Nitrogen regulator GlnR controls uptake and utilization of non ... - PNAS

Nitrogen regulator GlnR controls uptake and utilization of non-phosphotransferase-system carbon sources in actinomycetes Cheng-Heng Liaoa, Lili Yaoa, Ya Xua, Wei-Bing Liua, Ying Zhoua, and Bang-Ce Ye (叶邦策)a,b,1 a Lab of Biosystems and Microanalysis, State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China; and bSchool of Chemistry and Chemical Engineering, Shihezi University, Xinjiang 832000, China

The regulatory mechanisms underlying the uptake and utilization of multiple types of carbohydrates in actinomycetes remain poorly understood. In this study, we show that GlnR (central regulator of nitrogen metabolism) serves as a universal regulator of nitrogen metabolism and plays an important, previously unknown role in controlling the transport of non-phosphotransferase-system (PTS) carbon sources in actinomycetes. It was observed that GlnR can directly interact with the promoters of most (13 of 20) carbohydrate ATP-binding cassette (ABC) transporter loci and can activate the transcription of these genes in response to nitrogen availability in industrial, erythromycin-producing Saccharopolyspora erythraea. Deletion of the glnR gene resulted in severe growth retardation under the culture conditions used, with select ABC-transported carbohydrates (maltose, sorbitol, mannitol, cellobiose, trehalose, or mannose) used as the sole carbon source. Furthermore, we found that GlnR-mediated regulation of carbohydrate transport was highly conserved in actinomycetes. These results demonstrate that GlnR serves a role beyond nitrogen metabolism, mediating critical functions in carbon metabolism and crosstalk of nitrogen- and carbon-metabolism pathways in response to the nutritional states of cells. These findings provide insights into the molecular regulation of transport and metabolism of non-PTS carbohydrates and reveal potential applications for the cofermentation of biomass-derived sugars in the production of biofuels and bio-based chemicals.

|

actinomycetes GlnR nitrogen metabolism

| carbon sources utilization | ABC transport system |

and two nucleotide-binding domains with ATPase activity (6, 7). An additional component forms the substrate-binding proteins domains (BPDs) that are particularly required in prokaryotes (8). The most extensively characterized carbohydrate ABC transporter is the maltose transport system MalEFGK2 of E. coli (9, 10), which provides a prototypic model for the study of carbohydrate ABC transport systems. Actinomycetes, with some species serving as representative plant-biomass decomposers, use a wide variety of secondary carbon sources because of their natural habitat (i.e., soil-dwelling) and their considerably large gene sets for carbon ABC transporters encoded in their genomes (11, 12). In a model strain of Streptomyces, namely Streptomyces coelicolor A3 (2), the ABC transporters represent ∼87% of the entire set of carbohydrate transport systems encoded in the genome (11), whereas this proportion in Mycobacterium tuberculosis is 68% (12). As one of the largest bacterial genera, actinomycetes are well known as prolific producers of numerous antibiotics (13), biofuels, materials, and commodity chemicals. Owing to their capacity for transporting multiple carbon substrates, industrial actinomycetes are potential microbial cell factories for biorefinery and fermentation processes. Typically, agricultural and forest residues are abundant and economical carbon nutrient feedstocks consisting mainly of lignocellulose, which in turn is composed of cellulose, hemicellulose, and lignin (14). These components can be subsequently broken down into a heterogeneous mixture of Significance

M

icroorganisms that can simultaneously couse multiple carbohydrates are of considerable interest for the biological-based conversion of biomass to fuels and chemicals. Most microorganisms have evolved tailored carbon-utilization pathways and regulatory mechanisms [such as carbon catabolite repression (CCR) and other multiply coordinated mechanisms] for the sequential utilization of sugars from a combination of carbon sources, including lignocellulose-derived sugar mixtures. The CCR process ensures that microorganisms first use preferred (i.e., readily metabolized) carbon sources such as glucose, which are generally imported via the phosphotransferase system (PTS). Recently, it was found that, upon inactivation of the PTS system, an alternative glucose transport system (GalP permease) exists that can been used for the efficient and fast production of succinate in Escherichia coli (1). CCR influences carbon utilization through the repression of genes encoding enzymes involved in the uptake and catabolism of nonpreferred carbon sources (non-PTS carbon sources), which in turn increases the sugar-uptake capability and promotes faster growth (2). Ensuring the sequential utilization of sugars is a major technical challenge for increasing the yield and productivity of industrial microorganisms. Most less-preferred sugars are taken into cells by the ATP-binding cassette (ABC) transport systems, which are the largest group of carbohydrate-transport systems found in bacteria (3–5). The canonical architecture of the carbohydrate ABC transport systems consists of two transmembrane domains that form a substrate translocation channel

www.pnas.org/cgi/doi/10.1073/pnas.1508465112

Actinomycetes abundantly generate beneficial metabolic products. The efficient coutilization of heterogeneous carbon sources remains a major technical challenge for the industrial-scale production of drugs, chemicals, materials, and fuels by actinomycetes. Here, we present insights into the regulatory mechanisms of uptake and utilization of multiple carbohydrates in actinomycetes. GlnR (central regulator of nitrogen metabolism) was shown to regulate the control of ATP-binding cassette transport systems for secondary, non-phosphotransferase-system carbon sources. By integrating nitrogen signals to modulate the uptake and utilization of multiple carbon sources, GlnR mediates the interplay between nitrogen and carbon metabolism. These findings highlight the potential of actinomycetes in carbon utilization (especially cofermentation of biomass-derived sugars) for biorefinery applications. To our knowledge, our data represent the first systematic description of GlnR-mediated regulation of carbohydrate metabolism in actinomycetes. Author contributions: C.-H.L. and B.-C.Y. designed research; C.-H.L. performed research; C.-H.L., L.Y., Y.X., W.-B.L., Y.Z., and B.-C.Y. contributed new reagents/analytic tools; C.-H.L. and B.-C.Y. analyzed data; and C.-H.L. and B.-C.Y. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1

To whom correspondence should be addressed. Email: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1508465112/-/DCSupplemental.

PNAS Early Edition | 1 of 6

APPLIED BIOLOGICAL SCIENCES

Edited by Arnold L. Demain, Drew University, Madison, NJ, and approved November 11, 2015 (received for review April 30, 2015)

fermentable sugars, such as cellobiose, xylose, glucose, arabinose, mannose, and galactose. CCR and other regulatory mechanisms underlying the uptake and utilization of multiple carbohydrates represent major hurdles that need to be overcome to facilitate the more efficient use of biomass-derived sugar mixtures. The CCR regulatory mechanisms have been intensively studied in the model bacteria Bacillus subtilis and E. coli. However, the molecular mechanisms by which CCR and multiple interacting regulatory pathways influence carbohydrate metabolism in actinomycetes remain unknown. Recently, we found that GlnR (central regulator of nitrogen metabolism) serves as a universal regulator of nitrogen metabolism in the industrial erythromycin producer Saccharopolyspora erythraea (15) and might play a role in the crosstalk between nitrogen and carbon metabolism-associated pathways. To better understand how GlnR affects carbon metabolisms in actinomycetes, we investigated how GlnR-mediated gene transcription is involved in the uptake and metabolism of multiple carbohydrates. We found that GlnR plays an important and extraordinary role in controlling the transport of multiple non-PTS carbon sources in S. erythraea. Furthermore, it was observed that the GlnRmediated regulation of carbohydrates transport is highly conserved in actinomycetes. These findings provide insights into the regulatory mechanisms of the transport and metabolism of non-PTS carbohydrates and highlight potential applications for cofermentation of biomass-derived sugars in the production of biofuels and bio-based chemicals by actinomycetes. Results GlnR Directly Controls the Transport and Utilization of Maltose in S. erythraea. Previously, the maltose/maltodextrin-transport system

was intensively characterized in E. coli (9) and in a model strain of S. coelicolor (16), both of which use a canonical ABC transporter encoded by the gene cluster designated as malEFG (4). The malEFG homologs in the S. erythraea genome have been annotated in the KEGG (Kyoto Encyclopedia of Genes and Genomes) database (www.genome.jp/kegg/). The organization of the loci within the SACE_1664-1665-1666 gene cluster was similar to that observed in the genomes of S. coelicolor (16) and other actinomycetes (SI Appendix, Fig. S1A). The sequence of

the gene encoding the maltose-binding protein (SACE_1666) showed ∼27–42% sequence similarity to homologs from other actinomycetes. Quantitative real-time RT-PCR (qPCR) analysis showed that transcription of the homologous loci was exclusively induced when maltose, maltodextrin, or starch was used as the sole carbon source (SI Appendix, Fig. S1B). Thus, we herein designated the S. erythraea locus as malEFG. On the opposite DNA strand (relative to malEFG) lays a putative cluster-specific regulator gene (malR) and an alpha-glucosidase gene (aglA), as depicted in Fig. 1A. Cotranscription analysis by semiquantitative RT-PCR indicated that malEFG and malRaglA represent divergently transcribed operons (SI Appendix, Fig. S1C). Using a previous training set developed for GlnR binding-site prediction (15), two potential GlnR cis-elements (GlnR-1 and GlnR-2) were identified in the intergenic region of the malEFG and malR-aglA operons (Fig. 1A and SI Appendix, Fig. S2A). Electrophoretic mobility shift assays (EMSAs) revealed a direct binding interaction, as DNA probes containing the intergenic region clearly shifted following incubation with purified recombinant GlnR (Fig. 1B), indicating that the GlnR-1 cis-element was the major binding site in the mal operon (SI Appendix, Fig. S2 A and B). To elucidate the regulatory effect of GlnR on maltose utilization, we constructed S. erythraea glnR-deletion (ΔglnR) strains and glnR complementary (ΔglnR::glnR) strains, as described (15, 17). It was observed that glnR deletion resulted in severe growth retardation in liquid minimal media with maltose as the sole carbon source, whereas growth occurred normally in glucosecontaining medium (Fig. 1C). Complementation of the glnR gene restored growth on maltose (Fig. 1C). In addition, the ΔglnR strain also showed impaired growth on minimal agar medium, with maltose as the sole carbon source (SI Appendix, Fig. S3A). Scanning-electron microscope (SEM) imaging revealed the presence of withered mycelia in the ΔglnR strain grown on maltosecontaining agar, whereas normal mycelia formed during logarithmic phase in the presence of glucose (SI Appendix, Fig. S3B). These results indicated that GlnR plays a crucial role in maltose utilization in S. erythraea. The qPCR data showed that malEFG transcript levels were markedly decreased (12- to 39-fold) in the ΔglnR strain (Fig.

Fig. 1. GlnR directly controls the maltose-transport system in S. erythraea. (A) Putative GlnR cis-elements in the maltose/maltodextrin transporter locus in S. erythraea. The positions of the cis-elements are numbered relative to the translational start site of malE. (B) EMSA results revealed binding of GlnR to the malE-malR intergenic region. Biotin-labeled DNA fragments were incubated with increasing concentrations of His-GlnR (0, 0.2, 0.4, and 0.8 μM). (C) Growth of the wild-type (WT), ΔglnR, and ΔglnR::glnR strains in minimal medium supplemented with 0.5% glucose or maltose as the sole carbon source. (D) qPCR analysis of the relative transcription levels of the mal locus between the WT and ΔglnR strains. Cells were grown in nitrogen-balanced minimal medium supplemented with 0.5% maltose as the sole carbon source. The expression levels of genes in the WT strain were arbitrarily set to 1.0. Error bars indicate SD from three independent experiments. **P < 0.01; t test. (E) Sugar uptake in the WT, ΔglnR, and glnR overexpression (WT::glnR) strains. Cells grown on 1% maltose or glucose were collected at logarithmic phase and reincubated with 5 mM maltose or glucose. Residual sugar concentrations in each case were determined as described in Materials and Methods. The cell weights used were identical for each of the sugar-transport assays.

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1508465112

Liao et al.

GlnR Mediates the Transport and Utilization of Maltose in Response to Nitrogen Availability. To investigate the impact of nitrogen

availability on maltose utilization, we compared the transcriptional responses of the mal locus under defined nitrogen-rich (1% wt/vol tryptone) and nitrogen-limited (1 mM ammonium sulfate) conditions. We found that the transcript levels from the mal locus were approximately ninefold higher under the nitrogen-limited condition, relative to the nitrogen-rich condition (Fig. 2A). The maltose uptake rate of the cells subjected to nitrogen-limited cultivation was also higher than that in the nitrogen-rich condition (Fig. 2B). Under the maltose-free condition, transcription from the mal locus was induced by ∼120- to 200-fold under the nitrogen-limited condition (SI Appendix, Fig. S4A). Transcript levels of malE and glnR were also examined over the nitrogen down-shift. We found that malE and glnR were induced following nitrogen deprivation (Fig. 2 C and D). Notably, malE and glnR displayed similar transcription profiles at different time points. Moreover, glnR overexpression resulted in a 9- to 10-fold induction of the mal locus in the WT::glnR strain (SI Appendix, Fig. S4B), thus further indicating that GlnR plays a role in mal transcription in response to nitrogen availability. GlnR Exerts Regulatory Effects on Most Carbohydrate ABC Transport Systems in S. erythraea. Next, we investigated whether GlnR con-

trols the functions of other ABC transport systems involved in

Fig. 2. GlnR-mediated induction of malE during nitrogen starvation. (A) Induction of the mal locus during nitrogen starvation. S. erythraea (WT) was inoculated in minimal medium supplemented with 1% (wt/vol) tryptone or 1 mM ammonium sulfate, as nitrogen sources. The relative expression of malE in tryptone was set to 1.0. (B) Sugar uptake of S. erythraea (WT) grown in identical cultivated conditions as in A. Cells were collected at logarithmic phase and reincubated with 5 mM maltose. The cell weights used were identical for each of the sugar-transport assays. (C and D) Induction of malE and glnR during nitrogen deprivation. S. erythraea (WT) was grown in TSB medium with 1% maltose until logarithmic phase, after which the cells were collected, washed, and transferred to minimal medium supplemented with 1% maltose and 1 mM ammonium sulfate. RNA was isolated from mycelia collected at logarithmic phase and at 4and 8-h incubation after the transfer. Results are representative of three independent experiments. Error bars indicate SD from triplicates. **P < 0.01; *P < 0.05.

Liao et al.

carbohydrate uptake in S. erythraea. By searching the NCBI (National Center for Biotechnology Information) and KEGG databases, 19 additional putative carbohydrate ABC transporters were identified in the genome of S. erythraea (SI Appendix, Table S1 and Fig. S5). Similarly, potential corresponding GlnR cis-elements were discovered in the promoter regions of eight transporters (SI Appendix, Table S1). EMSA results showed that 13 probes (including 8 with potential GlnR cis-elements) were directly bound by GlnR (SI Appendix, Fig. S6). The transcription levels of genes encoding BPDs representative of each ABC transporter were examined in the WT, ΔglnR, and WT::glnR strains. We found that the BPD-encoding genes were down-regulated in the ΔglnR mutant, but up-regulated in the glnR-overexpression strain, except for SACE_0943, _1890, _2989, and _6230 (SI Appendix, Fig. S7). These results suggested that GlnR served as a general activator of most carbohydrate ABC transporters. To investigate the effect of GlnR on carbohydrate utilization, the growth of WT, glnR-deleted, and glnR-complementary S. erythraea strains were tested by using nine different carbohydrates as sole carbon sources (SI Appendix, Fig. S8). Compared with the WT strain, growth of the ΔglnR strain was impaired on several carbohydrate sources, including sorbitol, mannitol, cellobiose, trehalose, and mannose. Complementation of glnR restored growth on these carbohydrates. In contrast, no differences were observed when these strains were grown on ribose, xylose, arabinose, or lactose, indicating that GlnR-independent utilization pathways for these carbohydrates exist in S. erythraea. Overexpression of glnR increased the growth rate and biomass accumulation in these identified carbon sources whose utilization was significantly affected by glnR deletion (SI Appendix, Fig. S9). Furthermore, GlnR enhanced the utilization of multiple carbohydrates in mixed carbon conditions (such as glucose and xylose or cellobiose and xylose) (SI Appendix, Fig. S10). The substrates recognized and transported by all carbohydrate ABC transporters were determined by analyzing the transcriptional responses of BPD-encoding genes with the ABC systems in response to various carbohydrates as the sole carbon source (SI Appendix, Fig. S11). The carbohydrate ABC transport systems present in S. erythraea as well as a model of GlnR regulation are presented in Fig. 3. GlnR-Mediated Regulation of Carbohydrate ABC Transporters Is Conserved in Actinomycetes. Next, we examined whether GlnR

regulates the transcription of genes involved in carbohydrate uptake and utilization in other actinomycetes species. We screened all genes that putatively encode carbohydrate ABC transporters and searched for GlnR cis-elements in the promoter regions of six well studied actinomycetes species. Many GlnR cis-elements were found in the regulatory regions of genes encoding carbohydrate ABC transporters: 11 of 44 for Amycolatopsis mediterranei, 7 of 20 for Mycobacterium smegmatis, 12 of 37 for S. coelicolor, 11 of 29 for Streptomyces griseus, 11 of 28 for Streptomyces venezuelae, and 13 of 45 for Streptomyces avermitilis (SI Appendix, Tables S2–S7). GlnR ciselements derived from the ABC transporters of each species show high similarity and conservation (SI Appendix, Fig. S12 A and B). In silico analysis suggested that the regulation by GlnR of carbohydrate ABC transporters occurs in other actinomycetes species as well. We found that the maltose/maltodextrin transport systems in all investigated actinomycetes are highly conserved and harbor a typical GlnR cis-element (Fig. 4A). EMSA results showed that the S. coelicolor, S. avermitilis, and M. smegmatis GlnR regulators directly bound to the corresponding upstream regions of the mal operon (Fig. 4B). Similar to the findings for S. erythraea, deletion of the glnR gene resulted in severe growth retardation of S. coelicolor using maltose as the sole carbon source (Fig. 4C) and low transcript levels from the mal operon SCO2228-29-30-31, which were down-regulated by 2.5- to 4-fold compared with those observed with the WT M145 strain (Fig. 4D). The S. coelicolor ΔglnR mutant exhibited a deficiency of maltose uptake (Fig. PNAS Early Edition | 3 of 6


1D). The ΔglnR strain complemented with the glnR gene showed similar malEFG transcription levels, as did the wild-type (WT) strain (SI Appendix, Fig. S3C). Furthermore, low expression of the maltose transporter in the ΔglnR strain resulted in the loss of maltose-uptake capability, whereas additional glnR expression in the glnR overexpression strain WT::glnR (Materials and Methods) increased the maltose uptake rate relative to the WT strain (Fig. 1E). As a control, no clear difference in glucose uptake was observed among the strains (Fig. 1E). Analysis by qPCR revealed that aglA expression was also down-regulated in the ΔglnR strain, whereas it was threefold up-regulated in the WT::glnR strain compared with the WT strain (SI Appendix, Fig. S3D). Taken together, these results showed an essential role for GlnR in maltose transportation and degradation by direct control of the expression of genes involved in maltose metabolism.

Fig. 3. Summary of carbohydrate ABC transport systems in S. erythraea and a model of GlnR regulation. Specific substrates for each ABC transporter are indicated. A model for ABC transporter regulation by GlnR and its interplay with catabolite repression is proposed. The ABC transporters highlighted in light blue were up-regulated by GlnR, those highlighted in gray were bound directly by GlnR without an obvious regulatory effect, and those highlighted in white showed no effect. The schematic is organized based on the experimental data obtained in this study.

4E), suggesting that the maltose uptake system is a general target of GlnR in actinomycetes. The ABC transport systems involved in xylose/xylobiose transportation were also investigated in S. coelicolor. Three loci were shown to be bound by GlnR, as verified by EMSAs (Fig. 4F). qPCR analysis suggested that two of the three loci were down-regulated (Fig. 4G), and slight growth defects in xylose media were observed (Fig. 4H) upon glnR deletion. These results suggested that xylose utilization is also subjected to GlnR regulation. Discussion GlnR is an OmpR-like response regulator found extensively in actinomycetes and other Gram-positive bacteria that typically acts as a general regulator of genes related to nitrogen assimilation

and metabolism (18). Previous findings have identified GlnRcontrolled genes involved in carbon and phosphate metabolism, as well as secondary metabolism (19–21). In the present work, we found that GlnR positively regulates the transport and utilization of multiple sugars in S. erythraea. A set of 20 ABC transport systems was identified in the S. erythraea genome (Fig. 3 and SI Appendix, Fig. S5), over half of which were directly transcriptionally controlled by GlnR. Deletion of the glnR gene caused severe growth deficiency using several sole-carbon sources (maltose, mannose, trehalose, sorbitol, and mannitol), which are translocated by ABC transporters in S. erythraea and S. coelicolor. Bioinformatics analysis and biochemical assay results revealed that GlnR-mediated regulation of carbohydrate transport is highly conserved in actinomycetes. Furthermore, we found

Fig. 4. GlnR-dependent regulation of the maltose and xylose transport systems is conserved in S. coelicolor and other actinomycetes. (A) Alignment of GlnR cis-elements in the mal-operon promoter regions from various actinomycetes. (B) EMSA results confirmed the ability of GlnR to bind the mal loci in three actinomycetes. (C) Growth of the S. coelicolor strain M145 and the glnR mutant strain ΔglnR in minimal liquid and agar medium supplemented with 0.5% maltose as the sole carbon source. (D) qPCR analysis of the relative transcription levels of the S. coelicolor mal locus in the M145 and ΔglnR strains. Cells were grown in nitrogenbalanced minimal medium supplemented with 0.5% maltose as the sole carbon source. The expression levels of genes in the WT strain were set to 1.0. (E) Uptake of maltose in WT S. coelicolor and the ΔglnR mutant. (F) EMSA results revealed binding of GlnR to the xylose transporters in S. coelicolor. (G) qPCR analysis of the relative transcription levels of the S. coelicolor xylose loci in the M145 and ΔglnR strains. (H) Growth of the S. coelicolor M145 and ΔglnR strains in agar and minimal liquid medium supplemented with 0.5% xylose as the sole carbon source. Results are representative of three independent experiments. Error bars indicate SD from triplicates. n.s., not significant. *P < 0.05.


Liao et al.

Liao et al.

and CRP for the mal locus was indicated (SI Appendix, Fig. S13D). Therefore, the competition and crosstalk of GlnR with CSR regulators and with the CRP protein might reflect a complicated regulatory mechanism underlying carbohydrate metabolism. In the past decade, the uses of microbial cell factories for producing drugs, fuels, and various commodity chemicals from renewable biomass-derived sugars have been intensely studied, particularly those involving metabolic-engineering strategies based on CCR regulation and sugar transportation. However, the development of efficient strategies for the cofermentation of heterogeneous carbon sources remains a major technical challenge in the fermentation industry, owing to the overwhelming complexity of the multiple coordinating mechanisms underlying carbohydrate metabolism. The results from this study suggest interesting possibilities for enhancing the cofermentation efficiency of these carbon sources via nitrogen signal-stimulation through the global regulator GlnR. Specifically, two strategies could be put into practice in real applications. First, the genetic engineering approaches could be used for overexpression of the glnR gene, and thus enhance the utilization rate of sugars. Second, nitrogen could be used as a key regulatory signal during the fermentation process by changing the C/N ratio to obtain optimal sugar utilization at given phases. Materials and Methods Bacterial Strains and Media. The bacterial strains and plasmids used in this study are listed in SI Appendix, Table S8. All S. erythraea and S. coelicolor strains were grown and stored as described (29). R2YE agar plates were used for sporulation. An agar piece of ∼1 cm2 inoculated into a 250-mL flask containing 50 mL of tryptic soy broth (TSB) medium was grown for 48 h at 30 °C and 200 rpm for seed-stock preparation. Minimal Evans medium (30) supplemented with various amounts of carbon or nitrogen was used for phenotype and transcription studies. The E. coli strains DH5α and BL21(DE3) (Novagen) were used for DNA cloning and heterologous GlnR production, respectively. Production of GlnR homologs in E. coli. The glnR gene homologs were amplified by PCR using primers containing NcoI and HindIII restriction sites (SI Appendix, Table S9) from wild-type strains and cloned into the pET-28a(+) vector (with a His-tag) to generate the recombinant plasmids shown in SI Appendix, Table S8. After validation of the gene sequences by sequencing, the recombinant plasmids were introduced into E. coli BL21(DE3). E. coli cells were selected by growth in kanamycin LB medium and grown in 50 mL of LB medium at 37 °C with 25 mg·mL−1 kanamycin in an orbital shaker (250 rpm, 37 °C) to an OD 600 of 0.6. Protein expression was induced with isopropyl β-D-1-thiogalactopyranoside (final concentration 0.5 mM), followed by incubation at 20 °C for 6–8 h. For protein purification, cells were harvested by centrifugation, washed twice with PBS (pH 8.0), and lysed using an ultrasonic cell crusher. Cell debris and membrane fractions were separated from the soluble fraction by centrifugation (45 min, 17,000 × g, 4 °C). Proteins were purified by using Ni-NTA Superflow columns (Qiagen) and eluted with 250 mM imidazole in 50 mM NaH2PO4 and 300 mM NaCl (pH 8.0). The purified proteins were dialyzed overnight in protein-preservation buffer [50 mM Tris, 0.5 mM EDTA, 50 mM NaCl, 20% (vol/vol) glycerol, 1 mM DTT; pH 8.0] at 4 °C and stored at −80 °C until use. The quality of the purified proteins was determined by SDS polyacrylamide gel electrophoresis (SDS/PAGE). Protein concentrations were determined by using the Bradford assay method. Strain Constructions. Gene deletion and complementation experiments in S. erythraea NRRL23338 were performed as described (17). An in-frame deletion of the 771-nt fragment of the glnR gene (gene ID: SACE_7101) was replaced by the thiostrepton-resistance cassette. For glnR overexpression, the E. coli–S. erythraea integrative shuttle vector pIB-glnR (SI Appendix, Table S8) was introduced into wild-type S. erythraea NRRL23338 by polyethylene glycol-mediated transformation, generating a glnR-overexpression strain harboring an extra copy of glnR. All primers used for strain constructions are listed in SI Appendix, Table S9. The selected mutants were verified by PCR and DNA sequencing. The desired complementary and overexpression strains were screened by apramycin-resistance testing (17) and PCR. EMSAs. The putative promoter regions of ABC transporter loci were amplified by PCR using the primers listed in SI Appendix, Table S9. PCR products were

PNAS Early Edition | 5 of 6


that the expression of many genes encoding alpha-/beta-glucosidases (SI Appendix, Fig. S5) in S. erythraea is directly regulated by GlnR. Glucosidases are an important group of hydrolases that break down various polysaccharides during carbohydrate metabolism. Taken together, these data show that GlnR serves a crucial role as a master regulator of carbon metabolism in actinomycetes. Both carbon and nitrogen are essential components for microbial nutrition. To survive in a competitive environment, or to cope with a challenging nutrient niche, microorganisms have developed sophisticated mechanisms to sense various nutrient signals and adapt their metabolic pathways accordingly. Actinomycetes are widely distributed in various natural habitats, including soil. Inorganic nitrogen and glucose are the preferred nitrogen/ carbon nutrient sources for actinomycetes, which are frequently limited in soil. The regulation of carbohydrate metabolism in actinomycetes is under GlnR control, indicating that nitrogen starvation triggers GlnR-mediated responses, resulting in the induction of carbohydrate-metabolism genes. Crosstalk between nitrogen- and carbon-metabolism pathways may serve as a way to preserve limited cell resources and use alternative nutrients via the expression of different genes for nitrogen and carbon assimilation. Our observations demonstrate that the GlnR protein is an important mediator that balances cell nitrogen–carbon homeostasis by sensing and controlling nitrogen availability, thereby modulating utilization of biomass-derived sugars in soil. It is well established that GlnR integrates intracellular nitrogen signals to modulate (induce or repress) transcription of its target genes in response to nitrogen availability (15, 19). In this study, we found that the transcription of genes involved in the transport and utilization of multiple carbohydrates (such as maltose) were notably induced under a nitrogen-limited condition compared with a nitrogen-rich condition. Maltose can be transported into cells by the ABC transport system, encoded by the SACE_1666-1664 operon, and broken down into two glucose molecules by alpha-glucosidase (AglA), encoded by SACE_1668. The availability of a readily metabolizable carbon source such as glucose represses the synthesis and activities of enzymes necessary for the transport and metabolism of secondary carbon sources such as maltose and xylose (CCR regulation). Our results demonstrated that nitrogen has a regulatory effect on the utilization of secondary carbon sources beyond CCR regulation [via cluster-specific regulation or the cAMP receptor protein (CRP)]. GlnR directly activates the expression of carbohydrate-related genes under nitrogen starvation. The cluster-specific regulators (CSRs) are widely distributed in the operons involved in carbohydrate metabolism (SI Appendix, Fig. S5), i.e., the MalR repressor of the maltose/ maltodextrin transport system in S. coelicolor (22) and Streptomyces lividans (23), and the CebR repressor of the cellulose/ cellooligosaccharide transporter system in S. griseus (24). In the intergenic regions of malEFG and malR-aglA of S. erythraea, two putative MalR-binding sites were identified, one of which overlapped with the GlnR cis-element (SI Appendix, Fig. S2A), and verified to be bound by MalR (SI Appendix, Fig. S13A). CRP was originally identified in Gram-negative E. coli and was found to play a crucial role in CCR (25), although the functions of CRP homologs in actinomycetes are obscure. In a model strain of S. coelicolor, it was demonstrated that CCR regulation was not mediated by CRP (26, 27), but rather by the glycolytic enzyme, glucose kinase (28). However, the CRP homolog in S. erythraea was found to be a cAMP-dependent DNA-binding protein that binds the regulatory region of certain ABC transporters (SI Appendix, Fig. S13B); this regulatory feature was similar to the CRP homolog of E. coli rather than that of S. coelicolor (17). Thus, CRP might be involved in CCR regulation in some actinomycetes. Similar CRP-response elements also overlap with GlnR ciselements in some ABC transporter loci (SI Appendix, Fig. S13C). Consistent with these findings, a likely competition between GlnR

labeled with biotin by using a universal biotinylated primer (5′-biotin-AGCCAGTGGCGATAAG-3′). Biotin-labeled PCR products were purified with the PCR Purification kit (Shanghai Generay Biotech) as EMSA probes. For the analysis of GlnR cis-elements in the mal locus, three subfragments were synthesized and biotinylated directly, and equimolar amounts of the two complementary single-stranded oligonucleotides were denatured at 94 °C for 5 min and allowed to anneal with a temperature reduction of 1 °C per min in a standard PCR machine. EMSAs were performed by using a Chemiluminescent EMSA kit (Beyotime Biotechnology), according to the manufacturer’s protocol. The binding reaction contained 10 mM Tris HCl (pH 8.0), 25 mM MgCl2, 50 mM NaCl, 1 mM DTT, 1 mM EDTA, 0.01% Nonidet P-40, 50 μg/mL poly(d[I-C]), and 10% glycerol. Biotin-labeled DNA probes were incubated individually with varying amounts of proteins of interest at 25 °C for 20 min. Next, samples were resolved on 6% nondenaturing PAGE gels in ice-cold 0.5× Trisborate-EDTA at 100 V, and bands were detected by using the BeyoECL Plus kit (Beyotime). qPCR. All primers used for qPCR are listed in SI Appendix, Table S9. qRT-PCR was conducted by using a SYBR Premix Ex Taq GC kit (Takara), and all procedures were performed according to the manufacturer’s instructions. Total RNA (1 μg) was reverse-transcribed by using a PrimeScript RT Reagent kit with gDNA Eraser (Takara). DNase digestion was performed to remove genomic DNA before reverse transcription for 5 min at 42 °C. cDNA (∼100 ng) was added to a final PCR volume of 20 μL, and PCR assays were performed by using a CFX96 Real-Time System (Bio-Rad). The following thermocycling conditions were used: 95 °C for 5 min, followed by 40 cycles of 95 °C for 10 s, 60 °C for 10 s, and 72 °C for 30 s; with a final extension cycle at 72 °C for

1. Zhang X, et al. (2009) Metabolic evolution of energy-conserving pathways for succinate production in Escherichia coli. Proc Natl Acad Sci USA 106(48):20180–20185. 2. Görke B, Stülke J (2008) Carbon catabolite repression in bacteria: Many ways to make the most out of nutrients. Nat Rev Microbiol 6(8):613–624. 3. Tarr PT, Tarling EJ, Bojanic DD, Edwards PA, Baldán A (2009) Emerging new paradigms for ABCG transporters. Biochim Biophys Acta 1791(7):584–593. 4. Rees DC, Johnson E, Lewinson O (2009) ABC transporters: The power to change. Nat Rev Mol Cell Biol 10(3):218–227. 5. Higgins CF (1992) ABC transporters: From microorganisms to man. Annu Rev Cell Biol 8:67–113. 6. Hollenstein K, Dawson RJ, Locher KP (2007) Structure and mechanism of ABC transporter proteins. Curr Opin Struct Biol 17(4):412–418. 7. Eitinger T, Rodionov DA, Grote M, Schneider E (2011) Canonical and ECF-type ATPbinding cassette importers in prokaryotes: Diversity in modular organization and cellular functions. FEMS Microbiol Rev 35(1):3–67. 8. Davidson AL, Dassa E, Orelle C, Chen J (2008) Structure, function, and evolution of bacterial ATP-binding cassette systems. Microbiol Mol Biol Rev 72(2):317–364. 9. Nikaido H (1994) Maltose transport system of Escherichia coli: An ABC-type transporter. FEBS Lett 346(1):55–58. 10. Ehrmann M, Ehrle R, Hofmann E, Boos W, Schlösser A (1998) The ABC maltose transporter. Mol Microbiol 29(3):685–694. 11. Bertram R, et al. (2004) In silico and transcriptional analysis of carbohydrate uptake systems of Streptomyces coelicolor A3(2). J Bacteriol 186(5):1362–1373. 12. Titgemeyer F, et al. (2007) A genomic view of sugar transport in Mycobacterium smegmatis and Mycobacterium tuberculosis. J Bacteriol 189(16):5903–5915. 13. Mahajan GB, Balachandran L (2012) Antibacterial agents from actinomycetes - a review. Front Biosci (Elite Ed) 4:240–253. 14. Lynd LR, Weimer PJ, van Zyl WH, Pretorius IS (2002) Microbial cellulose utilization: Fundamentals and biotechnology. Microbiol Mol Biol Rev 66(3):506–577. 15. Yao LL, et al. (2014) GlnR-mediated regulation of nitrogen metabolism in the actinomycete Saccharopolyspora erythraea. Appl Microbiol Biotechnol 98(18):7935–7948. 16. van Wezel GP, White J, Bibb MJ, Postma PW (1997) The malEFG gene cluster of Streptomyces coelicolor A3(2): Characterization, disruption and transcriptional analysis. Mol Gen Genet 254(5):604–608. 17. Liao CH, Yao LL, Ye BC (2014) Three genes encoding citrate synthases in Saccharopolyspora erythraea are regulated by the global nutrient-sensing regulators GlnR, DasR, and CRP. Mol Microbiol 94(5):1065–1084. 18. Amon J, Titgemeyer F, Burkovski A (2010) Common patterns - unique features: Nitrogen metabolism and regulation in Gram-positive bacteria. FEMS Microbiol Rev 34(4):588–605. 19. Tiffert Y, et al. (2008) The Streptomyces coelicolor GlnR regulon: Identification of new GlnR targets and evidence for a central role of GlnR in nitrogen metabolism in actinomycetes. Mol Microbiol 67(4):861–880.


10 min. The methods used for cell cultivation and RNA preparation are described in detail in the SI Appendix, SI Materials and Methods. Sugar Uptake Assays. Sugar-uptake assays were performed as described (31), with modifications. Briefly, 50 mL of each strain was grown to logarithmic phase in TSB medium supplemented with 1% glucose or maltose as the sole carbon source. Next, strains were harvested and washed twice with ice-cold carbon-depleted Evans medium. Equal weights of cells were then resuspended in 50 mL of Evans medium containing 5 mM glucose or maltose and incubated at 30 °C with orbital shaking (200 rpm). Samples of cell suspensions were collected every 2 h and centrifuged at 10,000 × g for 20 min at 4 °C, after which the sugar concentrations remaining in the supernatants were determined as a measure of sugar uptake. Glucose or maltose concentrations were determined by using the Glucose Assay kit (Applygen Technologies) and the Maltose Assay kit (BioVision), respectively, according to the manufacturer’s instructions. Computational Analysis. MEME/MAST tools (meme-suite.org/) and PREDetector software (32) were used to search for GlnR cis-elements in the 5′ regions of the ABC transporter operons. A training set of sequences was used to generate weight matrices, and the prediction methods were performed as described (15, 33). ACKNOWLEDGMENTS. We thank Sebastien Rigali (University of Liège) for his kind assistance with this work and Wolfgang Wohlleben (University of Tuebingen) for providing the S. coelicolor ΔglnR mutant strain. This work was supported by China Natural National Science Foundation Grants 21276079, 21421004, and 21335003; Chinese Ministry of Education Grant SRFDP 20120074110009; National Key Technologies R&D Programs 2014AA021502; and by Fundamental Research Funds for the Central Universities.

20. Pullan ST, Chandra G, Bibb MJ, Merrick M (2011) Genome-wide analysis of the role of GlnR in Streptomyces venezuelae provides new insights into global nitrogen regulation in actinomycetes. BMC Genomics 12:175. 21. Jeßberger N, et al. (2013) Nitrogen starvation-induced transcriptome alterations and influence of transcription regulator mutants in Mycobacterium smegmatis. BMC Res Notes 6:482. 22. van Wezel GP, White J, Young P, Postma PW, Bibb MJ (1997) Substrate induction and glucose repression of maltose utilization by Streptomyces coelicolor A3(2) is controlled by malR, a member of the lacl-galR family of regulatory genes. Mol Microbiol 23(3):537–549. 23. Schlösser A, Weber A, Schrempf H (2001) Synthesis of the Streptomyces lividans maltodextrin ABC transporter depends on the presence of the regulator MalR. FEMS Microbiol Lett 196(1):77–83. 24. Marushima K, Ohnishi Y, Horinouchi S (2009) CebR as a master regulator for cellulose/ cellooligosaccharide catabolism affects morphological development in Streptomyces griseus. J Bacteriol 191(19):5930–5940. 25. Ishizuka H, Hanamura A, Inada T, Aiba H (1994) Mechanism of the down-regulation of cAMP receptor protein by glucose in Escherichia coli: Role of autoregulation of the crp gene. EMBO J 13(13):3077–3082. 26. Derouaux A, et al. (2004) Crp of Streptomyces coelicolor is the third transcription factor of the large CRP-FNR superfamily able to bind cAMP. Biochem Biophys Res Commun 325(3):983–990. 27. Derouaux A, et al. (2004) Deletion of a cyclic AMP receptor protein homologue diminishes germination and affects morphological development of Streptomyces coelicolor. J Bacteriol 186(6):1893–1897. 28. Gubbens J, Janus M, Florea BI, Overkleeft HS, van Wezel GP (2012) Identification of glucose kinase-dependent and -independent pathways for carbon control of primary metabolism, development and antibiotic production in Streptomyces coelicolor by quantitative proteomics. Mol Microbiol 86(6):1490–1507. 29. Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA (2000) Practical Streptomyces genetics. Int Microbiol 3(4):260–261. 30. Fink D, Weissschuh N, Reuther J, Wohlleben W, Engels A (2002) Two transcriptional regulators GlnR and GlnRII are involved in regulation of nitrogen metabolism in Streptomyces coelicolor A3(2). Mol Microbiol 46(2):331–347. 31. Choi KH, Hwang S, Cha J (2013) Identification and characterization of MalA in the maltose/maltodextrin operon of Sulfolobus acidocaldarius DSM639. J Bacteriol 195(8): 1789–1799. 32. Hiard S, et al. (2007) PREDetector: A new tool to identify regulatory elements in bacterial genomes. Biochem Biophys Res Commun 357(4):861–864. 33. Liao C, et al. (2014) Control of chitin and N-acetylglucosamine utilization in Saccharopolyspora erythraea. Microbiology 160(Pt 9):1914–1928.

Liao et al.