A Type IX Secretion System Substrate Involved in Crystalline Cellulose Degradation by Affecting Crucial Cellulose Binding Proteins in Cytophaga hutchinsonii

ABSTRACT

Cytophaga hutchinsonii is an abundant soil cellulolytic bacterium that uses a unique cellulose degradation mechanism different from those that involve free cellulases or cellulosomes. Though several proteins have been identified as important for cellulose degradation, the mechanism used by C. hutchinsonii to digest crystalline cellulose remains a mystery. In this study, chu_0922 was identified by insertional mutation and gene deletion as an important gene locus indispensable for crystalline cellulose utilization. Deletion of chu_0922 resulted in defects in crystalline cellulose utilization. The Δ0922 mutant completely lost the ability to grow on crystalline cellulose, even with extended incubation, and selectively utilized the amorphous region of cellulose, leading to increased crystallinity. As a protein secreted by the type IX secretion system (T9SS), CHU_0922 was found to be located on the outer membrane, and the outer membrane localization of CHU_0922 relied on the T9SS. Comparative analysis of the outer membrane proteins revealed that the abundance of several cellulose-binding proteins, including CHU_1276, CHU_1277, and CHU_1279, was reduced in the Δ0922 mutant. Further study showed that CHU_0922 is crucial for the full expression of the gene cluster containing chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C), which is essential for cellulose utilization. Moreover, CHU_0922 is required for the cell surface localization of CHU_3220, a cellulose-binding protein that is essential for crystalline cellulose utilization. Our study provides insights into the complex system that C. hutchinsonii uses to degrade crystalline cellulose.

IMPORTANCE The widespread aerobic cellulolytic bacterium Cytophaga hutchinsonii, belonging to the phylum Bacteroidetes, utilizes a novel mechanism to degrade crystalline cellulose. No genes encoding proteins specialized in loosening or disruption the crystalline structure of cellulose were identified in the genome of C. hutchinsonii, except for chu_3220 and chu_1557. The crystalline cellulose degradation mechanism remains enigmatic. This study identified a new gene locus, chu_0922, encoding a typical T9SS substrate that is essential for crystalline cellulose degradation. Notably, CHU_0922 is crucial for the normal transcription of chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C), which play important roles in the degradation of cellulose. Moreover, CHU_0922 participates in the cell surface localization of CHU_3220. These results demonstrated that CHU_0922 plays a key role in the crystalline cellulose degradation network. Our study will promote the uncovering of the novel cellulose utilization mechanism of C. hutchinsonii.

INTRODUCTION

Cellulose, one of the major components of plant cell walls, is abundant on the earth and is an ideal source of biofuel. Cellulose is insoluble, stable, and resistant to hydrolysis because of the strong hydrogen bonding, weaker hydrophobic interactions, and van der Waals forces between the microfibers, which produce the crystalline structure. Loosening and disruption of the crystalline structure of cellulose constitute the first and key step in its degradation (1–3). Microorganisms play vital roles in the degradation and bioconversion of cellulose using distinct strategies (4). The noncomplexed cellulase system is used by most aerobic cellulolytic microorganisms by secreting an extracellular free-cellulase system. Most anaerobic cellulolytic microorganisms digest cellulose by production of large enzymatic complex cellulosomes on the cell surface (2, 3, 5, 6). Inaccessibility to the crystalline cellulose chain is one of the reasons for the recalcitrance of cellulose. Cellulose-disrupting proteins, including proteins with carbohydrate-binding modules (CBMs), expansins, expansin-like proteins, and lytic polysaccharide monooxygenases (LPMOs), are thought to play important roles in loosening or disruption of the hydrogen-bonding networks in cellulose (3, 7–10). Two cellulolytic bacteria, including Cytophaga hutchinsonii, were proposed to use a third mechanism to degrade cellulose (4).

C. hutchinsonii is a widespread cellulolytic bacterium that belongs to the phylum Bacteroidetes and can efficiently digest crystalline cellulose in a direct-contact manner. C. hutchinsonii does not secrete a free-cellulase system or form cellulosomes (4, 11). Genome analysis of C. hutchinsonii found 18 predicted endoglucanases and five β-glucosidases that may be involved in cellulose utilization (10, 12–14). However, the genome of C. hutchinsonii lacks genes encoding predicted exoglucanases, which play crucial roles in the degradation of crystalline cellulose. No genes encoding homologous proteins involved in loosening or disruption of the hydrogen-bonding networks in cellulose were identified in the genome of C. hutchinsonii, and the predicted endoglucanases lack obvious CBMs (14, 15). The mechanism by which C. hutchinsonii degrades crystalline cellulose remains elusive.

Recently, in addition to the cellulases (10, 12), several proteins were identified as playing crucial roles in cellulose degradation, including the outer membrane proteins CHU_1276 (16) and CHU_1277 (17). Notably, our previous work identified a protein, CHU_3220, involved in the degradation of the crystalline region but not the amorphous region of cellulose (18). Moreover, previous works reported that deletion of the encoding genes of SprP, PorU, GldN, SprA, and SprT, which are the components of the type IX secretion system (T9SS), resulted in defects in cellulose utilization and cell motility (19–22). T9SS is a novel protein secretion system that is widely and exclusively distributed in the Bacteroidetes and plays important roles in degradation of complex biopolymers, gliding motility, and pathogenesis (23, 24). There are at least 147 putative T9SS substrates containing type A or type B C-terminal domains (CTDs) in C. hutchinsonii, the functions of which are largely unknown, except for the endoglucanases (19, 25). Our previous work reported that CHU_3220, a T9SS substrate with a variant T9SS signal, was involved in crystalline cellulose degradation (18, 21, 26). Recently, our group reported another putative T9SS substrate, CHU_1557, with a type B CTD, was required for the disruption of the crystalline cellulose and glucose assimilation (27). However, the functions of the other T9SS substrates have not been reported.

This study identified and characterized a new gene locus, chu_0922, encoding a typical T9SS substrate with a type A CTD that is essential for crystalline cellulose degradation. Further study found that CHU_0922 is essential for the cell surface localization of CHU_3220 and the normal transcription of chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C), which play important roles in the degradation of cellulose. The function of CHU_0922 in cellulose degradation and the subcellular localization of CHU_0922 were further studied.

RESULTS

Identification a gene locus, chu_0922, involved in filter paper degradation through transposon mutagenesis.

In order to investigate the unique cellulose degradation mechanism of C. hutchinsonii, the transposon HimarEM3 was used to identify genes essential for cellulose utilization. Among about 2,000 transposon insertional mutants, we found a mutant, TM155, defective in filter paper degradation (Fig. 1A). Inverse PCR was used to determine the transposon insertion site, which was found to be located within the gene locus chu_0922. To confirm this, a gene deletion mutant of chu_0922 was constructed as previously described (20) (the deletion process is shown in Fig. S1 in the supplemental material). The selected transformants were verified by PCR using primers 0922UF/0922UR and 0922UF/CFXR, and all of them had the expected band sizes (see Fig. S2 in the supplemental material). Like the transposon mutants, the Δ0922 mutant was defective in filter paper utilization (Fig. 1A).

FIG 1.

Cellulose utilization ability of the wild type and mutant. (A) Filter paper degradation assay. Equivalent amounts (3 μL) of cells were spotted on Whatman filter paper on Stanier agar, followed by incubation for 15 days at 30°C. This assay was performed in triplicate using three independent transformants (with the same results), and one representative result is shown. (B and C) Ability to utilize 0.2% RAC (B) and 0.2% Avicel (C). The mean values and SDs from at least three replicates are shown. WT, wild-type strain; TM155, mutant with chu_0922 disrupted by pHimarEM3; Δ0922, chu_0922 deletion mutant; Δ0922-C, mutant with the CTD of chu_0922 deleted; C0922, the Δ0922 mutant complemented with pHB0922.

Bioinformatics analysis revealed that chu_0922 encodes a hypothetical protein with an unknown function. The genes upstream and downstream of chu_0922 are a conserved hypothetical gene and pseudogene, respectively. CHU_0922 consists of 730 amino acids with a molecular mass of 82 kDa. An N-terminal signal peptide containing 21 amino acids was detected in CHU_0922 using SignalP 5.0. No conserved domain was detected in CHU_0922 except for the conserved type A CTD (see Fig. S3 in the supplemental material), revealing that CHU_0922 may be a substrate of the T9SS. To investigate the function of CTD, it was deleted as described in Materials and Methods, and the mutant was designated Δ0922-C. Though cells of Δ0922-C could turn the filter paper yellow, they could not completely digest the filter paper (Fig. 1A). Complementation of the Δ0922 mutant with pHB0922, which carries the wild-type chu_0922 under the control of the constitutive promoter, restored the ability of the Δ0922 mutant to digest filter paper (Fig. 1A). These results indicated that CHU_0922 is required for the degradation of filter paper.

CHU_0922 is essential for the degradation of the crystalline structure of cellulose.

To further investigate the function of CHU_0922, the ability to utilize cellulose with different degrees of crystallinity was determined in the Δ0922 mutant and the wild type. The Δ0922 mutant exhibited a lag phase of 3 days in utilization of the regenerated amorphous cellulose (RAC) (Fig. 1B). Moreover, the Δ0922 mutant showed an obvious defect in utilization of Avicel, and the maximum biomass was 37% of that of the wild type (Fig. 1C), which was consistent with the defect in degradation of filter paper. The complemented strain of the Δ0922 mutant regained the ability to normally degrade RAC and Avicel (Fig. 1B and C). To study changes in the degree of crystallinity of Avicel, cells of the wild type and Δ0922 mutant were incubated with Avicel for 48 h, and then the crystallinity of Avicel was detected by X-ray diffraction (XRD) as previously described (18). It was found that the crystallinity of Avicel was reduced from 66.3% to 62.4% after treatment with cells of the wild type (see Fig. S4A in the supplemental material), while it was increased from 67.2% to 74.0% after treatment with cells of the Δ0922 mutant (see Fig. S4B). In addition, the surface morphology of Avicel treated with cells of the wild type and Δ0922 mutant was observed by a scanning electron microscope (SEM). The surface of the untreated Avicel was rough (Fig. 2A). It turned smooth and flat after incubation with cells of the wild type (Fig. 2B), while it became rougher after being treated with cells of the Δ0922 mutant (Fig. 2C). These results demonstrated that cells of the Δ0922 mutant selectively utilize the amorphous region of Avicel, and CHU_0922 is essential for the degradation of the crystalline structure of cellulose.

FIG 2.

SEM observation the surface morphology of Avicel incubated with cells of the wild type and the Δ0922 mutant. (A) Avicel PH101 incubated without cells of C. hutchinsonii. (B) Avicel PH101 incubated with cells of the wild type for 48 h. (C) Avicel PH101 incubated with cells of the Δ0922 mutant for 48 h. Bars, 10 μm. This assay was performed in triplicate using three independent transformants (with the same results), and one representative result is shown.

Deletion of chu_0922 attenuated the cellulase activity.

Given the cell-associated cellulolytic activity of C. hutchinsonii, the cell surface and intracellular cellulase activities of the wild type and Δ0922 mutant were measured. The results showed that the cell surface endoglucanase activity of the Δ0922 mutant was reduced by 19%, and the intracellular endoglucanase activity of the Δ0922 mutant was reduced by 37%, compared with that of the wild type (Fig. 3A). Notably, the cell surface β-glucosidase activity of the Δ0922 mutant was decreased by 66% and its intracellular β-glucosidase activity was decreased by 23% compared with those of the wild type (Fig. 3B). Complementation of the Δ0922 mutant with pHB0922 restored the endoglucanase activity and β-glucosidase activity (Fig. 3A and B). These results indicated that deletion of chu_0922 obviously influenced the β-glucosidase activity.

FIG 3.

Cellulase activities of the wild type and Δ0922 mutant. (A) Endoglucanase activity of the wild type and the Δ0922 mutant. Endoglucanase activity was determined using sodium carboxymethyl cellulose (CMC-Na) as the substrate, and the concentration of the reducing ends was measured using the 3,5-dinitrosalicylic acid procedure. (B) β-Glucosidase activity of the wild type and the Δ0922 mutant. β-Glucosidase activity was determined using p-nitrophenyl-β-d-glucopyranoside (pNPG) as the substrate, and the released p-nitrophenol was determined by absorption at 410 nm. ***, P < 0.001. The data are the averages and SDs from three independent experiments.

CHU_0922 significantly affected the expression level of the cellulose-binding proteins CHU_1276, CHU_1277, and CHU_1279.

Previous studies reported that the outer membrane integrity is crucial for cellulose degradation in C. hutchinsonii, and some outer membrane proteins can bind to cellulose (15–17). We investigated changes in the outer membrane protein profile and outer membrane cellulose-binding protein profile of the Δ0922 mutant compared with that of the wild type. The outer membrane proteins were extracted as described in Materials and Methods and then separated by SDS-PAGE. As shown in Fig. 4A, at least three outer membrane proteins were differentially expressed in the Δ0922 mutant compared with the wild type and were identified by mass spectrometry. The results showed that CHU_1276, CHU_1277, and CHU_1279 were decreased in abundance in the outer membrane proteins of the Δ0922 mutant (Table 1). Moreover, the outer membrane cellulose-binding protein profile of the Δ0922 mutant was also different from that of the wild type (Fig. S5 in the supplemental material), and the cellulose-binding proteins expressed differentially between the wild type and Δ0922 mutant were also identified by mass spectrometry. The results demonstrated that CHU_1276, CHU_1277, and CHU_1279 are cellulose-binding proteins and decreased in abundance in the outer membrane cellulose-binding proteins of the Δ0922 mutant. The gene cluster containing chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C) is important for cellulose utilization (16, 17, 28), and these genes are transcribed in the same direction. To investigate the reason for the decreased abundance of CHU_1276, CHU_1277, and CHU_1279 in outer membrane proteins of the Δ0922 mutant, we determined the transcription level of these genes by real-time quantitative PCR (qPCR). The results showed that the transcription levels of chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C) were all significantly decreased, especially chu_1277, in the Δ0922 mutant compared with those in the wild type (Fig. 4B). These results demonstrated that CHU_0922 is crucial for the normal expression of the gene cluster containing chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C).

FIG 4.

Deletion of chu_0922 affected the expression of the outer membrane proteins essential for cellulose degradation. (A) SDS-PAGE of the outer membrane proteins of the wild type and the Δ0922 mutant. Loading samples were normalized by equal biomass. The differential proteins indicated by both asterisks and numbers were identified by mass spectrometry. (B) qPCR detection of the transcription levels of chu_1276 to chu_1280 in the Δ0922 mutant compared with those in the wild type. The transcription levels of these genes in the wild type were normalized to 1. The mean values and SDs from at least three replicates are shown.

TABLE 1.

Identification of outer membrane proteins^a differentially expressed between the wild type and the Δ0922 mutant by LC-MS/MS

Band	Locus	MW (kDa)b	Abundance decrease (%)c
1	CHU_1276	85.2	44
2	CHU_1277	65.4	72
3	CHU_1279	42.4	21

^aFor all loci, the predicted function was “hypothetical protein.”

^bMolecular mass of the primary product of translation, including the predicted signal peptide.

^cProteins that decreased in abundance in the outer membrane proteins of the Δ0922 mutant compared with the wild type were quantitatively analyzed by gray analysis using Image J.

CHU_0922 is essential for the outer membrane localization of CHU_3220.

CHU_3220, a cell surface cellulose-binding protein, is essential for the degradation of crystalline cellulose (18). Because the Δ0922 mutant was defective in utilization of the crystalline structure of cellulose, we investigated the localization of CHU_3220 in the Δ0922 mutant. CHU_3220 was detected in the whole-cell and outer membrane proteins of the wild type (Fig. 5A). However, the abundance of CHU_3220 was obviously reduced in the whole-cell proteins of the Δ0922 mutant, and CHU_3220 could not be detected in the outer membrane proteins of the Δ0922 mutant (Fig. 5A). Moreover, it was found that CHU_3220 was accumulated in the periplasmic space of the Δ0922 mutant (Fig. 5A). To investigate whether the reduced abundance of CHU_3220 in the whole-cell and outer membrane proteins of the Δ0922 mutant was due to the compromised transcription of chu_3220, we detected the transcription level of chu_3220 in the wild type and the Δ0922 mutant by real-time qPCR. The result showed that the transcription level of chu_3220 in the Δ0922 mutant was slightly increased compared with that of the wild type (Fig. 5B). Our previous work reported that T9SS is responsible for the secretion of CHU_3220 (21). However, the transcription level of most of the core genes of T9SS in the Δ0922 mutant was increased to some degree compared with that of the wild type (see Fig. S6A in the supplemental material), and the typical extracellular T9SS substrate of C. hutchinsonii CHU_0344 could be normally secreted in the Δ0922 mutant (see Fig. S6B in the supplemental material). These results demonstrated that the outer membrane localization of CHU_3220 depends on the presence of CHU_0922.

FIG 5.

Expression of CHU_3220 in the wild type and the Δ0922 mutant. (A) Abundance of CHU_3220 in the whole-cell proteins, outer membrane proteins, and periplasmic proteins of the wild type, the Δ0922 mutant, and the Δ3220 mutant detected by Western blotting. Loading samples were normalized by equal biomass. All measurements were carried out in triplicate with the same results. (B) qPCR detection of the transcription level of chu_3220 in the wild type and the Δ0922 mutant. The mean values and SDs from at least three replicates are shown.

CHU_0922 is a T9SS substrate located on the outer membrane.

Bioinformatics analysis revealed that CHU_0922 contains the N-terminal signal peptide and conserved type A CTD, which are characteristics of the T9SS substrates. The above result demonstrated that the CTD was important for the normal function of CHU_0922. To further verify whether CHU_0922 is secreted by the T9SS and to investigate its subcellular localization, we constructed strains in which the Twin-Strep tag was inserted between the signal peptide and the mature region of CHU_0922 in the wild type and the ΔgldN mutant (a T9SS mutant), as described in Materials and Methods, yielding the WT-0922Strep mutant and the ΔgldN-0922Strep mutant. GldN is a core component of the T9SS and is essential for the secretion of the T9SS substrates (21). The WT-0922Strep mutant retained the ability to efficiently digest cellulose (Fig. S7 in the supplemental material), implying that the insertion of the Twin-Strep tag did not affect the function of CHU_0922. The outer membrane and periplasmic proteins were extracted from the WT-0922Strep mutant, ΔgldN-0922Strep mutant, and wild type. The localization of CHU_0922 was detected by Western blotting using an antibody against the Strep tag. As shown in Fig. 6, CHU_0922 was detected in both of the outer membrane and periplasmic proteins extracted from the WT-0922Strep mutant. The molecular weight of the mature CHU_0922 was above 100 kDa, which was higher than the theoretical molecular weight of 82 kDa, suggesting that CHU_0922 might be modified by an unknown mechanism. However, CHU_0922 could be detected only in the periplasmic proteins extracted from the ΔgldN-0922Strep mutant. The outer membrane and periplasmic proteins of the wild type were used as the negative control. The results demonstrated that CHU_0922 is a substrate of the T9SS and that the mature CHU_0922 is located on the outer membrane.

FIG 6.

Subcellular localization of CHU_0922 detected by Western blotting using the antibody against the Strep tag. Cell fractions were isolated from the outer membrane and periplasmic space of the WT-0922Strep mutant, the ΔgldN-0922Strep mutant, and the wild type. Loading samples were normalized by equal biomass. P, periplasmic proteins; OM, outer membrane proteins. All measurements were carried out in triplicate with the same results.

Purification of CHU_0922 and identification of the interacting protein of CHU_0922.

In order to investigate the proteins interacting with CHU_0922, we purified CHU_0922 and its associated proteins from cells of the WT-0922Strep mutant using high-performance Strep-Tactin Sepharose resin as described in Materials and Methods and performed Western blotting using the antibody against the Strep tag and CHU_3220, respectively. The proteins purified from cells of the wild type were used as the negative control. As shown in Fig. S8A, the mature CHU_0922 was detected in proteins isolated from the WT-0922Strep and was absent in proteins isolated from the wild type. The result demonstrated that CHU_0922 could be purified from the WT-0922Strep mutant. Western blotting using the antibody against CHU_3220 did not detect any band in proteins purified from the WT-0922Strep mutant and wild type (Fig. S8B). Moreover, the proteins purified from the WT-0922Strep mutant and wild type were identified by liquid chromatography-tandem mass spectrometry (LC-MS/MS) to investigate the interacting protein of CHU_0922. The result demonstrated that CHU_0922 was abundant in proteins isolated from the WT-0922Strep mutant. No peptide spectrum of CHU_0922 was detected in proteins purified from the wild type. However, two peptide spectrums of CHU_3220 were identified in proteins purified from the WT-0922Strep mutant, and one peptide spectrum of CHU_3220 were identified in proteins isolated from the wild type (Data Set S1 in the supplemental material). These results indicated that there is no obvious interaction between CHU_0922 and CHU_3220.

DISCUSSION

Limited access to the crystalline structure of cellulose is the major barrier faced by cellulases in the bioconversion of cellulose to ethanol. Nonhydrolytic proteins, such as CBMs, expansins, swollenins, and LPMOs, can disrupt or loosen the crystalline cellulose and increase the cellulose surface area, which makes it more accessible to the cellulases (3). C. hutchinsonii can efficiently digest crystalline cellulose, but it contains none of these nonhydrolytic proteins, and the predicted endoglucanases lack obvious CBMs (14). The mechanism of disruption and digestion of the crystalline structure of cellulose remains largely unknown in C. hutchinsonii. As proposed by Wilson, it is possible that cellulose is bound and disrupted by a protein complex on the cell surface of C. hutchinsonii and then the individual cellulose molecules are transported into the periplasmic space, where they are degraded by endoglucanases and β-glucosidases (4). Wang et al. proposed that there might be a protein machine on the cell surface of C. hutchinsonii that can disrupt the crystalline structure of cellulose in an intact cell dependent manner (18). CHU_3220 was identified to be essential for the degradation of crystalline cellulose.

Here, we identified another protein, CHU_0922, required for the disruption of crystalline cellulose. However, the relationship between the two proteins essential for crystalline cellulose degradation is unclear. Through Western blot analysis we found that CHU_3220 was missing on the outer membrane proteins of the Δ0922 mutant, and CHU_3220 was accumulated in the periplasmic space of the Δ0922 mutant. Deletion of chu_0922 did not affect the transcription level of chu_3220. This phenomenon is similar to that of the ΔgldN mutant, in which the defect in secretion of CHU_3220 to the cell surface results in its accumulation in the periplasmic space of the ΔgldN mutant (21). However, the transcription levels of the core genes of T9SS in the Δ0922 mutant were almost consistent with that of the wild type, and the typical extracellular T9SS substrate of C. hutchinsonii, CHU_0344, could be normally secreted in the Δ0922 mutant (20, 22). These results demonstrated that CHU_0922 participated in the cell surface localization of CHU_3220 but not by affecting the function of T9SS. However, no obvious protein interactions between CHU_0922 and CHU_3220 were identified by Western blotting and LC-MS/MS using the purified CHU_0922 and its associated proteins, indicating that CHU_0922 indirectly affects the outer membrane localization of CHU_3220. Further studies focusing on the detailed relationship between CHU_0922 and CHU_3220 will uncover the working mechanism of the crystalline cellulose degradation machine of C. hutchinsonii.

Comparative analysis of the outer membrane proteins of the wild type and the Δ0922 mutant revealed that the abundance of several cellulose-binding proteins, including CHU_1276, CHU_1277, and CHU_1279, was reduced in the outer membrane of the Δ0922 mutant. Further study showed that deletion of chu_0922 dramatically reduced the transcription level of chu_1276, chu_1277, chu_1278, chu_1279, and chu_1280 (cel9C), indicating that CHU_0922 is crucial for the transcription of these genes. It was reported that chu_1276, chu_1277, chu_1278, and chu_1279 form a predicted operon that is located immediately upstream of cel9C and is essential for cellulose utilization (15). Cel9C is a periplasmic endoglucanase that is crucial for cellulose utilization (10). Previous studies reported that CHU_1276 and CHU_1277 are outer membrane cellulose-binding proteins and are required for cellulose degradation (16, 17). CHU_1278 and CHU_1279 were also proved to be indispensable for cellulose utilization (our unpublished data). The organization of these genes is reminiscent of the polysaccharide utilization loci (PUL) except for the absence of susC-like and susD-like genes (29–32). In many members of the phylum Bacteroidetes, SusC-like and SusD-like proteins can bind oligosaccharides and transport them into the periplasmic space for further digestion. However, the only C. hutchinsonii SusC-like and SusD-like proteins are not required for efficient cellulose utilization (33). The gene cluster containing chu_1276, chu_1277, chu_1278, chu_1279, and cel9C is speculated to be a novel type of PUL that plays a vital role in cellulose degradation in C. hutchinsonii (15, 28). However, the regulatory system of this putative PUL is unclear. Our study found that CHU_0922 is indispensable for the normal transcription of this new predicted PUL.

Bioinformatics analysis revealed that CHU_0922 contains the N-terminal signal peptide and the conserved type A CTD. Moreover, the CTD is essential for the normal function of CHU_0922. This study showed that the mature CHU_0922 is located on the outer membrane, which requires the normal function of T9SS, proving that CHU_0922 is a substrate of the T9SS. CHU_0922 is a T9SS substrate with a type A CTD that has been proved to be essential for crystalline cellulose degradation. The different phenotypes of the Δ0922 mutant and Δ0922-C mutant in filter paper degradation might result from the trace leakage of CHU_0922 from the periplasmic space of the Δ0922-C mutant; thus, cells of the Δ0922-C mutant could turn the filter paper yellow. Furthermore, this study revealed that CHU_0922 not only participates in the cell surface localization of CHU_3220 but also influences the expression of the putative PUL, indicating that CHU_0922 plays a key role in the crystalline cellulose degradation network. Further studies focusing on the detailed mechanism will undoubtedly contribute to uncovering the novel mechanism of crystalline cellulose degradation of C. hutchinsonii.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

All strains and plasmids used in this study are listed in Table 2. C. hutchinsonii ATCC 33406 was used as the wild-type strain and was cultured in PY6 medium (6 g/L peptone, 0.5 g/L yeast extract, 4 g/L glucose; pH 7.3), Stanier medium (10 mM KNO₃, 4.4 mM K₂HPO₄, 0.8 mM MgSO₄, 0.07 mM FeCl₃, 0.9 mM CaCl₂, 2 g/L glucose; pH 7.3), and PYT medium (6 g/L peptone, 0.5 g/L yeast extract, 0.9 mM CaCl₂, 0.8 mM MgSO₄, 4 g/L glucose; pH 7.3). Liquid cultures were incubated with shaking at 160 rpm at 30°C. Solid media with 10 g/L agar were incubated in a constant-temperature incubator at 30°C. Escherichia coli strains were cultivated in Luria-Bertani medium at 37°C with shaking at 170 rpm. Antibiotics were used at the following concentrations when required: ampicillin, 100 μg/mL; cefoxitin, 15 μg/mL; and erythromycin, 30 μg/mL. The primers used in this study are listed in Table S1.

TABLE 2.

Strains and plasmids used in this study

Strain or plasmid	Descriptiona	Source or reference
E. coli strain DH5α	Strain used for gene cloning	TaKaRa
C. hutchinsonii strains
ATCC 33406	Wild type	ATCC
TM155	Transposon insertion in chu_0922	This study
Δ0922 mutant	chu_0922 deleted	This study
Δ0922-C mutant	CTD of chu_0922 deleted	This study
Δ3220 mutant	chu_3220 deleted	18
C0922	Complementation of the Δ0922 mutant with pHB0922	This study
ΔgldN mutant	chu_0174 deleted	21
WT-0922Strep	The Twin-Strep tag was inserted between the signal peptide and the mature CHU_0922 in the wild type	This study
ΔgldN-0922Strep	The Twin-Strep tag was inserted between the signal peptide and the mature CHU_0922 in the ΔgldN mutant	This study
Plasmids
pHimarEM3	Carries HimarEM1, Apr (Emr), vector used for transposon mutagenesis	34
pCFX	Gene deletion template plasmid carrying cfxA flanked by two MCS; Apr (Cfxr); derived from pCFXSK with removal of the FRT sites	21
pSKSO8TG	Used for complementation of the Δ0922 mutant with oriC; Apr (Emr)	20
pHB0922	Plasmid constructed from pSKSO8TG used for complementation of the Δ0922 mutant; Apr (Emr)	This study
pUC18	Plasmid used for gene cloning; Apr	TaKaRa
pYT313	sacB-containing suicide vector; Apr (Emr)	36

^aMCS, multiple cloning sites; FRT, FLP recombination target; Ap^r, ampicillin resistance; Cfx^r, cefoxitin resistance; Em^r, erythromycin resistance. Phenotypes in parentheses are expressed in C. hutchinsonii, and phenotypes not in parentheses are expressed in E. coli.

HimarEM3 transposon mutagenesis and identification of insertion sites.

The transposon-containing plasmid (pHimarEM3) was transformed into C. hutchinsonii ATCC 33406 by electroporation to screen genes involved in cellulose degradation. PY6 plates with erythromycin were used to culture the transformants, and then the transformants were transferred onto Stanier plates covered with filter paper for 15 days. The mutants defective in filter paper degradation were chosen to identify the insertion sites by inverse PCR (34).

Genetic construction.

The chu_0922 deletion mutant was constructed as previously reported (22). A 1.5-kbp fragment containing the initial 114 bp of chu_0922 and the region directly upstream was amplified from the C. hutchinsonii genome with primers 0922H1F and 0922H1R, referred to as 0922H1. 0922H1 was digested with BamHI and PstI and cloned into plasmid pCFX, generating pCFXH1. A 1.5-kbp fragment containing the last 778 bp of chu_0922 and the region directly downstream was amplified with primers 0922H2F and 0922H2R, referred to as 0922H2. 0922H2 was digested with SalI and SacI and ligated into the corresponding sites of pCFXH1. The gene-targeting cassette was amplified with primers 0922H1F and 0922H2R and purified with a Cycle Pure kit. A total of 1.5 μg PCR product was transformed into 100 μL competent cells of the wild type by electroporation. The transformants were cultured using PY6 plates with cefoxitin. The transformants were verified by PCR with two sets of primers, 0922UF/0922UR and 0922UF/CFXR. The C-terminal domain (CTD) of CHU_0922 was deleted using the same method and plasmid with the primers (0922CHIF, 0922CHIR, 0922CH2F, and 0922CH2R) listed in Table S1.

The mutants WT-0922Strep and ΔgldN-0922Strep were constructed as described by Lauber et al. with some modifications (35). The Twin-Strep tag was inserted between the signal peptide and the mature CHU_0922 using overlap PCR and the Gibson assembly method. A 2.0-kp fragment containing the sequence upstream of chu_0922 and the signal peptide sequence was amplified from the genome of C. hutchinsonii with primers 0922S1F and 0922S1R, referred to as S1. The fragment of 0922S1 was generated by overlap PCR using S1 as the template with primers 0922S1F and 0922S1R2. A 1.8-kp fragment containing the region of chu_0922 after the signal peptide was amplified from the genome of C. hutchinsonii with primers 0922S2F and 0922S2R, referred to as S2. The fragment of 0922S2 was generated by overlap PCR using S2 as the template with primers 0922S2F2 and 0922S2R. The fragments of 0922S1 and 0922S2 were assembled by the Gibson assembly mixture using pUC18, which was digested with BamHI and SalI. The transformants were verified by sequencing. Then, the target fragment was digested with BamHI and SalI and ligated to the suicide plasmid pYT313 (36). The suicide plasmid was transformed into 100 μL competent cells of the wild type and the ΔgldN mutant by electroporation. Cells with chromosomally integrated plasmids were selected using erythromycin resistance. The resulting clones were cultured in PY6 medium without antibiotics for 24 h for the loss of the plasmid backbone and then plated on PY6 agar containing 5% sucrose. The colonies resistant to sucrose were screened by PCR for the presence of the Twin-Strep tag and verified by sequencing.

Complementation of the Δ0922 mutant.

The complementation of the Δ0922 mutant was performed as previously described (22). Briefly, a fragment containing the constitutive promoter was amplified with primers P1284F and P1284R. chu_0922 was amplified with primers C0922F and C0922R. The fragment containing the constitutive promoter and chu_0922 was generated by fusion PCR using primers P1284F and C0922R. The fragment was then digested with SalI and SacI and ligated into the corresponding sites of pSKSO8TG, generating pHB0922. The plasmid was transformed into the Δ0922 mutant by electroporation. Erythromycin was used to select the transformants. The complemented strain of Δ0922 with pHB0922 was named C0922.

Cellulose degradation assay.

Strains were precultured in PY6 medium to mid-log phase. Cells were collected by centrifugation, washed with PY6 medium without a carbon source, and then adjusted to an optical density at 600 nm (OD₆₀₀) of 1.0 for inoculation. Because Avicel and regenerated amorphous cellulose (RAC) are insoluble and cells of C. hutchinsonii can adhere to Avicel and RAC, the growth state of C. hutchinsonii in Stanier medium supplemented with Avicel and RAC could not be determined using the OD₆₀₀. In the case of Avicel and RAC being used as the sole carbon source, cellular protein concentration indicating the growth status was detected at set intervals as previously described (21). To determine the filter paper utilization ability, equivalent amounts of cells of the wild type and mutants were spotted on Stanier agar covered with Whatman filter paper. The plates were incubated at 30°C, and the degradation of filter paper was recorded by a Canon camera at set intervals.

Measurement of cellulose crystallinity by XRD.

To determine the crystallinity of Avicel PH101 (Sigma-Aldrich) incubated with C. hutchinsonii cells, strains were cultured in Stanier medium to mid-exponential phase. Cells were harvested, washed twice with fresh Stanier medium, and then transferred into Stanier medium supplemented with 0.2% (vol/vol) Avicel PH101. Cells were cultured at 30°C with shaking at 160 rpm. Samples were taken at set intervals (0 h, 24 h, and 48 h) and centrifuged at 5,000 × g for 10 min to harvest the cell pellets. The cell pellets were suspended with 0.2 M NaOH, boiled for 10 min, and centrifuged at 12,000 × g for 10 min to remove the supernatant. The pellets were then washed twice with distilled water and dried at 60°C overnight. The X-ray diffraction (XRD) of samples was observed with a D8 Advance system diffractometer (Bruker, Germany). The XRD crystallinity index (CI_XRD) was calculated as previously described (18): CI_XRD (%) = (I₀₀₂ − I_am)/I₀₀₂ × 100. I₀₀₂ is the height of the crystalline peak at 22°, and I_am is the intensity of the peak at 18°.

Microscopic observation the surface morphology of Avicel.

The surface morphology of Avicel PH101 was observed by scanning electron microscope (SEM) as previously described (18). Briefly, strains were cultivated in Stanier medium with 0.4% Avicel PH101 as the sole carbon source for 48 h at 30°C with shaking at 160 rpm. The samples were centrifuged at 5,000 × g for 10 min to harvest the pellets. The pellets were resuspended in 0.2 M NaOH, boiled for 10 min, and centrifuged at 12,000 × g for 10 min to remove the supernatant. The pellets were then washed twice with distilled water and dried at 60°C overnight. Avicel PH101 not treated with C. hutchinsonii was used as the negative control.

Cellulase activity assay.

Cellulase activity assays were carried out as previously described (18, 21). Briefly, cells were cultivated in Stanier medium to mid-log phase and harvested by centrifugation at 5,000 × g for 5 min. Cell pellets were washed with Na₂HPO₄-KH₂PO₄ buffer (100 mM, pH 6.8) and centrifuged at 5,000 × g for 5 min to collect cell pellets. For intact cell samples, cell pellets were resuspended with Na₂HPO₄-KH₂PO₄ buffer (100 mM, pH 6.8). For cell extract samples, cell pellets were resuspended with Na₂HPO₄-KH₂PO₄ buffer (100 mM, pH 6.8) containing 2% (vol/vol) Triton X-100 for about 4 h at 4°C, and 0.5 mg/mL phenylmethylsulfonyl fluoride (PMSF) was added to deactivate the proteases. Sodium carboxymethyl cellulose (CMC-Na; 1% [wt/vol]) and 2 mM p-nitrophenyl-β-d-glucopyranoside (pNPG) were used as substrates to measure endoglucanase activity and β-glucosidase activity. To measure endoglucanase activity, a mixture containing 500 μL of 1% CMC-Na and 500 μL of the resuspended cells was incubated at 30°C for 30 min. 3,5-Dinitrosalicylic acid was used to determine the reducing ends. To determine the β-glucosidase activity, a mixture containing 500 μL of 2 mM pNPG and 500 μL of the resuspended cells was incubated at 30°C for 30 min. The released p-nitrophenol was determined by measuring absorption at 410 nm. The intracellular cellulase activity was calculated by the cellulase activity of the cell extracts minus that of the intact cells. The protein concentration was measured as described by Bradford (37). All the measurements were carried out in triplicate.

Cell fractionation and Western blot analysis.

Strains were cultivated in Stanier medium. The cell pellets were collected by centrifugation at 5,000 × g for 10 min at 4°C, then resuspended with 50 mM piperazine-N,N-bis(2-ethanesulfonic acid) (PIPES) buffer (pH 7.0) supplemented with 1 mM PMSF, and disrupted by sonication on ice. The cell debris was removed by centrifugation at 12,000 × g for 10 min at 4°C. The supernatant was collected as the whole-cell proteins. Extracellular proteins were isolated as previously described (20). Strains were cultivated in PY6 medium to mid-log phase, the culture was centrifuged at 5,000 × g for 10 min at 4°C, and the supernatant was filtered through a 0.22-μm-pore-size polyvinylidene difluoride (PVDF) filter. The cell-free supernatant was concentrated using Amicon 10-kDa Ultra-15 centrifugal filter units (Millipore, MA, USA). Outer membrane proteins were extracted as previously described (21). Strains were cultivated in PY6 medium or Stanier medium to mid-log phase; then cells were harvested by centrifugation at 5,000 × g for 10 min at 4°C. The cell pellets were washed once with 50 mM PIPES buffer, resuspended with PIPES buffer supplemented with 0.5 M NaCl, and incubated at 4°C for 30 min with shaking at 150 rpm. Cells were removed by centrifugation at 12,000 × g for 10 min at 4°C, and the supernatant containing the buffer-washed proteins was ultracentrifuged (100,000 × g, 30 min, 4°C). The sediment was resuspended in PIPES buffer as outer membrane proteins.

The periplasmic proteins were extracted as previously described (21, 38). Briefly, strains were cultured to mid-log phase, and cells with equal biomass were collected by centrifugation at 5,000 × g for 10 min at 4°C. The cell pellets were resuspended by 0.3 M Tris-HCl (pH 8.0) containing 20% sucrose and 1 mM EDTA and incubated on ice for 15 min. Cells were collected by centrifugation at 8,000 × g for 10 min at 4°C. The cell pellets were resuspended vigorously by cold double-distilled water (ddH₂O), and incubated on ice for 15 min. The supernatant was collected by centrifugation at 12,000 × g for 10 min at 4°C, and was concentrated using Amicon 10-kDa Ultra-15 centrifugal filter units (Millipore, MA, USA). Proteins with equal biomass were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then stained by Coomassie brilliant blue R-250. Deferential bands between the wild type and the Δ0922 mutant were identified by mass spectrometry. Western blotting was performed as previously described (21). Whole-cell proteins, outer membrane proteins, and periplasmic proteins of different strains with equal biomass were separated by SDS-PAGE and then transferred onto a PVDF membrane. Antibody against CHU_3220 was the same as previously reported (21). Antibody against the Strep tag was purchased from ABclonal.

Isolation of cellulose-binding proteins in the outer membrane proteins of C. hutchinsonii.

The outer membrane cellulose-binding proteins were prepared as previously described (39) with some modifications. The outer membrane proteins were extracted as described above and solubilized in PIPES buffer containing 2% (vol/vol) Triton X-100 and 1 mM PMSF at 4°C for 12 h. Then the samples were centrifuged at 100,000 × g for 30 min at 4°C to remove the insoluble materials. The supernatant was collected, mixed with the same volume of autoclaved Avicel PH101, incubated with shaking for 2 h at 4°C, and centrifuged at 10,000 × g for 5 min at 4°C. The sediment was washed three times with PIPES buffer. Proteins bound to Avicel PH101 were eluted with SDS-PAGE sample buffer and boiled for 10 min. The proteins were separated by SDS-PAGE and stained with Coomassie brilliant blue R-250. The differential bands between the wild type and the Δ0922 mutant were identified by mass spectrometry.

Real-time qPCR analysis.

Strains were cultivated in Stanier medium to mid-exponential phase. Four milliliters of the culture were centrifuged at 5,000 × g for 5 min at 4°C to collect the cell pellets. Total RNA extraction and quantitative PCR (qPCR) were performed as previously described (21, 40). qPCR was performed on a LightCycler 480 System with SYBR green Premix Pro Taq HS qPCR kit (Accurate Biotechnology, Hunan, China). The data were analyzed using the relative quantitation/comparative threshold cycle (ΔΔC_T) method (41) and normalized to an endogenous control (16S rRNA gene). Three biological repeats were set for all assays.

Purification of CHU_0922.

The purification of CHU_0922 was performed as previously described with some modification (35). The WT-0922Strep mutant and wild type were cultured in 4 L PY6 medium to mid-log phase. Cells were collected by centrifugation at 5,000 × g for 10 min at 4°C. The cell pellets were resuspended with binding buffer (100 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1 mM EDTA) plus 1 mM PMSF. Cells were disrupted by sonication, and the cell debris was removed by centrifugation at 12,000 × g for 10 min at 4°C. The supernatant was collected, and the membrane proteins were solubilized with 1% (wt/vol) N-dodecyl-β-d-maltoside (DDM) on ice for 2 h. The mixture was centrifuged at 150,000 × g for 45 min at 4°C to remove the insoluble material. The supernatant was incubated with high-performance Strep-Tactin Sepharose resin (GE Healthcare) on ice for 1 h and then flowed through the column. The column was washed with 15 column volumes (CV) of binding buffer, and 6 CV of binding buffer containing 2.5 mM desthiobiotin was used to elute the bound proteins. The eluate was freeze-dried to concentrate, and then it was analyzed by Western blotting and LC-MS/MS.

Statistical analysis.

Statistical analysis was performed using Student's t test. Three biological replicates were undertaken for each analysis. Reported results and errors are means and standard deviations, respectively, for these replicates.