Cry and Cyt proteins, respectively, which are toxic to a broad range of insect larvae, making B. thuringiensis the most commonly used biological pesticide worldwide (2). When grown to late sporulation stages, the mother cell of B. thuringiensis goes through autolysis and subsequently releases the mature spore and Cry proteins. However, the released Cry proteins are frequently inactivated by environmental factors such as UV light when B. thuringiensis is applied in the field (3). One approach to protect Cry proteins against UV light inactivation is to mutate the sigK gene, which encodes a sigma factor essential for the transcription of many late sporulation genes, including the amidase genes involved in mother cell lysis. Deletion of sigK thus leads to a block of mother cell lysis, Cry protein encapsulation in the mother cell, and increased insecticidal persistence (4). However, a serious problem caused by sigK disruption is that the expression of some cry genes is sharply decreased during late sporulation phase because of the regulation of those genes by SigK (5, 6). A similar but potentially more effective approach is to construct engineered strains by mutating genes that encode cell wall hydrolases involved in mother cell lysis without affecting sporulation frequency or Cry protein production. Bacterial cell wall hydrolases (also known as autolysins) form a large and highly diverse group of enzymes that function to hydrolyze the bacterial cell wall by cleaving bonds in polymeric peptidoglycans (7). These hydrolases play important roles in mother cell lysis during the sporulation of B. subtilis and B. cereus group strains.

B. subtilis is a well-established model for studying mother cell lysis (8–11). Three main cell wall lytic enzymes, CwlB (also named LytC), CwlC, and CwlH, were identified in B. subtilis. CwlB is the major vegetative autolysin produced at the end of exponential growth phase, and it is also present during sporulation (8, 12). CwlC and CwlH are sporulation-specific autolysins whose production is SigK dependent (10, 13). Although the cwlH, cwlC, or cwlB single mutation did not affect mother cell lysis, cwlB cwlC and cwlC cwlH double deletion mutants showed defects in the initiation of mother cell lysis, while the cwlB cwlC cwlH triple deletion mutant had a significant decrease in mother cell lysis (10, 14).

Compared with B. subtilis, little is known about autolysins and their regulation in the B. cereus group, which contains B. cereus, B. thuringiensis, B. anthracis, and four other Bacillus species. In previous studies, a few B. cereus proteins have been found to participate in cortex degradation, vegetative cell wall metabolism, and mother cell lysis (15–17). For instance, CwlB, which shows low sequence identity with B. subtilis autolysins, is a newly identified autolysin found to be involved in mother cell lysis in the B. cereus group (18). Deletion of the gene that encodes CwlB in B. thuringiensis caused a delay in mother cell lysis (estimated to be about 2 h) (18). However, other putative and presumably more important cell wall hydrolases involved in mother cell lysis in the B. cereus group remain unknown.

In this study, we characterized a new SigK-dependent and sporulation-specific gene, cwlC, that encodes a putative 27.9-kDa N-acetylmuramoyl-l-alanine amidase (MurNAc-LAA) that is essential for mother cell lysis in B. thuringiensis. We present evidence here that CwlC binds to and shows hydrolytic activity against the Bt cell wall. Disruption of the cwlC gene completely blocked mother cell lysis without impacting sporulation efficiency, crystal protein production, or insecticidal activity.

RESULTS

Bioinformatic analysis of RS15875, which encodes a putative cell wall hydrolase.

To identify cell wall hydrolases essential for mother cell lysis in B. thuringiensis HD73, we screened putative cell wall hydrolase-encoding genes with high transcriptional activities at the late sporulation phase in Schaeffer's sporulation medium (SSM) on the basis of transcriptome sequencing data. Here we found one such candidate gene (RS15875). The open reading frame (ORF) of RS15875 in the B. thuringiensis HD73 genome (NCBI accession no. NC_020238.1) was analyzed. RS15875 is 735 bp in length and is predicted to encode a protein that resembles a cell wall hydrolase. Conserved-domain analysis revealed that the putative cell wall hydrolase encoded by the RS15875 gene contains an N-terminal MurNAc-LAA domain and a C-terminal amidase02 domain (Fig. 1A). The MurNAc-LAA domain is conserved among the N-acetylmuramoyl-l-alanine amidases. For instance, CwlB of B. thuringiensis, CwlC and CwlB of B. subtilis, and CwlM of B. licheniformis all contain a MurNAc-LAA domain (Fig. 1A). The amino acid sequence identity shown by the predicted proteins of the RS15875 gene with CwlB of B. thuringiensis is 12%, and those with CwlC and CwlB of B. subtilis are 21 and 9%, respectively (data not shown), suggesting that RS15875 may be a new N-acetylmuramoyl-l-alanine amidase-encoding gene in B. thuringiensis. Since there were cell wall hydrolase genes previously named cwlA and cwlB in B. thuringiensis (18), RS15875 was tentatively designated cwlC (cell wall lytic enzyme).

FIG 1.

Bioinformatic analysis of cell wall lytic N-acetylmuramoyl-l-alanine amidases. (A) Domain organization of CwlC and CwlB of B. thuringiensis HD73, CwlC and CwlB of B. subtilis 168, and CwlM of B. licheniformis ATCC 9789. MurNAc-LAA (blue bar) represents the N-acetylmuramoyl-l-alanine amidase domain, SH3 (purple bar) stands for Src homology 3 domain, SPOR (yellow bar) stands for sporulation-related domain, and LytB (red bar) stands for cell wall binding domain. (B) Comparison of the amino acid sequences of the catalytic domains of the amidases (CwlC and CwlB of B. thuringiensis HD73, CwlC and CwlB of B. subtilis 168, and CwlM of B. licheniformis ATCC 9789). The two conserved critical glutamic acid residues are highlighted in yellow.

It was demonstrated that in B. subtilis, the cell wall lytic N-acetylmuramoyl-l-alanine amidases contain two conserved glutamic acid residues that act as critical catalytic sites (19). Although CwlC of B. thuringiensis HD73 shows only 21% amino acid sequence identity with CwlC of B. subtilis 168, the critical catalytic residues of CwlC of B. thuringiensis HD73 (glutamic acids located at positions 24 [E24] and 140 [E140], corresponding to E24 and E141 of CwlC of B. subtilis 168, respectively) are conserved (Fig. 1B, highlighted in yellow) on the basis of sequence analysis. In summary, our bioinformatic analysis of the conserved domain and the critical catalytic residues suggested that CwlC of B. thuringiensis HD73 could be an N-acetylmuramoyl-l-alanine amidase.

Characterization of CwlC as a cell wall hydrolase.

To investigate the function of CwlC in B. thuringiensis, the recombinant CwlC-His proteins were expressed in Escherichia coli and purified by affinity chromatography. SDS-PAGE showed that the molecular size of the purified protein is approximately 28 kDa, which is in agreement with the predicted size of CwlC (244 amino acids) plus 6 amino acids from the histidine tag (Fig. 2A, lane 2).

FIG 2.

Characterization of the CwlC protein as a cell wall hydrolase. (A) SDS-PAGE analysis of the CwlC proteins expressed in E. coli BL21(pETCwlC) cells and purified by nickel affinity column chromatography. Lanes: M, protein molecular size markers; CwlC, wild-type CwlC protein; E24A, CwlC protein with a point mutation at position 24 (replacement of glutamic acid with alanine); E140A, CwlC protein with a point mutation at position 140 (replacement of glutamic acid with alanine); E24A E140A, CwlC protein with point mutations at both positions 24 and 140 (replacement of glutamic acids with alanines). (B) Western blot analysis to determine CwlC binding to the Bt cell wall. CwlC-His (lane 1) and SigK-His (lane 4, a control for exclusion of histidine tag interference) were incubated for 30 min on ice in the absence of cell wall and not centrifuged. CwlC-His (lane 3) was incubated alone for 30 min on ice and centrifuged to test if CwlC-His would precipitate. CwlC-His (lane 2) and SigK-His (lane 5, a control for exclusion of histidine tag interference) were also incubated for 30 min on ice in the presence of cell wall and then centrifuged. The pellet, suspended in 20 μl of sterile distilled water, was subjected to SDS-PAGE and visualized by Western blot analysis with an anti-His antibody. A positive control without cell wall was also included, and that sample was not centrifuged. (C) Digestion of B. thuringiensis cell wall by purified CwlC protein. CwlC protein (0.27 μM; wild-type CwlC, red squares; E24A CwlC, green triangles; E140A CwlC, purple circles; E24A/E140A CwlC, orange asterisks) was mixed with isolated B. thuringiensis cell wall resuspended in 0.05 M TK buffer (pH 7.0) and incubated at 37°C. Subsequently, 500-μl aliquots were removed at various time intervals to measure the turbidity of the samples at 540 nm. Cell wall incubated without addition of CwlC protein was used as a negative control (blue diamonds; obscured behind the green, purple, and orange symbols). (D) Quantitation of the polymeric cell wall. After hydrolysis of the B. thuringiensis cell wall by purified CwlC protein (wild-type CwlC, red column; E24A CwlC, green column; E140A CwlC, purple column; E24A/E140A CwlC, orange column) at 37°C for 1 h, 1 ml of the mixture was centrifuged at 4°C for 5 min at 27,000 × g and then freeze-dried for ∼48 h until the pellet became lyophilized powder to quantify polymeric cell wall degradation. Cell wall incubated without added CwlC protein was used as a negative control (blue column). The asterisk indicates a statistically significant difference in the dry weight of the polymeric cell wall after incubation with wild-type CwlC compared with the control (P < 0.05, Student t test). Error bars represent the standard error of the mean.

We first tested the cell wall binding ability of CwlC. To do so, we incubated a mixture of purified CwlC-His protein and the B. thuringiensis cell wall on ice. The SigK-His protein was used as a control to eliminate possible interference from the histidine tag (20). The CwlC-His protein alone was incubated in distilled water without addition of the cell wall as another control to determine whether CwlC-His itself would precipitate. After incubation, the above-described reaction mixtures were centrifuged at 4°C (see Materials and Methods). Western blot analysis was performed to detect His-tagged proteins by using an anti-His antibody. The results showed that both the CwlC-His (Fig. 2B, lane 1) and SigK-His (Fig. 2B, lane 4) proteins could be detected in the supernatant of samples without addition of the cell wall. In the sediment, CwlC-His was detected only in the sample with the cell wall added (Fig. 2B, lane 2), while neither SigK-His incubated with the cell wall (Fig. 2B, lane 5) nor CwlC-His in the absence of the cell wall (Fig. 2B, lane 3) was found in the sediment. These results suggested that the CwlC protein is able to bind to the cell wall of B. thuringiensis. However, no competition was observed when excessive unlabeled CwlC was used as a competitor for binding of the biotin-labeled CwlC protein to the Bt cell wall in a homologous competition assay (see Fig. S1 in the supplemental material). We assumed that the binding sites might not be specific because of the complexity of the cell wall.

To test the cell wall lytic ability of the CwlC protein, a mixture of the CwlC-His protein and the B. thuringiensis cell wall was incubated at 37°C. The turbidity of the sample, which corresponds to the amount of polymeric cell wall, was recorded at 540 nm at the indicated time points during the reaction. The optical density (OD) of the mixture decreased by approximately 50% within 60 min after the wild-type CwlC-His protein was added (Fig. 2C, red squares), suggesting that CwlC-His was able to effectively degrade the polymeric B. thuringiensis cell wall. We also measured the dry weight of the freeze-dried polymeric cell wall after hydrolysis, showing that the cell wall was significantly decreased after incubation with wild-type CwlC protein (Fig. 2D, red column).

To test whether the two conserved glutamic acids (E24 and E140) located in the MurNAc-LAA domain of CwlC are critical for the catalytic activity of the enzyme, site-directed mutagenesis was performed to replace these two residues in CwlC. We introduced two different single amino acid substitutions (E24A [GAA to GCA] and E140A [GAA to GCA]) and one double amino acid substitution (E24A/E140A) into the CwlC-expressing strains. SDS-PAGE showed that the reconstituted CwlC proteins with point mutations were also well expressed, with the correct size (Fig. 2A, lanes 3 to 5). We next tested the cell wall lytic abilities of the mutant CwlC proteins. The result, based on the rate of disappearance of turbidity, suggested that all three mutant CwlC proteins (E24A, green triangles; E140A, purple circles; E24A/E140A, orange asterisks) did not show any lytic activity, compared with the wild-type CwlC protein (Fig. 2C). Also, the dry weight of the freeze-dried polymeric cell wall after hydrolysis with CwlC protein containing a point mutation (Fig. 2D, green, purple, and orange columns) was similar to that of the control (Fig. 2D, blue column). Thus, our results confirmed that CwlC employs the two conserved glutamic acids located in the MurNAc-LAA domain as the key catalytic residues. This again supported the idea that CwlC of B. thuringiensis is an N-acetylmuramoyl-l-alanine amidase.

Transcriptional regulation of the cwlC gene.

In the B. thuringiensis HD73 genome, the cwlC gene (Fig. 3A, blue) is clustered with an upstream gene, RS15870 (Fig. 3A, yellow), that encodes a hypothetical protein, and a downstream gene, RS15880 (Fig. 3A, orange), that is annotated as encoding a putative N-acetylmuramoyl-l-alanine amidase. Since these three ORFs (RS15870, cwlC, and RS15880) are arranged in the same orientation in the genome, we wanted to test if cwlC is transcribed alone or together with the other two genes as a polycistronic unit. The total RNA of B. thuringiensis HD73 cells grown in SSM to T₁₅ (T₀ is the end of the exponential growth phase; T_n is n hours after the end of the exponential growth phase) was extracted and subject to reverse transcription-PCR (RT-PCR). Three pairs of primers (RTcwlC-5/RTcwlC-3, RTC55-5/RTC55-3, and RTC57-5/RTC57-3; Table 1) were designed to examine the transcription of cwlC and the immediate flanking regions (Fig. 3A). RT-PCR results indicated that the RNA corresponding to an internal region of cwlC was present, while the RNAs corresponding to the immediate intergenic regions were not detected (Fig. 3B). This result suggested that the cwlC gene is transcribed as a monocistronic unit (Fig. 3A).

FIG 3.

Analyses of cwlC transcription in B. thuringiensis HD73. (A) Map of the RS15870-RS15880 locus in the B. thuringiensis HD73 genome. The DNA region deleted to disrupt the cwlC gene is indicated. The bent arrow represents the promoter. Dashed lines annotated by letters correspond to RT-PCR amplicons (see lanes in panel B). ORFs are indicated by large open arrows. The scale bar corresponds to 300 bp. (B) RT-PCR analysis of gene expression in the RS15870-RS15880 locus of B. thuringiensis HD73. PCRs without template cDNA were used as negative controls. Positive controls are the PCR with genomic DNA as the template. The letters refer to the positions of the RT-PCR products in the locus, as depicted in panel A. Total RNA was extracted from cells harvested at T₁₅ in SSM. (C) Sequence analysis of the intergenic region (289 bp) between the RS15870 and cwlC genes of B. thuringiensis HD73. The predicted transcription start site (+1) and the putative −35 and −10 motifs are marked. The predicted stop codon of RS15870 and the translation start codon of cwlC are singly and doubly underlined, respectively. (D) Assays of β-galactosidase activity were performed to compare the activities of the cwlC promoter in three different strains [HD73, red squares; HD(ΔgerE), blue circles; HD(ΔsigK), purple triangles] at the indicated time points after incubation in SSM at 30°C with shaking at 220 rpm; Tn, n hours after T₀ (the end of the exponential growth phase). Each value represents the mean of at least three independent replicates. Error bars show standard deviations.

TABLE 1.

Oligonucleotides used in this study

To determine the transcription start of the cwlC gene, a 5′ rapid amplification of cDNA ends (RACE)-PCR experiment was performed (see Materials and Methods). The experiment confirmed that the cwlC gene transcription start site is a G located 21 nucleotides upstream of the cwlC translational start codon (ATG) (Fig. 3C) on the basis of sequences from 12 randomly selected clones obtained by 5′ RACE-PCR (see Fig. S1).

We next decided to investigate the transcriptional regulation of cwlC in B. thuringiensis. The putative cwlC promoter contains the sequences AGCA and AATAAGATA, located in the upstream −35 and −10 regions from the cwlC transcription start site (Fig. 3C). These sequences resemble the consensus sequences seen in the σ^K-dependent promoters (HDCA and CATANNNDD; H is A/C/T, D is A/G/T, and N is A/C/G/T) (21). Our analyses indicated that the cwlC promoter might be recognized by σ^K and cwlC could be regulated in a σ^K-dependent manner. In B. subtilis, some of the σ^K-dependent genes were also reported to be coregulated by the gerE gene (22), which encodes a transcriptional activator for several coat protein genes (10). To investigate the transcriptional regulation of the cwlC gene, a vector containing the P_cwlC-lacZ fusion was constructed and introduced into the wild-type HD73, HD(ΔsigK), and HD(ΔgerE) strains, respectively. β-Galactosidase activity assays showed that expression of the reporter fusion in the parent strain, HD73, started at T₈, reached a maximum at T₁₄, and then sharply decreased (Fig. 3D, red squares). This result indicated that expression of the cwlC gene starts at the late sporulation stage. The expression of P_cwlC was completely blocked in the ΔsigK mutant (Fig. 3D, purple triangles). In the ΔgerE mutant (Fig. 3D, blue circles), the expression of P_cwlC increased slowly from T₈ but was significantly lower than that in the wild-type strain (Fig. 3D, red squares). Our results confirmed that the cwlC gene is expressed in the late sporulation stage in a SigK-dependent manner and is also positively regulated by GerE.

Deletion of the cwlC gene completely blocked mother cell lysis.

Cell wall hydrolases play various roles in cell morphology in different growth stages. We were interested in characterizing the function of CwlC in B. thuringiensis HD73. We constructed a cwlC deletion mutant, HD(ΔcwlC), by replacing the cwlC coding sequence with the kanamycin (Kan) resistance gene kan (Fig. 3A). The deletion of cwlC did not affect the vegetative growth of B. thuringiensis HD73 cells, as shown by the growth curve (data not shown). We next examined the morphology of the wild-type and mutant cells grown in SSM to different growth phases by bright-phase microscopy. Our results revealed that the cell morphology of HD(ΔcwlC) at T₀ and T₁₆ was indistinguishable from that of wild-type HD73 grown to the same time points (Fig. 4A). For instance, at T₁₆, mother cells of both strains contained endospores (Fig. 4A). However, at T₂₄, while a majority of the mother cells of wild-type strain HD73 had autolyzed and mature spores were released, the HD(ΔcwlC) mutant showed little mother cell lysis (<5%, Fig. 4A). We further complemented the ΔcwlC mutant strain with the wild-type cwlC gene controlled by its own promoter and integrated into the pHT315 plasmid. Cells of complemented strain HD(ΔcwlC::cwlC) began to autolyze at T₁₆, and virtually all of the spores were released at T₂₄, similar to the wild-type strain (Fig. 4A).

FIG 4.

Observation of mother cell lysis by optical microscopy. (A) Lysis of B. thuringiensis HD73 (wild-type strain), HD(ΔcwlC) mutant, and genetically complemented HD(ΔcwlC::cwlC) mother cells was observed by optical microscopy at T₀, T₁₆, and T₂₄ after incubation in SSM at 30°C with shaking at 220 rpm. Scale bars, 10 μm. (B) Lysis of the HD(ΔcwlC) mutant mother cells was also observed by optical microscopy at days 3, 7, and 15 after incubation in SSM at 30°C with shaking at 220 rpm. (C) Mother cell lysis of the genetically complemented strains with wild-type CwlC [HD(ΔcwlC::cwlC)] and reconstituted CwlC [HD(ΔcwlC::cwlC^E24A), HD(ΔcwlC::cwlC^E140A), and HD(ΔcwlC::cwlC^E24A/E140A)] was observed by optical microscopy at T₂₄ after incubation in SSM at 30°C with shaking at 220 rpm. Scale bars, 10 μm.

To further examine the mother cell lysis phenotype of the HD(ΔcwlC) mutant, we continued to grow the mutant cells in SSM for 15 days. Surprisingly, the mother cells of the mutant did not autolyze, even after 15 days of growth, and most of the cells were found to harbor encapsulated mature spores (Fig. 4B). To further test whether those encapsulated mature spores are heat resistant, ΔcwlC mutant cells, after growing for 15 days in SSM, were serially diluted with Luria-Bertani (LB) medium. The diluted cell samples were heated in a 65°C water bath for 20 min, plated on LB agar plates, and incubated at 30°C for 12 h. We found that these 15-day-old heat-treated encapsulated spores could regrow on LB agar plates, indicating that these cells could start a vegetative cycle again because of the mature spores encapsulated inside the mother cells.

Finally, we also tested the mother cell lysis phenotypes of the three reconstituted CwlC complementary strains with point mutations (Table 2). We observed that at T₂₄, the reconstituted CwlC complementary strains (E24A, E140A, and E24A/E140A) lost the ability to autolyze (Fig. 4C), showing the same phenotype seen in the ΔcwlC deletion mutant. This result was consistent with our conclusion that CwlC uses the two conserved glutamic acids in the MurNAc-LAA domain as the catalytic residues.

TABLE 2.

Strains and plasmids used in this study

Strain or plasmid	Characteristic(s)	Reference(s) or source
E. coli strains
TG1	Δ(lac-proAB) supE thi hsd-5 (F′ traD36 proA+ proB+ lacIq lacZΔM15)	28
SCS110	F− dam-13::Tn9 dcm-6 hsdM-hsdR recF143 zjj-202::Tn10 galK2 galT22 ara14 pacY1 xyl-5 leuB6 thi-1	28
BL21(DE3)	F− dcm ompT hsdS (rB−mB−) galλ(DE3)	10
BL21(pETcwlC)	BL21(DE3) with pETcwlC plasmid	This study
BL21(pETcwlC-M1)	BL21(DE3) with pETcwlC-M1 plasmid	This study
BL21(pETcwlC-M2)	BL21(DE3) with pETcwlC-M2 plasmid	This study
BL21(pETcwlC-M12)	BL21(DE3) with pETcwlC-M12 plasmid	This study
B. thuringiensis strains
HD73	Wild type containing cry1Ac gene	29, 30
HD(ΔsigK)	HD73 ΔsigK mutant	41
HD(ΔgerE)	HD73 ΔgerE mutant	This study
HD(ΔcwlC)	HD73 ΔcwlC mutant	This study
HD(PcwlC-lacZ)	HD73 strain containing plasmid pHTPcwlC	This study
HDΔsigK(PcwlC-lacZ)	ΔsigK mutant containing plasmid pHTPcwlC	This study
HDΔgerE(PcwlC-lacZ)	ΔgerE mutant containing plasmid pHTPcwlC	This study
HD(ΔcwlC::cwlC)	HD(ΔcwlC) genetic complementation strain carrying pHTHFcwlC plasmid; Eryr	This study
HD(ΔcwlC::cwlCE24A)	HD(ΔcwlC) genetic complementation strain carrying pHTHFcwlCE24A plasmid; Eryr	This study
HD(ΔcwlC::cwlCE140A)	HD(ΔcwlC) genetic complementation strain carrying pHTHFcwlCE140A plasmid; Eryr	This study
HD(ΔcwlC::cwlCE24A, E140A)	HD(ΔcwlC) genetic complementation strain carrying pHTHFcwlCE24A, E140A plasmid; Eryr	This study
Plasmids
pHT304-18Z	Promoterless lacZ vector, Eryr, Ampr, 9.7 kb	36
pET-21b	Expression vector, Ampr, 5.4 kb	Novagen
pHT315	B. thuringiensis-E. coli shuttle vector	35
pRN5101	Temperature-sensitive plasmid, 8.0 kb	34
pHTPcwlC	pHT304-18Z carrying PcwlC, Ampr, Eryr	This study
pETcwlC	pET-21b containing cwlC gene, Ampr	This study
pETcwlC-M1	pET-21b containing cwlCE24A gene, Ampr	This study
pETcwlC-M2	pET-21b containing cwlCE140A gene, Ampr	This study
pETcwlC-M12	pET-21b containing cwlCE24A, E140A gene, Ampr	This study
pRN5101ΩcwlC	pRN5101 carrying partial cwlC deletion gene	This study
pHTHFcwlC	pHT315 containing PcwlC-cwlC	This study
pHTHFcwlCE24A	pHT315 containing PcwlC-cwlCE24A	This study
pHTHFcwlCE140A	pHT315 containing PcwlC-cwlCE140A	This study
pHTHFcwlCE24A, E140A	pHT315 containing PcwlC-cwlCE24A, E140A	This study

Deletion of the cwlC gene did not impact the sporulation frequency, Cry protein production, or insecticidal activity.

Previously, it was shown that in the B. subtilis cell wall, hydrolases had no effect on the sporulation frequency (10). We were curious to know whether deletion of the cwlC gene could impact sporulation in B. thuringiensis. Since B. thuringiensis functions as a typical insecticidal agent by producing Cry proteins, we also decided to test the potential impact of cwlC deletion on Cry protein production, as well as insecticidal activity. We assessed the sporulation frequency and Cry protein production of the wild-type, deletion mutant, and complemented strains in late sporulation phase (T₂₄). The sporulation frequencies of wild-type HD73 (Fig. 5A, blue column), the cwlC mutant HD(ΔcwlC) (Fig. 5A, orange column), and complemented strain HD(ΔcwlC::cwlC) (Fig. 5A, yellow column) showed little difference, indicating that blocking mother cell lysis did not impact the formation of mature, heat-resistant spores. The abundance of Cry proteins was determined by SDS-PAGE after most of the Cry proteins and spores were released from the mother cells of wild-type strain HD73. The HD(ΔcwlC) mutant produced Cry proteins of ∼130 kDa, similar to those found in wild-type cells (Fig. 5B). The abundance of Cry proteins from HD(ΔcwlC) cells was also comparable to that of Cry proteins from wild-type cells. On the basis of the results described above, we concluded that deletion of cwlC had no effect on Cry protein production in B. thuringiensis. To determine the insecticidal activity of the cwlC deletion mutant HD(ΔcwlC), second-instar Plutella xylostella larvae were fed cabbage pretreated with a suspension of a mixture of HD73 spores and crystals and with sporangium from cwlC deletion mutant HD(ΔcwlC). The LC₅₀s (50% lethal concentrations) of HD73 and HD(ΔcwlC) for P. xylostella were 22.28 and 25.46 μg of protein/ml, respectively (Table 3), which are not significantly different, suggesting that the toxicity of HD(ΔcwlC) for P. xylostella is similar to that of wild-type strain HD73. Therefore, we concluded that deletion of the cwlC gene had no effect on the insecticidal activity of B. thuringiensis.

FIG 5.

Comparison of the sporulation frequency and crystal protein production of wild-type HD73 and the HD(ΔcwlC) mutant. (A) Deletion of the cwlC gene did not impact the sporulation frequency. The error bars represent standard deviations. The results are representative of three independent assays. (B) SDS-PAGE showing that deletion of the cwlC gene does not affect Cry1Ac crystal protein production. Lane M, molecular size markers.

TABLE 3.

Insecticidal activities of B. thuringiensis strains against P. xylostella

Strain	LC50 (μg of protein/ml)	95% confidence interval
HD73	22.28	15.43–37.02
HD(ΔcwlC)	25.46	19.09–37.73

DISCUSSION

The main finding of this study is the characterization of CwlC, a new N-acetylmuramoyl-l-alanine amidase (MurNAc-LAA) of B. thuringiensis. CwlC of B. thuringiensis is a novel cell wall hydrolase that plays an essential role in Bt mother cell lysis. In B. subtilis, none of the single autolysin gene deletions affected mother cell lysis because of the functional redundancy of those lytic enzymes (10, 14). In B. thuringiensis, it was previously shown that deletion of the cwlB gene delayed but did not block mother cell lysis (18). In this study, we revealed that disruption of the cwlC gene completely blocked B. thuringiensis mother cell lysis (Fig. 4). Our findings suggest that the mechanism underlying mother cell lysis may differ between B. subtilis and B. thuringiensis, at least regarding the participation of functionally redundant lytic enzymes.

The three major cell wall hydrolases known to be involved in mother cell lysis in B. subtilis are CwlB (LytC) (12), CwlC (13), and CwlH (10), all of which are MurNAc-LAAs. The CwlB autolysin of B. thuringiensis is a MurNAc-LAA (18). Those proteins contain a highly conserved MurNAc-LAA domain (18). Here, we revealed a new potential autolysin in B. thuringiensis, named CwlC. CwlC contains an N-terminal MurNAc-LAA domain and a C-terminal amidase02 domain (Fig. 1A). This protein shows relatively low sequence identity with CwlB of B. thuringiensis and CwlB and CwlC of B. subtilis (12, 9, and 21%, respectively; data not shown). It was reported that the C-terminal boundary of the catalytic domain of B. subtilis CwlC is located between amino acid positions 161 and 176 and the two conserved critical catalytic amino acid residues are glutamic acids E24 and E141 (19). CwlC of B. thuringiensis also contains the two critical glutamic acid residues at the corresponding positions (E24, E140) (Fig. 1B). We confirmed that E24 and E140 are the critical catalytic residues in CwlC of B. thuringiensis (Fig. 2C and 4C). This further supports the notion that CwlC is a MurNAc-LAA. By searching the NCBI database, homologs of the CwlC protein were found in the B. cereus group, often with >90% amino acid sequence identity (Fig. S2). This implies that CwlC may be an essential cell wall hydrolase distributed among the members of the B. cereus group.

B. thuringiensis has been widely used as a biological pesticide, relying on produced Cry proteins to kill its insect larval host (23). When B. thuringiensis is applied in the field, notable problems are its instability and short persistence because of inactivation of Cry proteins by environmental factors such as UV sunlight, desiccation, and high temperature (24). In previous studies, it was shown that a sigK deletion mutant (ΔsigK) failed to trigger mother cell lysis and release crystals. When the recombinant Cry expression vector was introduced into the ΔsigK mutant, mature crystals were produced, but they were encapsulated in B. thuringiensis mother cells, which led to increased insecticidal activity against a broad range of pests (4). However, since most B. thuringiensis cry genes are classified as sporulation-dependent genes whose transcription is controlled by σ^E, σ^K, or both (5), the biggest concern with deletion of the sigK gene is that the expression of some cry genes during late sporulation stages may decrease significantly in the ΔsigK mutant strain (5, 6, 18, 25). In this study, we generated similar encapsulated crystal proteins in a cwlC deletion strain. More importantly, deletion of cwlC completely blocked mother cell lysis without impacting sporulation frequency, crystal production, or insecticidal activity against P. xylostella under laboratory conditions. Compared with ΔsigK, deletion of the cwlC gene provides a more promising solution to protect crystal proteins, with the advantage that cwlC deletion does not affect crystal production or insecticidal activity. In the engineered B. thuringiensis ΔcwlC mutant strain, it is possible that the crystal proteins are less sensitive to UV radiation and thus have an extended duration of activity, as reported for sigK deletion mutants (3, 4, 26). Further investigation may confirm this hypothesis.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

The strains and plasmids used in this study are summarized in Table 2. E. coli strains TG1 and BL21 were used as hosts for molecular cloning and protein expression, respectively. E. coli SCS110 (also called ET12567) was used to generate unmethylated plasmid DNA for transformation into B. thuringiensis cells (27, 28). All E. coli strains were grown at 37°C in LB medium or on LB agar plates supplemented with either chloramphenicol (5 μg/ml) or ampicillin (Amp; 100 μg/ml) when required. B. thuringiensis HD73 was used as the recipient strain to monitor gene transcriptional activity and manipulate gene cloning of B. thuringiensis (29, 30). HD73 and its derivatives were routinely grown at 30°C in LB broth or on LB agar plates supplemented with either erythromycin (Ery; 5 μg/ml) or Kan (100 μg/ml) when required. SSM [8 g of nutrient broth, 0.12 g of MgSO₄, 1 g of KCl, 0.5 mM NaOH, 1 mM Ca(NO₃)₂, 0.01 μM MnCl₂, and 1 μM FeSO₄ per liter of broth] was used for sporulation assays (31).

DNA manipulation.

PCR was performed with Taq DNA polymerase (BioMed, Beijing, China) and PrimeSTAR HS DNA polymerase (TaKaRa Biotechnology Corporation, Beijing, China). The amplified fragments were purified with the AxyPrep PCR cleanup kit (Axygen Biotechnology Corporation, Beijing, China). A single B. thuringiensis colony was suspended in 20 μl of double-distilled water (ddH₂O), boiled for 10 min at 100°C, and then centrifuged at 12,000 × g for 1 min, and the supernatant was used as the template DNA for PCR assays. Restriction enzymes and T4 DNA ligase (TaKaRa Biotechnology Corporation, Beijing, China) were used in accordance with the manufacturer's instructions. Oligonucleotides were synthesized by Sangon Biotech (Beijing, China). Plasmid DNA was extracted from E. coli cells with the AxyPrep Plasmid Miniprep kit (Axygen Biotechnology Corporation, Beijing, China). All of the plasmids constructed were validated by DNA sequencing (BGI, Beijing, China). Plasmids were introduced into E. coli and B. thuringiensis cells by chemical transformation (32) and electroporation (33), respectively. All of the oligonucleotides used in this study are listed in Table 1.

Strain construction.

Long flanking PCR mutagenesis was employed to construct cwlC deletion mutant HD(ΔcwlC). Briefly, a 709-bp upstream region containing a 27-bp overlap with the 5′ end of cwlC (cwlC fragment A) and a 548-bp downstream region containing a 15-bp overlap with the 3′ end of cwlC (cwlC fragment B) were amplified from B. thuringiensis HD73 genomic DNA with the cwlC-a/cwlC-b and cwlC-c/cwlC-d primer sets, respectively. A 1,473-bp Kan resistance gene (kan) fragment including the P_aphA3 promoter was PCR amplified with the pDG780 plasmid (28) and the CKm-a/CKm-b primer set. Subsequently, long flanking PCR was performed with cwlC fragment A, the kan fragment, and cwlC fragment B as templates and the cwlC-a/cwlC-d primer set to generate a long (2,730-bp) fragment in that order. The resulting DNA fragment was doubly digested with BamHI and SalI and cloned into Ery-resistant, temperature-sensitive suicide plasmid pRN5101 (34), which was also doubly digested with BamHI and SalI, thereby generating the pRN5101_cwlC recombinant plasmid. Next, the pRN5101_cwlC plasmid was electroporated into HD73 cells. Transformants obtained on LB agar plates containing Ery and Kan were verified by PCR with the pRN5101-f/pRN5101-r primer set. Allelic replacement of the pRN5101_cwlC plasmid in HD73 cells was achieved as reported previously (18). Mutant strain HD(ΔcwlC), with a deletion of codons 10 to 240 of the cwlC ORF and an insertion of the Kan resistance gene at the same locus, was verified by PCR with the cwlC-a/cwlC-d primer set (Table 1).

To genetically complement ΔcwlC in HD(ΔcwlC), a 1,423-bp fragment containing the cwlC promoter and the ORF (RS15870) was first amplified by PCR with genomic DNA from B. thuringiensis HD73 as the template and the HFcwlC-F/HFcwlC-R primer set. The PCR product was digested with HindIII and SalI and ligated into the pHT315 shuttle vector (35) to generate pHTHFcwlC. The recombinant pHTHFcwlC plasmid was then transformed into HD(ΔcwlC) to generate complemented strain HD(ΔcwlC::cwlC).

To construct a reporter vector for cwlC, the 717-bp promoter region of cwlC (P_cwlC), located upstream of the cwlC gene, was amplified from B. thuringiensis HD73 genomic DNA with the PcwlC-5/PcwlC-3 primer set. The PCR product was digested with PstI and BamHI and then cloned into linearized vector pHT304-18Z, which harbors a promoterless lacZ gene (36). Recombinant plasmid pHTP_cwlC was then transformed into cells of HD73, as well as the sigK and gerE deletion mutants (18, 25), generating HD(P_cwlC-lacZ), HD(ΔsigK)(P_cwlC-lacZ), and HD(ΔgerE)(P_cwlC-lacZ), respectively.

To construct a CwlC-expressing E. coli strain, we first constructed a plasmid that encodes CwlC with a C-terminal 6× histidine tag. A PCR product containing the cwlC gene was amplified from B. thuringiensis HD73 chromosomal DNA with the cwlC-F/cwlC-R primer set. The PCR product was then cloned into the BamHI and SalI sites of pET21b. Recombinant plasmid pETcwlC was then transformed into E. coli BL21(DE3) cells (37) for protein expression.

Site-directed mutagenesis of cwlC (E24A, E140A, and E24A/E140A) was performed with the Fast Mutagenesis System kit (TransGen, Beijing, China). For example, to create the E24A substitution, the CwlC expression plasmid pETcwlC was used as a template for PCR amplification. Two complementary mutagenic DNA oligomers, E24A-F and E24A-R, were designed for the replacement of Glu-24 (codon, GAA) with Ala (codon, GCA). The resulting plasmid, pETcwlC-M1, was directly introduced into BL21(DE3) by transformation, generating strain BL21(pETcwlC-M1). Successful replacement of Glu with Ala at position 24 of cwlC in the above-mentioned strain was confirmed by PCR amplification of the region and DNA sequencing. Construction of the CwlC(E140A)- and CwlC(E24AE140A)-expressing plasmids and that of the CwlC(E24A), CwlC(E140A), and CwlC(E24AE140A) complementary plasmids were similar to what was described above, except that pHTHFcwlC was used for the complementary plasmid construction and mutagenesis primers E140A-F and E140A-R were used in the PCR to introduce the replacement of Glu-140 (codon, GAA) with Ala (codon, GCA). All of the primers used are listed in Table 1.

Purification of recombinant CwlC-His proteins.

The recombinant CwlC-His proteins were purified in accordance with the following procedure. E. coli strain BL21(pETcwlC) was used to express the CwlC-His fusion proteins. Briefly, 300-ml cultures were grown at 37°C in LB broth supplemented with Amp (100 μg/ml) to an OD at 600 nm (OD₆₀₀) of 1.0. Isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 0.5 mΜ, and the cultures were incubated at 18°C for another 12 h with shaking at 150 rpm. Cells were harvested by centrifugation at 10,000 × g for 10 min at 4°C and washed three times with distilled water. Cell pellets were suspended in 50 mM Tris-HCl (pH 8.3) and disrupted on ice by ultrasonication for 5 min at 70% power (CP750 ultrasonic processor; Cole-Parmer Instruments). Cell lysates were centrifuged at 10,000 × g for 10 min to remove cell debris. Supernatant containing the soluble CwlC-His fusion protein was filtered through a 0.45-mm-pore-size membrane filter (Nalgene) and then loaded onto a nickel HiTrap chelating column (1 ml; Pharmacia). This was repeated three times. After protein binding, the column was washed with a 50 mM imidazole solution (50 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl, pH 8.3) to remove nonspecific binding proteins. The CwlC-His protein was eluted from the column with a 250 mM imidazole solution (250 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl, pH 8.3). The purified CwlC-His protein was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (38).

Cell wall preparation.

Cell wall material was prepared as described by Yang et al., with some modifications (18). Briefly, B. thuringiensis HD73 cells were cultured to exponential growth phase in LB medium, harvested by centrifugation at 16,000 × g for 10 min, and washed three times with deionized water. The cell pellet was then resuspended in TK buffer (0.05 M Tris-HCl, 0.05 M KCl, pH 7.0) and disrupted with a BeadBeater (Biospec Products, Inc., Bartlesville, OK) in a 2-ml centrifuge tube containing an appropriate volume of glass beads (0.1-mm diameter). After low-speed centrifugation at 1,000 × g and 4°C for 10 min, the supernatant was transferred to a new Beckman vessel and the crude cell wall was pelleted by high-speed centrifugation at 27,000 × g for 5 min at 4°C. The pellet was suspended in 5 ml of 4% (wt/vol) SDS and boiled for 10 min. After three sequential washes with 1 M NaCl and deionized water, the cell wall sample was stored at −70°C until needed (39, 40).

Cell wall binding ability of CwlC.

The cell wall binding ability of the CwlC protein was determined as described by Nugroho et al. (10). Briefly, purified CwlC-His or SigK-His protein (a control for exclusion of histidine tag interference) was mixed with prepared B. thuringiensis cell wall in distilled water and incubated on ice for 30 min. Meanwhile, CwlC-His protein alone was added to distilled water and incubated on ice for 30 min as another control. All reaction mixtures were pelleted by centrifugation. Proteins in the pellet were separated by electrophoresis (4% polyacrylamide stacking gel and 10% polyacrylamide separating gel), followed by Western blot analysis after transfer of the proteins to a polyvinylidene difluoride membrane. The CwlC-His protein was detected with an anti-His antibody.

Homologous competition assay.

CwlC protein (1 mg) was labeled with EZ-Link Sulfo-NHS-SS-Biotin in accordance with the manufacturer's instructions (Thermo Fisher). A 4.5-μg (1,000 nM) sample of CwlC was added to a mixture of cell wall (50 μl, OD₅₄₀ of 0.3) in ddH₂O in a 160-μl final volume and incubated for 30 min on ice. After that, excess unlabeled CwlC protein (10,000 to 320,000 nM) was added to the above mixture for incubation for another 30 min. The mixture was centrifuged at 4°C for 5 min at 27,000 × g and then washed once with ddH₂O. The pellets containing the cell wall were analyzed by Western blot analysis with Pierce high-sensitivity streptavidin-horseradish peroxidase (Thermo Fisher) to detect the bound biotinylated CwlC.

Cell wall hydrolysis by CwlC.

B. thuringiensis cell wall hydrolysis by CwlC was performed as previously described (10, 18). Briefly, purified CwlC-His protein was added to the prepared B. thuringiensis cell wall suspended in 0.05 M TK buffer (pH 7.0) at a final OD₅₄₀ of 0.3. The mixture was incubated at 37°C. The OD₅₄₀ of the mixture was measured at the designated time point. After hydrolysis, 1 ml of the mixture was centrifuged at 4°C for 5 min at 27,000 × g and then freeze-dried for ∼48 h until the pellet became lyophilized powder to quantify polymeric cell wall degradation.

Total RNA isolation and RT-PCR.

Total RNA was extracted from B. thuringiensis HD73 cells cultured in SSM at T₁₅, and RT-PCR was performed as described previously (41). The primers used for RT-PCR analysis to identify cwlC gene expression are shown in Table 1.

Determination of the transcription start site.

5′ RACE was performed to determine the transcription start site of the cwlC gene in accordance with the manufacturer's instructions (Clontech Laboratories, Inc., TaKaRa Biotechnology Corporation, Beijing, China). Briefly, cDNA was synthesized as follows. A mixture of 1 μl of RNA, 1 μl of 5′ CDS primer A, and 1.75 μl of diethyl pyrocarbonate-treated water was subject to 72°C for 3 min and 42°C for 2 min. To the cDNA mixture we added 1 μl of SMARTer IIA oligonucleotide and 5.25 μl of a mixture of 2 μl of 5× first-strand buffer, 1 μl of dithiothreitol (20 mM), 1 μl of a deoxynucleoside triphosphate mixture (10 mM), 0.25 μl of RNase inhibitor (40 U/μl), and 1 μl of SMARTScribe reverse transcriptase. The reaction mixture was heated to 42°C for 90 min and 72°C for 10 min. The RACE mixture (50 μl) was prepared as follows. A 41.5-μl volume of master mix 2.5 μl of 2× SeqAMP buffer, 1 μl of SeqAMP DNA polymerase, 15.5 μl of ddH₂O, 1 μl of cDNA, 5 μl of 10× universal primer mix, 1 μl of gene-specific primer (cwlCRACE-R), and 1.5 μl of RNase-free water. The reaction mixture was subjected to 25 cycles of 94°C for 30 s, 68°C for 30 s, and 72°C for 3 min. The oligonucleotides used in this study are listed in Table 1.

β-Galactosidase assays.

Two-milliliter samples of cells cultured in SSM were collected at each time point (from T₈ to T₂₀ at 1-h intervals), and cells were centrifuged at 10,000 × g for 2 min to measure β-galactosidase activity in the pelleted cells as described previously (42). β-Galactosidase activity is expressed in Miller units. The data shown are the mean values of at least three independent experiments.

Microscopic analysis and assays of sporulation frequency.

The wild-type strain (HD73) and the cwlC deletion mutant [HD(ΔcwlC)] were cultured in 100 ml of SSM at 30°C with shaking at 220 rpm. One-microliter samples were collected at designated time points (T₀, T₁₆, T₂₄, and days 3, 7, and 15) and centrifuged. The pellets were resuspended in a final volume of 100 μl of deionized water. One microliter of each cell sample was spotted onto the center of a glass slide and covered with a coverslip. Cell samples were analyzed with a BX61 optical microscope (Olympus, Japan). The sporulation frequency was calculated as described previously (18, 20). Briefly, the total number of cells in the sample taken at T₁ was determined. A one-microliter sample was collected at T₂₄ and heated at 65°C for 20 min to kill the vegetative cells, serially diluted, and plated on LB agar plates. The sporulation frequency was defined as the ratio of the number of colonies after heat treatment at T₂₄ to the number of colonies at T₁.

Quantification of Cry1Ac protein production.

The wild-type and cwlC deletion strains of B. thuringiensis were cultured in SSM at 30°C to T₂₄. Subsequently, 2-ml samples of bacterial cells were centrifuged, resuspended in 500 μl of Tris-HCl (50 mM, pH 8.0), and disrupted with a BeadBeater (Biospec Products, Inc.). The supernatant was then mixed with 5× protein-reducing sample buffer (42) and boiled for 5 min for subsequent total-protein quantitation and SDS-PAGE as described previously (18).

Bioassay of insecticidal activity.

Biological assays were performed as described by Zhou et al. (43), by using equivalent bacterial biomass concentrations of the HD73 and cwlC deletion mutant strains. Briefly, B. thuringiensis cells were grown to T₂₄ in 50 ml of fresh SSM at 30°C with shaking (220 rpm), centrifuged at 4°C for 10 min at 8,000 rpm, and then freeze-dried for ∼48 h until the pellets became lyophilized powders. An appropriate volume of ddH₂O was added to all samples to adjust them to equivalent bacterial biomass concentrations (mg/ml). Insecticidal activity was tested by exposing second-instar larvae of the diamondback moth (P. xylostella) to an artificial diet incorporating one of seven dilutions (bacterial lyophilized powder concentrations of 2.5, 5, 7.5, 10, 20, 30, and 60 μg/ml) of each preparation in water (44). A 6-cm-diameter cabbage leaf disc was pretreated with a gradient of bacterial concentrations as described above and then transferred to a new plastic culture dish, after which 30 second-instar larvae were placed in each dish. The surviving larvae were counted after 3 days, and the LC₅₀ was calculated by probit analysis. Each concentration was tested in triplicate.