Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Function and Evolution of Two Forms of SecDF Homologs in Streptomyces coelicolor

  • Zhan Zhou,

    Affiliations Province Key Laboratory of Microbial Biochemistry and Metabolism Engineering, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China, Department of Biology and South Texas Center for Emerging Infectious Diseases, University of Texas at San Antonio, San Antonio, Texas, United States of America, State Key Laboratory of Genetic Engineering and Ministry of Education Key Laboratory of Contemporary Anthropology, School of Life Sciences, Fudan University, Shanghai, China

  • Yudong Li , (Y-QL); (YL)

    Affiliation Department of Bioengineering, School of Food Sciences and Biotechnology, Zhejiang Gongshang University, Hangzhou, Zhejiang, China

  • Ning Sun,

    Affiliation Province Key Laboratory of Microbial Biochemistry and Metabolism Engineering, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

  • Zhihao Sun,

    Affiliation Province Key Laboratory of Microbial Biochemistry and Metabolism Engineering, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

  • Longxian Lv,

    Affiliation Province Key Laboratory of Microbial Biochemistry and Metabolism Engineering, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

  • Yufeng Wang,

    Affiliation Department of Biology and South Texas Center for Emerging Infectious Diseases, University of Texas at San Antonio, San Antonio, Texas, United States of America

  • Libing Shen,

    Affiliation State Key Laboratory of Genetic Engineering and Ministry of Education Key Laboratory of Contemporary Anthropology, School of Life Sciences, Fudan University, Shanghai, China

  • Yong-Quan Li (Y-QL); (YL)

    Affiliation Province Key Laboratory of Microbial Biochemistry and Metabolism Engineering, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

Function and Evolution of Two Forms of SecDF Homologs in Streptomyces coelicolor

  • Zhan Zhou, 
  • Yudong Li, 
  • Ning Sun, 
  • Zhihao Sun, 
  • Longxian Lv, 
  • Yufeng Wang, 
  • Libing Shen, 
  • Yong-Quan Li


The general secretion (Sec) pathway plays a prominent role in bacterial protein export, and the accessory component SecDF has been shown to improve transportation efficiency. Inspection of Streptomyces coelicolor genome reveals the unexpected presence of two different forms of secDF homologous genes: one in fused form (secDF) and the other in separated form (secD and secF). However, the functional role of two SecDF homologs in S. coelicolor has not yet been determined. Transcriptional analysis of secDF homologs reveals that these genes are constitutively expressed. However, the transcript levels of secD and secF are much higher than that of secDF in S. coelicolor. Deletion of secDF or/and secD/secF in S. coelicolor did result in reduced secretion efficiency of Xylanase A and Amylase C, suggesting that they may have redundant functions for Sec-dependent translocation pathway. Moreover, our results also indicate that SecD/SecF plays a more prominent role than SecDF in protein translocation. Evolutionary analysis suggests that the fused and separated SecDF homologs in Streptomyces may have disparate evolutionary ancestries. SecD/SecF may be originated from vertical transmission of existing components from ancestor of Streptomyces species. However, SecDF may be derived from bacterial ancestors through horizontal gene transfer. Alternately, it is also plausible that SecDF may have arisen through additional gene duplication and fusion events. The acquisition of a second copy may confer a selective benefit to Streptomyces by enhancing protein transport capacity. Taken together, our results provide new insights into the potential biological function and evolutionary aspects of the prokaryotic SecDF complex.


Streptomycetes are soil-dwelling Gram-positive bacteria that have a large (6.5–11.9 Mb) linear chromosome with high GC content, containing a central region that is highly conserved throughout the genus, and terminal regions that are variable in composition and organization [1], [2]. Streptomyces species secreted large quantities of proteins via the general secretory (Sec) pathway and twin-arginine translocation (Tat) pathway to their living niches [3][6]. These systems could facilitate nutrient acquisition by secreting chitinases and cellulases to degrade insoluble nutrient sources, and regulate morphological development of Streptomyces by secreting peptidases and their associated inhibitors [5]. Moreover, Streptomyces has become an important production host for heterologous expression of recombinant proteins due to its excellent secretion capacity, which makes the downstream processing much simpler [7][10]. Therefore, understanding of the mechanism of its protein export systems is extremely important for the economical application of Streptomyces.

Protein transport across the bacterial membrane is mediated by different translocation systems, of which the general protein secretion (Sec) system plays a prominent role in protein export and membrane insertion [11][15]. The Sec translocation machinery is a protein complex comprised of SecYEG (SecY, SecE, SecG), the ATPase SecA, and the accessory factor SecDF (SecD and SecF) [14][16]. The roles of the Sec pathway components have been extensively studied, but the function of the SecDF complex in protein secretion is still poorly understood. The SecDF complex has been shown to be involved in the cycling of SecA and the release of the translocated protein from the translocation channel [17], [18]. Recently, the SecDF complex was proposed to be a membrane integrated chaperone that uses proton motive force (PMF) to complete protein translocation through the SecYEG channel via the control of SecA cycling [19].

The SecDF protein complex belongs to the resistance-nodulation-cell division (RND) family of multidrug export pumps, which is widely distributed and conserved in all three major kingdoms of life. However, the SecDF complex is only found in kingdoms Bacteria and Archaea [20]. The SecDF complex is comprised of two structurally related proteins, SecD and SecF, which have previously been shown to have arisen via an internal gene duplication event [21]. The secD and secF genes are adjacent in genomes and widely distributed in bacteria, such as Escherichia coli [22]. Interestingly, the fusion form of secD and secF genes, secDF, has also been observed in bacteria, such as Bacillus subtilis and Staphylococcus aureus [23], [24]. These studies showed, for example, that both SecDF in B. subtilis [23] and SecD/SecF in E. coli [25] are involved in catalyzing protein translocation. The existence of the two forms of secDF genes shows different evolutionary pathways of SecDF complex. However, the evolutionary processes still remain obscure.

Here, we report the presence of two different forms of secDF genes, one separated (secD/secF, hereafter secD-F) and one fused (secDF) in the genome of S. coelicolor [26]. To further understand the auxiliary functions of SecDF and SecD-F, these genes were deleted together or separately. Our results here show that the simultaneous depletion of these genes results in severe inhibition of Xylanase A (XlnA) and Amylase C (AmlC) secretion, suggesting that SecDF and SecD-F may have redundant functions in the Sec protein translocation pathway. Moreover, phylogenetic analysis supported the proposal that the fused and separated secDF homologs of Streptomyces may have different evolutionary ancestries.

Materials and Methods

Bacterial strains, plasmids and growth conditions

Bacterial strains and plasmids used in this study are listed in Table 1, and primers are listed in Table S1 in File S1. The wild type strain M145 of S. coelicolor A3(2) strain was used for construction of a series of mutant strains in this study (Table 1). E. coli TG1 was used for gene cloning and plasmid construction. The non-methylating E. coli ET12567 containing plasmid pUZ8002 was used for conjugation with Streptomyces for transduction of plasmids.

E. coli strains were routinely cultured in Lennox broth (LB) liquid medium or on agar plates at 37°C. S. coelicolor strains were cultured at 30°C in liquid tryptic soy broth (TSB) or YEME medium for mycelium preparation in the primary metabolism, or in M14 medium [27] for the xylanase activity assay. Solid MM, SMMS and R2YE medium were used for cell differentiation, and MSF medium for spore preparation of S. coelicolor [28].

Construction of S. coelicolor mutant strains

The secD, secF and secDF genes were disrupted by in frame deletion via double-crossover homologous recombination [28] with the help of the temperature sensitive plasmid pKC1139 [29]. The secD and secF genes were knocked out together. About 1 kb of the upstream sequence of secD gene and the downstream sequence of secF gene were cloned and linked into pKC1139 to construct the secD-F gene disruption plasmid pKC1139a. About 1 kb of sequences both upstream and downstream of secDF gene were cloned and linked into pKC1139 to construct the secDF gene disruption plasmid pKC1139b. Disruption plasmids pKC1139a and pKC1139b were conjugated into wild-type strain M145 through E. coli ET12567/pUZ8002, to knock out secD-F and secDF, respectively (Figure 1A). Plasmid pKC1139a was also introduced into ZJUZ2 (secDF null mutant) to construct the double mutant ZJUZ3, likewise plasmid pKC1139b was introduced into ZJUZ1 to create ZJUZ3. The gene disruption achieved in each mutant stain was verified by PCR and Southern blot analysis [30].

Figure 1. The gene knock out processes and gene expression profiles of secDF homologous genes in S. coelicolor.

(A) Schematic representation of the in frame deletion of secD-F and secDF by double-crossover homologous recombination in S. coelicolor M145. Plasmids pKC1139a and pKC1139b were used to knock out secD-F (ZJUZ1) and secDF (ZJUZ2), respectively. Both pKC1139a and pKC1139b were used to create the ZJUZ3, which lacks both secD-F and secDF genes. (B) Gene expression analysis of secD, secF and secDF in S. coelicolor wild type and mutant strains by RT-PCR with 30 and 35 cycles. Gene hrdB was used as a reference gene.

The 300 bp promoter regions of secD-F and secDF were cloned from genomic DNA of M145, and inserted into the promoter detecting vector pIJ8630 [31] to generate the plasmid pZJ9 and pZJ10, respectively. The plasmids were introduced into M145 to detect the activity of the secD-F and secDF promoters. The plasmid pIJ8630 was used as negative control and the plasmid pLM1 with promoter ermEp* was used as positive control [32].

The expression plasmid pZJ2 was reconstructed from pIJ8668 [31] and pHZ1272 [33], with reppIJ101 and reppUC replicon, oriT fragments and antibiotic resistance genes tsr and aac3(IV). The xlnA gene (SCO5931) was amplified from genomic DNA of M145 with its own promoter and His6 tag in C terminal, and was then inserted into pZJ2 to create the xlnA expression plasmid pZJ6. Plasmid pZJ6 was introduced into the wild type strain M145, and into the secD-F and/or secDF disrupted strains ZJUZ1, ZJUZ2 and ZJUZ3. The amlC gene (SCO7019) was amplified from genomic DNA of M145, then inserted into pLM1 and replaced the egfp gene, resulting in the amlC expression plasmid pZJ7. Plasmid pZJ7 was introduced into the wild type strain M145, and into the secD-F and/or secDF disrupted strains ZJUZ1, ZJUZ2 and ZJUZ3.

Protein analysis and enzyme assay

For detecting the promoter activity, S. coelicolor mycelia were collected after incubation of spores in liquid TSB medium or on cellophane overlaid on R2YE medium. The collected mycelia were resuspended in lysis buffer (100 mM NaH2PO4, 10 mM Tris, pH 8.0, 1 mM phenylmethylsulfonyl fluoride), then lysed by ultrasonication [32]. Total protein concentration was determined using Bradford Protein Assay Kit (Invitrogen). About 20 µg total protein was separated by SDS-PAGE for Coomassie Blue staining and Western blot analysis [30] with anti-green fluorescent protein (α-GFP) antibody (Proteintech Group).

For assay of xylanase activity, S. coelicolor strains were grown in liquid M14 medium. Xylanase activity in supernatants was measured by the dinitrosalicylic acid method. One unit (IU) of enzyme activity is defined as the amount of enzyme required to release 1 mmol of reducing sugars (D-xylose) in 1 min at 57°C [27]. XlnA protein in supernatants was also analyzed by SDS-PAGE for Coomassie Blue staining and Western blot analysis [30] with anti-His antibody (Tiangen).

For semi-quantitatively assay of extracellular amylase activity, 20 µl spores (1×1010 cfu/ml) of S. coelicolor strains were cultured on solid MM medium (containing 0.2% soluble starch as the carbon source) at 30°C. The inoculated plates were grown for 120 hours, then they were stained with Lugol’s solution (5% I2 and 10% KI mixed in distilled water) for 30 min. The diameters of the transparent zones were measured to determine the relative activity [34], [35]. All measurements were depicted as a percentage of the extracellular amylase activity of ZJUZ27, which contains the amlC expression plasmid pZJ7 in the wild-type S. coelicolor strain. Each data point was measured by three times.

RNA isolation, RT-PCR and quantitative RT-PCR

The total RNA of S. coelicolor was isolated from the mycelia grown in YEME medium. RNA was prepared with RNAprep pure Cell/Bacteria Kit (Tiangen) according to the manufacturer’s instructions. The genomic DNA was removed by RNase-free DNase I (Takara), and the RNA concentration was determined by measuring the OD260/OD280 in a spectrophotometer.

RT-PCR was performed in two steps: cDNA was made from 2 µg total RNA using M-MLV Reverse Transcriptase (Takara) according to the manufacturer’s manual. Then cDNA was amplified by PCR with rTaq DNA Polymerase (Takara). Amplification and detection by quantitative RT-PCR (qPCR) were performed on an Eppendorf realplex 2 Mastercycler machine (Eppendorf), and the synthesized cDNA from the first step was amplified with SYBR Premix Ex Taq (Takara) according to the manufacturer’s instructions. The principal sigma factor coding gene (hrdB), which is a housekeeping gene in S. coelicolor, was used as the internal control: all values were normalized as the relative transcript level to hrdB. All reactions were performed in triplicate.

Bioinformatic analysis

The putative SecD, SecF and SecDF protein sequences of S. coelicolor were retrieved from StrepDB (, and were used as query sequences to detect SecDF homologs in the microbial proteins of non-redundant protein database (nr) from NCBI website [36], and the Integrated Microbial Genomes (IMG) database [37] by BLASTP searches (e-value cut-off = 10−10, coverage = 70%). The SecD and SecF homologs of other taxa were obtained from clusters of orthologous groups (COG) family COG0341 and COG0342, respectively [38]. Protein alignment was performed using ClustalW with default parameters [39]. Protein domain predictions were made using Pfam [40] and SMART analysis [41] (Simple Modular Architecture Research Tool; Prediction of transmembrane helices in proteins was performed using TMHMM [42]. For promoter analysis of secDF and secD-F genes, the sequences 300 bp upstream of the coding regions were analyzed using the regulatory sequence analysis tools website [43] (

Evolutionary analysis

The amino acid sequences of SecD and SecF were concatenated into single sequence SecD-F. Full-length sequences were aligned with ClustalW [39] with default settings. Then, poorly aligned regions were removed using Gblocks v0.91b [44] with the following settings: maximum number of contiguous non-conserved positions allowed = 8; minimum length of a block allowed = 5. The neighbor-joining (NJ) tree was constructed from the alignment using the MEGA version 6.0 [45] with the parameters: p-distance model for protein sequences and pairwise deletion option for gaps. A maximum likelihood (ML) tree was built from the alignment by PhyML version 3.0, which was available at PhyML webservice ( [46]. Branch supports for phylogenetic trees were determined using 1000 bootstrap replicates. The model archaeal strain (Haloferax volcanii) was used as outgroup.

To identify genes neighboring the two SecDF types, genomic synteny among 14 Streptomyces species with complete genome sequences was investigated using the gene neighborhood tool in the Integrated Microbial Genomes system provided by the US Department of Energy Joint Genome Institute [37] (

Results and Discussion

Two forms of SecDF homologs occur in S. coelicolor

Analysis of the complete genome sequence of S. coelicolor [26] identified three genes closely related to secD (SCO1516), secF (SCO1515) and secDF (SCO6160), with GC content 69.6%, 67.4% and 73.0%, respectively (Figure 2A). The location of secD and secF ORFs (2,838 nucleotides in length) are adjacent to each other in the conserved chromosome core region. The two ORFs are separated by an intergenic sequence of 3 nucleotides (GAG). In contrast, secDF is located in the variable terminal region. The secD gene encodes a protein of 570 amino acids, which is similar to E. coli SecD (26% identity). The secF gene encodes a protein of 373 amino acids, which is similar to E. coli SecF (31% identity). S. coelicolor secDF encodes a predicted membrane protein of 795 amino acids, which shows 30% identity to the SecDF protein of B. subtilis. Moreover, analysis of 17 Streptomyces species with complete genome sequences revealed that SecD and SecF are presented in all these species, but SecDF is absent from seven of these species (Table S2 in File S1). These results suggest that SecD/SecF homologs might play a more important role than SecDF in Streptomyces and undergo stabilizing selection.

Figure 2. The gene location and predicted topology of S. coelicolor SecDF homologs.

(A) Schematic diagrams of the location of secDF homologous genes in S. coelicolor genome. The core region of S. coelicolor is indicated in gray box. (B) The predicted topology of S. coelicolor SecDF protein. The conserved domains D1–D6 and F1–F4, which are presented in all known SecD and SecF proteins, are highlighted in black color.

The SecD, SecF and SecDF homologs of S. coelicolor were analyzed using TMHMM [42] to predict their membrane topologies. Consistent with the experimentally derived topologies of E. coli SecD and SecF, the S. coelicolor SecD and SecF were predicted to encode integral membrane proteins with six transmembrane segments. The fused SecDF is predicted to contain 12 transmembrane helices and two large extracellular domains (Figure 2B). The amino and carboxyl termini of these proteins are predicted to be located in the cytoplasm, and substantial extracellular domains are present between transmembrane helices 1 and 2 in all of them, besides 7 and 8 in SecDF. The first large size periplasmic loop has been shown to be involved in facilitating the formation of stable, folded conformations in the translocated Sec substrates [47].

Transcriptional analysis of the secDF homologous genes

For comparative analysis of secD, secF and secDF transcription, total RNA was isolated from S. coelicolor M145 after 36 h cultivation in YEME medium. The transcript abundance for secD-F and secDF genes was measured by RT-PCR and calibrated with hrdB gene (Figure 1B). Both secD and secF were highly transcribed in the wild type strain M145, whereas the transcripts of secDF were not detectable after the same 30 cycles of RT-PCR. This result indicated the level of secDF transcripts was too low to be detected by RT-PCR. However, when the PCR cycle number was increased to 35, the secDF transcript became detectable, suggesting that secDF gene was barely expressed in the wild-type strain. Overall, in the wild-type strain, the transcript levels of secD and secF are much higher than those of secDF.

Furthermore, promoter prediction for secD-F gene identified a putative promoter sequence (tcggggcgtgacgacccgctcacgcggggcacattcgccggagcaggaac) that starts from 115 bp upstream of the secD start codon, but no putative promoter sequence was found for the secDF gene. To validate the promoter activity, we cloned a 300 bp fragment of the DNA upstream of the secD-F and secDF start codons (ATG), and inserted them in the upstream of egfp gene on the Streptomyces promoter-probe plasmid pIJ8630 [31], yielding plasmids pZJ9 and pZJ10, respectively. The plasmids were introduced into M145 to create strains ZJUZ40 and ZJUZ41. The strains LM6 [32], carrying a plasmid with the constitutive promoter ermEp* [48], and ZJU39, carrying a plasmid with no promoter upstream of egfp gene, were used as positive and negative controls, respectively. Comparison of the transcription levels by qPCR revealed that the promoter activity of secD-F was about 60% of the strong promoter activity of ermEp*, whereas the promoter activity of secDF was barely detectable (Figure S1A in File S1). Western blot analysis was performed to analyze the expression of EGFP protein after 24, 36, 48 and 72 hours of growth (Figure S1B in File S1). The expression of EGFP in ZJUZ40 was nearly constant over time, but EGFP was barely detected by western blot in ZJUZ41. Therefore, the distinct expression pattern of secDF and secD-F was reflected by their promoter activities. The different promoter activities of secDF and secD-F indicated different transcription levels between them, suggesting secDF and secD-F might have different functions in the protein translocation process.

To evaluate whether the transcription of secDF genes depends on the growth phase, we carried out time-course expression analyses of S. coelicolor wild type strain M145. The strain was cultivated in YEME medium, and total RNA was isolated after 18, 24, 36 and 60 hours (Figure 3A). qPCR revealed that secD, secF and secDF were transcribed throughout the time of cultivation with nearly constant transcription levels, suggesting that these genes are expressed constitutively. The constant expression levels of these auxiliary components of Sec systems indicated their important roles in different growth phases of Streptomyces.

Figure 3. qPCR analysis of gene expression profiles of secD, secF and secDF in S. coelicolor wild type and mutant strains.

(A) qPCR analysis of gene expression profile of secD, secF and secDF in S. coelicolor M145 alone the time of cultivation. (B) qPCR analysis of secD and secF gene expression in M145 and ZJUZ2. (C) qPCR analysis of secDF gene expression in M145 and ZJUZ1. Gene hrdB was used as a reference gene.

Disruption of SecDF homologous genes and their functional effects

It has been shown that while mutations in the E. coli secDF-yajC operon result in a general protein secretion defect, this operon is not essential for viability [18], [25]. To determine the function of two SecDF homologs in S. coelicolor, we deleted each secDF homologous gene, creating three knock-out strains ZJUZ1, ZJUZ2 and ZJUZ3 with secD-F, secDF, and both secD-F and secDF deleted respectively (Figure 1A). These mutants were constructed by homologous recombination approach (see Materials and Methods) and were verified by PCR and Southern blot (Figure S2 in File S1).

The growth of obtained mutants with secD-F and/or secDF genes disrupted were not impaired when inoculating on four culture media, MM, SMMS, MSF, and R5, which demonstrated that the secDF homologous genes were not essential for the growth of S. coelicolor under the conditions tested (Figure S3 in File S1). As both E. coli strains with disrupted secD and/or secF genes, and B. subtilis strains with a disrupted secDF gene are cold-sensitive for growth and barely viable at 37°C [23], [49], [50], we tested the growth of the S. coelicolor mutants with disrupted secDF homologous genes at several temperatures (23°C, 30°C, 37°C). However, the mutants did not show reduced growth rates compared with the wild type strain at each temperature (data not shown). The fact that the growth of the S. coelicolor mutant with secD-F and secDF genes disrupted together is not severely impaired indicates that neither SecD-F nor SecDF is required for translocation of the essential proteins for viability and growth in these surveyed conditions.

Deletion of the S. coelicolor secDF homologous genes results in a Sec-specific protein export defect

In order to test the effects of the SecDF (ΔsecDF) or SecD-F (ΔsecD-F) absence on protein translocation in S. coelicolor mutants, we tested transport efficiency using the Sec-dependent substrate Xylanase A (XlnA) and Amylase C (AmlC) as a reporter protein. XlnA is secreted by the Sec-dependent pathway in Streptomyces to hydrolyze the β-1,4-D-glycosidic bonds of xylan in plant cell walls [51]. A vector, designated pZJ6, was constructed to express XlnA with His6 tag on its C terminus. The vector pZJ6 was transformed into S. coelicolor M145, ZJUZ1, ZJUZ2 and ZJUZ3, and the transformants were designated ZJUZ23, ZJUZ24, ZJUZ25, and ZJUZ26, respectively. When this construct was expressed in the four strains after cultivation for 96 hours, the culture supernatants have comparable protein profiles and similar amounts of proteins, as detected by SDS-PAGE, as shown in the lower part of Figure 4A. As SecDF was generally supposed to affect protein translocation efficiency, which can only be detected by pulse-chase method, but not by SDS-PAGE. Therefore, there is no significant difference detected between mutant and wild-type strains in this study.

Figure 4. Assay of extracellular XlnA and AmlC activity and protein amounts.

(A) Detecting XlnA protein secretion by SDS-PAGE and Western blot. WB: Western blot; CB: Commassie Blue Staining. ZJUZ23-26 strains were cultivated for 96 h. (B) Assay of extracellular XlnA activity. ZJUZ23-26 strains were cultivated for 72 h and 96 h. *: p<0.05, **: p>0.05, Chi-squared test. (C) Relative amylase activity from ZJUZ27, ZJUZ28, ZJUZ29 and ZJUZ30. The diameters of the transparent zones around colonies were measured to determine the relative activity.

Western blot analysis was performed to detect the secretion level of XlnA-His6 in the supernatant of S. coelicolor cultures (upper part of Figure 4A). As it was expected, XlnA was found to be efficiently translocated into the culture supernatant of S. coelicolor strain ZJUZ23, which was constructed from the wild type strain M145. As compared to that of strain ZJUZ23, the S. coelicolor strain ZJUZ25 (ΔsecDF) secreted levels of XlnA into the medium that were comparable to the wild-type construct. In contrast, the levels of mature protein in the culture supernatant of the strain ZJUZ24 (ΔsecD-F) and ZJUZ26 (ΔsecD-F & ΔsecDF), were present at substantially reduced levels relative to the wild-type strain, suggesting a translocation defect. Moreover, the levels of protein XlnA in the supernatant of strain ZJUZ26, which lacks both secDF and secD-F, were reduced more dramatically than that in the strain ZJUZ24, which lacks only secD-F. This suggests that SecDF plays an important role in promoting protein translocation when the secD-F was depleted.

To confirm these results, the XlnA activity in supernatant of each strain was measured at two time points (72 and 96 hours, respectively) (Figure 4B). There is no significant difference in the amount of secreted xylanase between strain ZJU25 with secDF deleted and strain ZJU23 (P>0.05, Chi-squared test). However, the strain ZJUZ24 lacking SecD-F has a significantly lower XlnA activity as compared to strain ZJUZ23 (P<0.05, Chi-squared test). The amount of secreted xylanase was reduced to 54% (72 h) and 66% (96 h), respectively. As expected, the XlnA activity was dramatically reduced to about 0.1 U/mg dry weight (about 30% of ZJUZ23) in strain ZJU26 with both secD-F and secDF deleted. As S. coelicolor contains at least three xylanases (XlnA, XlnB and XlnC), the XlnA and XlnB are secreted by Sec pathway, whereas the XlnC is secreted by TAT pathway [52]. The basal enzyme activity of total secreted xylanase in S. coelicolor is about 0.1 U/mg dry weight (data not shown). As the reduced amounts of extracellular xylanase activity in strain ZJUZ26 are similar to this basal value, it seems probable that disruption of both secD-F and secDF genes almost completely inhibited XlnA protein secretion.

We further tested the secretion of AmlC that belongs to the 1,4-a-D-glucan glucanohydrolase family in S. coelicolor [53]. The α-amylase was also secreted by Sec pathway in bacteria [54], [55]. A vector, designated pZJ7, was constructed to overexpress AmlC protein. The vector pZJ7 was transformed into S. coelicolor M145, ZJUZ1, ZJUZ2 and ZJUZ3, and the transformants were designated ZJUZ27, ZJUZ28, ZJUZ29, and ZJUZ30, respectively. The extracellular amylase activity was detected by Lugol’s solution stain in a semi-quantitative manner (Figure S4 in File S1) [34]. The relative amylase activity was also decreased in secD-F and/or secDF mutants compared to the wild-type strain, and showed the similar pattern to that of XlnA (Figure 4C). Therefore, the secD-F and/or secDF mutants caused the secretion defect of XlnA and AmlC, two Sec-specific proteins.

The SecDF complex is the accessory component of the Sec translocation system. Generally, SecDF plays important role in protein export by affecting protein translocation efficiency [23], [25]. The two forms SecDF homologs in S. coelicolor are required for efficient secretion of XlnA and AmlC. Since the two SecDF homologs in S. coelicolor seem to be redundant, we examined whether disruption of secDF homologous genes affects the transcription level of each other. qPCR analyses showed the transcription level of secD/secF in ZJU2 (ΔsecDF) and secDF in ZJU1 (ΔsecD-F) was relatively higher than that in the wild-type strain M145 (Figure 3B, C). Taken together, these results suggest that there is some degree of functional redundancy between the two SecDF homologs. However, SecD-F has a much more prominent role than SecDF in the Sec-dependent protein export and the complementation of two forms of SecDF homologs may contribute to higher secretion efficiency of S. coelicolor.

Evolution and origin of SecDF homologous genes in S. coelicolor

Previous studies suggested that membrane proteins, including SecD and SecF, might have evolved through gene duplication and fusion events [21], [56]. To investigate the origin of two copies of secDF homologous genes in S. coelicolor, we performed BLASTp searches for SecD, SecF and SecDF homologs in various organisms (see Methods). The BLAST results were sampled to include representatives from major bacterial lineages, especially those species whose genomes have been completely sequenced. Maximum likelihood (ML) and neighbor-joining (NJ) phylogenetic trees were reconstructed for the S. coelicolor SecDF homologs, and both trees were congruent in topology (Figure 5A). The trees for the SecDF homologs from various taxa showed that a cluster of proteins were grouped into two well-supported distinct clades, SecDF and SecD-F (Figure 5A). The group SecDF is mainly composed of the fusion protein SecDF, while the group SecD-F is comprised exclusively of separated proteins SecD and SecF. Notably, both clades were comprised of those representative species from the phylum Actinobacteria, except Dehalobacter sp. FTH1 which belongs to the phylum Firmicutes. Therefore, these two forms of SecDF homologs in S. coelicolor seem to have evolved early in or before the divergence of extant Actinobacteria species.

Figure 5. Phylogeny and gene synteny of Streptomyces SecDF homologs.

(A) Phylogenetic tree based on the amino acid sequences of SecDF homologues identified by BLAST searches (see Methods) from those species with whole genome sequences available. Background colors refer to two different clades of SecDF and SecD-F. Bootsrtap support values greater than 50% are depicted in tree branches. Streptomyces branches are shown red, Actinobacteria branches are blue and Dehalobacter branches are pink. Species names are abbreviated as given in Table S2 in File S1. (B) Chromosome regions encompassing secDF gene are depicted by black lines for some species in the phylogenetic tree. The secD or secDF homologous genes are shown as red arrow in the center, and the orthologous genes between these species are depicted in the same color according to IMG database (see Methods).

Observation of the gene order within the genomic vicinity of secD-F in several Streptomyces genomes reveals similar neighboring gene identities and orders for both secD-F homologous types (Figure 5B). In addition, the genes surrounding secD-F in all species of the clade SecD-F show a high degree of synteny. In contrast, the genomic neighborhood of secDF gene in S. coelicolor does not show synteny with that of the fused type from other closely related taxa. Our previous studies have shown that a core region of genome has been extraordinarily conserved and undergone strong purifying selection to maintain gene contents in Streptomyces [1], [2]. As the secD and secF genes are located in the core region, they tended to be conserved in Streptomyces genomes. In contrast, the secDF genes in many Streptomyces species are located at terminal plastic regions of genomes, and they tend to be acquired or lost by chromosomal rearrangement events. In addition, the phylogenetic analysis shows that the fused SecDF from the phylum Actinobacteria were more closely related to those homologs from other bacteria (E. coli, B. subtilis and Dehalobacter sp.) than to the separated SecD-F from same phylum (Figure 5). Taken together, these results indicated that secD-F may have arisen through the divergence and vertical transmission of existing components from an ancestral species of Streptomyces, while secDF may have been introduced into the genome of an ancestral species of Actinobacteria by horizontal gene transfer (HGT). The acquisition of a second secDF copy may render a functional benefit to the organism by enhancing protein transport capacity. However, if the secDF gene performs a redundant function it would easily become non-essential when the environment changed. This would lead to a relaxed constraint on the secDF gene, and it could have been lost in some species (such as S. avermitilis) during the evolution of Streptomyces.

Interestingly, Dehalobacter sp. FTH1, which belongs to the phylum Firmicutes, contains two copies of separated SecD and SecF homologs, and the phylogenetic analysis shows one of them is closely related to the fused SecDF and the other is related to the separated SecD/SecF of Streptomyces (Figure 5). The alternative scenario is that the secD and secF genes in the common ancestor of Firmicutes and/or Actinobacteria may have undergone additional gene duplication, and one pair of the duplicate genes were further fused in the Actinobacteria ancestor, leading to the two forms of secDF homologous genes in S. coelicolor.

Taken into consideration of earlier studies [20], [56], a potential evolutionary history of the two forms of secDF homologs in Streptomyces species could be hypothesized as follows (Figure 6). A single primitive gene (ancestor of secD/F) existed first. This gene was then duplicated to form two homologs, secD and secF (such as in E. coli). In a few genomes, a fusion of these two genes into secDF occurred later (such as in B. subtilis). The Actinobacteria ancestor inherited the separated form of secD and secF genes via vertical transmission, but acquired an additional copy of fused secDF gene by HGT. In the later evolutionary process, the fused secDF gene may have become redundant and been lost in some Streptomyces species. Alternatively, it is also plausible that both the fused and separated forms existing in Streptomyces may have been derived from the sequential gene duplication and fusion events occurring in an early progenitor of Actinobacteria. However, this seems unlikely as most extant bacteria have only one form of secDF homologous genes, either in separated or fused form. Furthermore, the separated and fused secDF genes, regardless of their origin, have diverged from each other, as reflected in their striking sequence differences. Clearly, more work is required to confirm this evolution hypothesis and to determine whether other functionalities are influenced by the presence or absence of SecDF proteins.

Figure 6. Proposed schematic model of secDF homologs evolution in Streptomyces.

Supporting Information

File S1.

Figures S1–S4, Tables S1 and S2. Figure S1. Identify the activities of secD-F and secDF gene promoters and gene expression profile. (A) Expression level of egfp in LM6 and ZJUZ39-41 by qPCR, the expression of egfp in LM1 was set as reference. (B) Detecting the expression of EGFP protein alone the time of cultivation by SDS-PAGE and Western blotting. WB: Western blotting; CB: Commassie Blue Staining. Figure S2. PCR and Southern blot verification of gene knock-out. (A) PCR analysis for disruption of secD-F genes. The primer pair secD_p_F vs secF_right_arm_R (Table S1) was used. (B) PCR analysis for disruption of secDF gene. The primer pair secDF_p_F vs secDF_right_arm_R (Table S1) was used. (C) Southern blot analysis for disruption of secD-F genes, digested by BamHI. (D) Southern blot analysis for disruption of secDF gene, digested by BamHI. Figure S3. Phenotypic analysis on morphogenesis between S. coelicolor wild type and mutants. (A) MM 60 h, (B) SMMS 84 h, (C) MSF 72 h, (D) R5 36 h. Figure S4. Semi-quantitative assay of extracellular AmlC activity. Strains ZJUZ27-30 were grown on MM media containing 0.2% soluble starch for 5 days before staining with Lugol’s solution. Table S1. Oligonucleotides used in this study. Table S2. Distribution of SecDF homologs in Streptomyces species and other species depicted in Figure 4.



We thank Dr. Yiling Du from the University of British Columbia for critical reading of this manuscript.

Author Contributions

Conceived and designed the experiments: ZZ YDL YQL. Performed the experiments: ZZ YDL NS ZS. Analyzed the data: ZZ YDL. Contributed reagents/materials/analysis tools: ZZ NS ZS LL. Contributed to the writing of the manuscript: ZZ YDL LS YW YQL.


  1. 1. Zhou Z, Gu J, Du YL, Li YQ, Wang Y (2011) The -omics Era- Toward a Systems-Level Understanding of Streptomyces. Curr Genomics 12: 404–416.
  2. 2. Zhou Z, Gu J, Li YQ, Wang Y (2012) Genome plasticity and systems evolution in Streptomyces. BMC Bioinformatics 13 Suppl 10S8.
  3. 3. Hopwood DA (2006) Soil to genomics: the Streptomyces chromosome. Annu Rev Genet 40: 1–23.
  4. 4. Widdick DA, Dilks K, Chandra G, Bottrill A, Naldrett M, et al. (2006) The twin-arginine translocation pathway is a major route of protein export in Streptomyces coelicolor. Proc Natl Acad Sci U S A 103: 17927–17932.
  5. 5. Chater KF, Biro S, Lee KJ, Palmer T, Schrempf H (2010) The complex extracellular biology of Streptomyces. FEMS Microbiol Rev 34: 171–198.
  6. 6. Morosoli R, Shareck F, Kluepfel D (1997) Protein secretion in streptomycetes. FEMS Microbiol Lett 146: 167–174.
  7. 7. Anne J, Vrancken K, Van Mellaert L, Van Impe J, Bernaerts K (2014) Protein secretion biotechnology in Gram-positive bacteria with special emphasis on Streptomyces lividans. BBA-Mol Cell Res 1843: 1750–1761.
  8. 8. Anne J, Maldonado B, Van Impe J, Van Mellaert L, Bernaerts K (2012) Recombinant protein production and streptomycetes. J Biotechnol 158: 159–167.
  9. 9. Vrancken K, Anne J (2009) Secretory production of recombinant proteins by Streptomyces. Future Microbiol 4: 181–188.
  10. 10. Li YD, Zhou Z, Lv LX, Hou XP, Li YQ (2009) New approach to achieve high-level secretory expression of heterologous proteins by using Tat signal peptide. Protein Pept Lett 16: 706–710.
  11. 11. Driessen AJ, Nouwen N (2008) Protein translocation across the bacterial cytoplasmic membrane. Annu Rev Biochem 77: 643–667.
  12. 12. Driessen AJ, Fekkes P, van der Wolk JP (1998) The Sec system. Curr Opin Microbiol 1: 216–222.
  13. 13. Chatzi KE, Sardis MF, Karamanou S, Economou A (2013) Breaking on through to the other side: protein export through the bacterial Sec system. Biochem J 449: 25–37.
  14. 14. Lycklama ANJA, Driessen AJ (2012) The bacterial Sec-translocase: structure and mechanism. Philos Trans R Soc Lond B Biol Sci 367: 1016–1028.
  15. 15. du Plessis DJ, Nouwen N, Driessen AJ (2011) The Sec translocase. BBA-Biomembranes 1808: 851–865.
  16. 16. Hsieh YH, Zhang H, Wang H, Yang H, Jiang C, et al. (2013) Reconstitution of functionally efficient SecA-dependent protein-conducting channels: transformation of low-affinity SecA-liposome channels to high-affinity SecA-SecYEG-SecDF.YajC channels. Biochem Biophys Res Commun 431: 388–392.
  17. 17. Economou A, Pogliano JA, Beckwith J, Oliver DB, Wickner W (1995) SecA membrane cycling at SecYEG is driven by distinct ATP binding and hydrolysis events and is regulated by SecD and SecF. Cell 83: 1171–1181.
  18. 18. Duong F, Wickner W (1997) The SecDFyajC domain of preprotein translocase controls preprotein movement by regulating SecA membrane cycling. EMBO J 16: 4871–4879.
  19. 19. Tsukazaki T, Mori H, Echizen Y, Ishitani R, Fukai S, et al. (2011) Structure and function of a membrane component SecDF that enhances protein export. Nature 474: 235–238.
  20. 20. Eichler J (2003) Evolution of the prokaryotic protein translocation complex: a comparison of archaeal and bacterial versions of SecDF. Mol Phylogenet Evol 27: 504–509.
  21. 21. Tseng TT, Gratwick KS, Kollman J, Park D, Nies DH, et al. (1999) The RND permease superfamily: an ancient, ubiquitous and diverse family that includes human disease and development proteins. J Mol Microbiol Biotechnol 1: 107–125.
  22. 22. Gardel C, Johnson K, Jacq A, Beckwith J (1990) The secD locus of E. coli codes for two membrane proteins required for protein export. EMBO J 9: 4205–4206.
  23. 23. Bolhuis A, Broekhuizen CP, Sorokin A, van Roosmalen ML, Venema G, et al. (1998) SecDF of Bacillus subtilis, a molecular Siamese twin required for the efficient secretion of proteins. J Biol Chem 273: 21217–21224.
  24. 24. Quiblier C, Zinkernagel AS, Schuepbach RA, Berger-Bachi B, Senn MM (2011) Contribution of SecDF to Staphylococcus aureus resistance and expression of virulence factors. BMC Microbiol 11: 72.
  25. 25. Pogliano JA, Beckwith J (1994) SecD and SecF facilitate protein export in Escherichia coli. EMBO J 13: 554–561.
  26. 26. Bentley SD, Chater KF, Cerdeno-Tarraga AM, Challis GL, Thomson NR, et al. (2002) Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature 417: 141–147.
  27. 27. Page N, Kluepfel D, Shareck F, Morosoli R (1996) Effect of signal peptide alterations and replacement on export of xylanase A in Streptomyces lividans. Appl Environ Microbiol 62: 109–114.
  28. 28. Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA (2000) Practical Streptomyces Genetics. Norwich, England: The John Innes Foundation.
  29. 29. Bierman M, Logan R, O'Brien K, Seno ET, Rao RN, et al. (1992) Plasmid cloning vectors for the conjugal transfer of DNA from Escherichia coli to Streptomyces spp. Gene 116: 43–49.
  30. 30. Green MR, Sambrook J (2012) Molecular Cloning: A Laboratory Manual. New York: Cold Spring Harbor Laboratory Press.
  31. 31. Sun J, Kelemen GH, Fernandez-Abalos JM, Bibb MJ (1999) Green fluorescent protein as a reporter for spatial and temporal gene expression in Streptomyces coelicolor A3(2). Microbiology 145: 2221–2227.
  32. 32. Mao XM, Zhou Z, Hou XP, Guan WJ, Li YQ (2009) Reciprocal regulation between SigK and differentiation programs in Streptomyces coelicolor. J Bacteriol 191: 6473–6481.
  33. 33. Yang R, Hu Z, Deng Z, Li J (1998) Construction of Escherichia coli-Streptomyces shuttle expression vectors for gene expression in Streptomyces. Chin J Biotechnol 14: 1–8.
  34. 34. Widdick DA, Eijlander RT, van Dijl JM, Kuipers OP, Palmer T (2008) A facile reporter system for the experimental identification of twin-arginine translocation (Tat) signal peptides from all kingdoms of life. J Mol Biol 375: 595–603.
  35. 35. Widdick DA, Dilks K, Chandra G, Bottrill A, Naldrett M, et al. (2006) The twin-arginine translocation pathway is a major route of protein export in Streptomyces coelicolor. Proc Natl Acad Sci U S A 103: 17927–17932.
  36. 36. Boratyn GM, Camacho C, Cooper PS, Coulouris G, Fong A, et al. (2013) BLAST: a more efficient report with usability improvements. Nucleic Acids Res 41: W29–33.
  37. 37. Markowitz VM, Chen IM, Palaniappan K, Chu K, Szeto E, et al. (2014) IMG 4 version of the integrated microbial genomes comparative analysis system. Nucleic Acids Res 42: D560–567.
  38. 38. Tatusov RL, Fedorova ND, Jackson JD, Jacobs AR, Kiryutin B, et al. (2003) The COG database: an updated version includes eukaryotes. BMC Bioinformatics 4: 41.
  39. 39. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948.
  40. 40. Finn RD, Bateman A, Clements J, Coggill P, Eberhardt RY, et al. (2014) Pfam: the protein families database. Nucleic Acids Res 42: D222–230.
  41. 41. Letunic I, Doerks T, Bork P (2012) SMART 7: recent updates to the protein domain annotation resource. Nucleic Acids Res 40: D302–305.
  42. 42. Krogh A, Larsson B, von Heijne G, Sonnhammer EL (2001) Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305: 567–580.
  43. 43. Reese MG (2001) Application of a time-delay neural network to promoter annotation in the Drosophila melanogaster genome. Comput Chem 26: 51–56.
  44. 44. Talavera G, Castresana J (2007) Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol 56: 564–577.
  45. 45. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013) MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30: 2725–2729.
  46. 46. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, et al. (2010) New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59: 307–321.
  47. 47. Nouwen N, Piwowarek M, Berrelkamp G, Driessen AJ (2005) The large first periplasmic loop of SecD and SecF plays an important role in SecDF functioning. J Bacteriol 187: 5857–5860.
  48. 48. Bibb MJ, Janssen GR, Ward JM (1985) Cloning and analysis of the promoter region of the erythromycin resistance gene (ermE) of Streptomyces erythraeus. Gene 38: 215–226.
  49. 49. Nouwen N, Driessen AJ (2005) Inactivation of protein translocation by cold-sensitive mutations in the yajC-secDF operon. J Bacteriol 187: 6852–6855.
  50. 50. Pogliano KJ, Beckwith J (1993) The Cs sec mutants of Escherichia coli reflect the cold sensitivity of protein export itself. Genetics 133: 763–773.
  51. 51. Morosoli R, Dupont C (1999) Secretion of xylanase A2 in Streptomyces lividans: dependence on signal peptides length, number and composition. FEMS Microbiol Lett 179: 437–445.
  52. 52. Geukens N, Rao CVS, Mellado RP, Frederix F, Reekmans G, et al. (2006) Surface plasmon resonance-based interaction studies reveal competition of Streptomyces lividans type I signal peptidases for binding preproteins. Microbiology 152: 1441–1450.
  53. 53. Yin XH, Gerbaud C, Francou FX, Guerineau M, Virolle MJ (1998) amlC, another amylolytic gene maps close to the amlB locus in Streptomyces lividans TK24. Gene 215: 171–180.
  54. 54. Lammertyn E, Desmyter S, Schacht S, Van Mellaert L, Anne J (1998) Influence of charge variation in the Streptomyces venezuelae alpha-amylase signal peptide on heterologous protein production by Streptomyces lividans. Appl Microbiol Biotechnol 49: 424–430.
  55. 55. Mulder KC, Bandola J, Schumann W (2013) Construction of an artificial secYEG operon allowing high level secretion of alpha-amylase. Protein Expr Purif 89: 92–96.
  56. 56. Hennerdal A, Falk J, Lindahl E, Elofsson A (2010) Internal duplications in alpha-helical membrane protein topologies are common but the nonduplicated forms are rare. Protein Sci 19: 2305–2318.