110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins but major differences for saccharide transporters

JPROT-02321; No of Pages 9 Journal of Proteomics xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Proteomics journal homepage: www.elsevier.com/locate/jprot

Comparative proteome analysis of Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins but major differences for saccharide transporters Sergej Wendler a, Andreas Otto b, Vera Ortseifen a,c, Florian Bonn b, Armin Neshat c, Susanne Schneiker-Bekel a, Timo Wolf a,c, Till Zemke d, Udo F. Wehmeier e, Michael Hecker b, Jörn Kalinowski c, Dörte Becher b, Alfred Pühler a a

Senior Research Group in Genome Research of Industrial Microorganisms, Center for Biotechnology, Bielefeld University, Universitätsstraße 27, 33615 Bielefeld, Germany Institute for Microbiology, Ernst-Moritz-Arndt-University of Greifswald, F.-L. Jahnstrasse 15, 17489 Greifswald, Germany c Microbial Genomics and Biotechnology, Center for Biotechnology, Bielefeld University, Universitätsstraße 27, 33615 Bielefeld, Germany d Product Supply, Bayer Pharma AG, Friedrich Ebert Str. 217-475, 42117 Wuppertal, Germany e Department for Sportsmedicine, University of Wuppertal, Pauluskirchstr. 7, 42285 Wuppertal, Germany b

a r t i c l e

i n f o

Article history: Received 30 July 2015 Received in revised form 13 October 2015 Accepted 20 October 2015 Available online xxxx Keywords: Actinoplanes Comprehensive proteomics Metabolic labeling Subcellular fractionation Acarbose

a b s t r a c t Actinoplanes sp. SE50/110 is known for the production of the α-glucosidase inhibitor and anti-diabetic drug acarbose. Acarbose (acarviosyl-maltose) is produced as the major product when the bacterium is grown in medium with maltose, while acarviosyl-glucose is the major product when glucose is the sole carbon source in the medium. In this study, a state-of-the-art proteomics approach was applied combining subcellular fractionation, in vivo metabolic labeling and shotgun mass spectrometry to analyze differences in the proteome of Actinoplanes sp. SE50/110 cultures grown in minimal medium containing either maltose or glucose as the sole carbon source. To study proteins in distinct subcellular locations, a cytosolic, an enriched membrane, a membrane shaving and an extracellular fraction were included in the analysis. Altogether, quantitative proteome data was obtained for 2497 proteins representing about 30% of the ca. 8270 predicted proteins of Actinoplanes sp. SE50/110. When comparing protein quantities of maltose- to glucose-grown cultures, differences were observed for saccharide transport and metabolism proteins, whereas differences for acarbose biosynthesis gene cluster proteins were almost absent. The maltose-inducible α-glucosidase/maltase MalL as well as the ABC-type saccharide transporters AglEFG, MalEFG and MstEAF had significantly higher quantities in the maltose growth condition. The only highly abundant saccharide transporter in the glucose condition was the monosaccharide transporter MstEAF, which may indicate that MstEAF is the major glucose importer. Taken all findings together, the previously observed formation of acarviosyl-maltose and acarviosyl-glucose is more closely connected to the transport of saccharides than to a differential expression of the acarbose gene cluster. Biological significance: Diabetes is a global pandemic accounting for about 11% of the worldwide healthcare expenditures (N 600 billion US dollars) and is projected to affect 592 million people by 2035 (www.idf.org). Whether Actinoplanes sp. SE50/110 produces type 2 diabetes drug acarbose (acarviosyl-maltose) or another acarviose metabolite such as acarviosyl-glucose as the major product depends on the offered carbon source. The differences observed in this proteome in this study suggest that the differences in the formation of acarviosyl-maltose and acarviosyl-glucose are more closely connected to the transport of saccharides than to a differential expression of the acarbose gene cluster. In addition, the present study provides a comprehensive overview of the proteome of Actinoplanes sp. SE50/110. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The actinobacterium Actinoplanes sp. SE50/110 is known for the production of the secondary metabolite acarbose [1,2]. Acarbose inhibits α-glucosidases and is therefore used in the treatment of type 2 diabetes mellitus [3]. The drug belongs to the so-called acarviose metabolites that consist of an acarviose core unit and a variable carbohydrate/ saccharide unit, which in the case of acarbose (acarviosyl-maltose) is

maltose [1,2,4,5]. Different acarviose metabolites can be obtained from cultures grown in media with different carbon sources [1,6,7]. For the natural producer Actinoplanes sp. SE50/110, acarviose metabolites were suggested to provide an advantage by ensuring an exclusive access to carbohydrates that are present in the environment, which is achieved by inhibiting carbohydrate metabolizing enzymes of competitors' [1,8] and by acting as carriers of carbohydrates, so-called carbophores [2,9].

http://dx.doi.org/10.1016/j.jprot.2015.10.023 1874-3919/© 2015 Elsevier B.V. All rights reserved.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

2

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

A more recent progress in the acarbose research was the establishment and improvement of the genome sequence of Actinoplanes sp. SE50/110 [10,11]. The genome has a size of 9.24 Mbp, contains ca. 8270 predicted protein coding sequences and harbors five large secondary metabolite gene clusters [10]. Studying the transcriptome with RNAsequencing revealed that the acarbose biosynthesis gene cluster (acb) was transcribed in maltose-containing minimal medium [12]. As a next step, gel-based methods for the analysis of the cytosolic and extracellular proteome were established to study Actinoplanes sp. SE50/110 cultures grown in minimal medium with maltose [13]. Here, 9 of 22 acb gene cluster proteins and a total of about 180 proteins were identified. Recently, state-of-the-art proteomics that combined subcellular fractionation and LC-MS/MS were applied to achieve a comprehensive analysis of the proteome of Actinoplanes sp. SE50/110 including membrane protein fractions [14]. Here, all 22 acb gene cluster proteins and a total of about 1600 proteins were identified, which demonstrates a remarkable progress compared to the previous gel-based analysis. Moreover, since the biosynthetic acb proteins were found in substantial abundance not only in the cytosolic but also in the membrane fractions, it was concluded that acarbose is synthesized at the membrane [14]. The relevance of the carbon source for the production of acarviose metabolites is known since a couple of decades [1,6]. However, systematic analyses of the carbon source dependence using fully synthetic media and highly sensitive MS-detection were published only recently [5]. Interestingly, Actinoplanes sp. SE50/110 was found to produce acarviosyl-maltose as the major product when grown in medium with maltose, while producing mainly acarviosyl-glucose when grown in medium with glucose [5]. Eventhough the underlying biosynthesis pathway has been proposed more than a decade ago [2], there is little knowledge about the last biosynthesis steps, in which the variable saccharide units are attached. Our hypothesis explaining the different products is that there could either be a differential expression of (1) specific genes of the acb gene cluster and/or (2) genes outside of the acb cluster such as saccharide transporters that could provide variable saccharide units for different acarviose metabolites. Conclusively, the aim of this work is to explain the formation of the different products by elucidating differences in the proteome of Actinoplanes sp. SE50/110 cultures grown in minimal medium containing either maltose or glucose as sole carbon source. For this, a state-of-the-art proteomics approach [15,16] was applied that combined subcellular fractionation with in vivo metabolic labeling [17] and shotgun mass spectrometry. 2. Material and methods 2.1. Cultivation of Actinoplanes sp. SE50/110 in minimal medium containing maltose or glucose Actinoplanes sp. SE50/110 (ATCC 31044; CBS 674.73) cultures were grown in triplicates per condition as described by Wendler et al. [14]. This included three stages of cultivation with two shake flask precultures and one bioreactor main culture. In the second pre-culture, cells were adapted to the minimal main culture medium containing 2.4 C-mol of either maltose (200 mM) or glucose (400 mM). The main cultures were carried out in 1 l Biostat Qplus bioreactors (Sartorius AG, Göttingen, Germany) with 800 ml medium, an automatically controlled pH of 6.5, temperature set to 30 °C and a dissolved oxygen level set to 50%. The growth was monitored through determination of the cell dry weight and the acarbose production by HPLC as described previously [13]. 2.2. Quantification strategy for proteomic samples For the quantification of proteins, a strategy relying on metabolic labeling to obtain an internal standard and 14N/15N relative quantification according to MacCoss et al. [17] was applied [15,16]. The 15N standard

was composed of a mixed pool of proteins from maltose- as well as glucose-grown cultures. The unlabeled 14 N samples were mixed with equal amounts of the labeled 15N standard, which ensured that the steps of the proteome sample preparation affected the sample and standard proteins equally. 2.3. Preparation of proteomic samples For the relative quantification, cells were harvested after 46 h of cultivation. The subcellular fractionation and preparation of the different subcellular fractions was performed as described by Wendler et al. [14] and Otto et al. [15]. A summarizing workflow illustrating the crucial steps in the preparation of the proteome samples is provided in Fig. 1. Applying the subcellular fractionation approach, four different subproteome fractions were obtained: an extracellular fraction (ExF), a cytosolic fraction (CyF), an enriched membrane fraction (EMF) and a membrane shaving fraction (MSF). The ExF was obtained through a separation of cells and supernatant by centrifugation followed by a phenol extraction of proteins from the latter. The CyF was obtained through disruption of cells and phenol extractions of proteins. The bacterial membranes used for the EMF and MSF were obtained through ultracentrifugation of disrupted cells. For the relative quantification, equal protein amounts of unlabeled 14N samples and the labeled 15N standard were mixed at an early stage. Proteins of the CyF, EMF and ExF were fractionated using 1D SDS-gels and subsequently in-gel digested with trypsin. In the case of the membrane shaving fraction, cellular membranes were “shaved” by digesting soluble loops of proteins integral to the membrane with Proteinase K. The remaining membranes containing peptides of the transmembrane domains were subsequently subjected to a buffer containing the MS-compatible detergent RapiGest (Waters Corporation, Milford, MA, USA) and digested with chymotrypsin. Proteome samples of all fractions were measured with high-resolution and high mass accuracy LC-MS/MS. 2.4. Mass spectrometry measurements Sample preparation was carried out as described by Otto et al. [15] and LC-MS/MS measurements were performed as described by Bonn et al. [18]. Peptide mixtures resulting from the proteome sample preparation were subjected to reversed phase C18 column chromatography operated on a Proxeon EASYnLC (Thermo Fisher Scientific, Waltham, MA, USA). MS and MS/MS data were acquired with a LTQ-Orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) online coupled to the LC system. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral. proteomexchange.org) via the PRIDE partner repository [19] with the dataset identifier PXD001497. 2.5. Data analysis The data analysis was performed as described previously [15,16]. For this, the *.dta files were extracted from *.raw files and searched in two iterations with Sorcerer SEQUEST (SageN) against a target-decoy protein sequence database composed of all protein sequences of the Actinoplanes sp. SE50/110 genome [10]. With the two iterations, masses of amino acids with either 14N- or 15N-nitrogen were considered. For all searches following parameters were applied: peptide tolerance, 10 ppm; tolerance for fragment ions, 1 amu; b- and y-ion series; variable modification, oxidation of methionine (15.99 Da) with up to three modifications per peptide. A carboxyamidomethylation of cysteine (57.02 Da) was also allowed as a variable modification for the membrane shaving fraction. The digestion type was set to No Enzyme for samples of the membrane shaving fraction and to Full digest using Trypsin allowing two missed cleavage sites for the other three fractions. For the relative quantification, respective mass traces of the two cognate isotopologue peptides (mass range of

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

3

Fig. 1. Workflow illustrating the most important steps in the preparation of the proteome fractions. Actinoplanes sp. SE50/110 cultures were grown in minimal medium containing either maltose or glucose in bioreactors in three replicates. Cultures were separated by centrifugation in cells and supernatant (extracellular fraction; ExF). Thereafter cells were disrupted to obtain a cytosolic fraction (CyF). Membranes were obtained from the disrupted cell solution through ultracentrifugation to prepare an enriched membrane fraction (EMF) through depletion of cytosolic proteins (washing) and a membrane shaving fraction (MSF) through treatment with Proteinase K. In the 14N/15N mixing, labeled and unlabeled protein pools were mixed in equal amounts to obtain quantitative data. Prior to the LC separation, proteins of the CyF, EMF and ExF were fractionated using 1D SDS-gels. The proteomic analysis (LC-MS/MS) was performed with high-resolution and high mass accuracy mass spectrometry.

10 ppm) were extracted with the Census software [20]. Here, ratios of the peak intensities of the lighter 14N sample mass trace and the heavier 15N standard heavy mass trace were calculated for all identified peptides, combined to protein ratios and exported (R2 value N 0.7, only unique peptides). Proteins that could not be quantified automatically were subsequently manually curated, which mostly concerned proteins that were exclusively present in the 14N or the 15N mass trace. Only proteins for which quantitative data was obtained for at least two unique peptides were considered as quantified. All protein ratios of the same sample were median-centered and log2 transformed. From these values mean log2 values of three replicates were calculated. To obtain the values that are indicated in the results section as the Δlog2 values, the mean log2 values of the glucose-grown culture samples were subtracted from mean log2 values of maltose-grown culture samples, which also eliminated the 15N internal standard. The resulting Δlog2 values indicate different relative quantities: positive, higher relative quantity in the maltosecondition; negative, higher relative quantity in the glucose-condition. The indicated fold changes are the exponentials of the Δlog2 values. To determine whether the protein quantities were significantly different, a

standard pair-wise Student's t-test (with a p-value below 0.01 and 0.05) was performed with the TIGR Multiexperiment Viewer (MeV) v. 4.9.0. (www.tm4.org/mev.html). 2.6. Relative quantification of proteins by spectral counting Normalized spectral abundance factors [21] (NSAF) were calculated as mean values of three replicates to estimate the relative abundance of proteins and to rank proteins according to their abundance within a proteomic fraction. NSAF values (SpC/(L·∑ SpC)) are the number of MS/MS spectra (SpC) assigned to a protein divided by protein length (L) and the total number of spectra assigned to proteins (∑SpC) within one sample. For one of the six MSF sample replicates NSAF values could not be calculated due to a high proportion of 15N standard proteins. 2.7. Prediction of subcellular locations of As previously described for Actinoplanes sp. SE50/110 [14], the Locate P v.2.0 pipeline [22] was applied to predict the subcellular location

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

4

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

of proteins and categorized as follows: (1) intracellular {intracellular}; (2) integral membrane {multitransmembrane, multitransmembrane (lipid-modified N termini)}; (3) membrane-associated {lipid anchored, N-terminally anchored (no cleavage site), N-terminally anchored (with cleavage site), C-terminally anchored (with cleavage site), intracellular/ TMH start after 60}; (4) secreted {secretory (released) (with cleavage site)}. 3. Results and discussion 3.1. 2563 different proteins were identified in the four subcellular proteome fractions We have previously demonstrated that Actinoplanes sp. SE50/110 produces acarviosyl-maltose (acarbose) when grown in medium with maltose and acarviosyl-glucose when grown in medium with glucose [5]. Conclusively, the aim of this work was to analyze differences in the proteome of Actinoplanes sp. SE50/110 cultures grown in minimal medium containing either maltose or glucose as sole carbon sources. The results should explain the formation of the two occurring acarviose metabolites and test the hypothesis if there is either a differential expression of (1) specific genes of the acb gene cluster and/or (2) genes outside of the acb cluster such as saccharide transporters. For this, cultures were grown under controlled conditions in bioreactors in three biological replicates. To reveal differences in protein quantities in distinct locations of the cell, a comprehensive proteome analysis combining subcellular fractionation and in vivo 14N/15N metabolic labeling was carried out. Samples were taken from the bioreactors during exponential growth after 46 h of cultivation. Subsequent fractionation allowed accessing soluble cytosolic proteins particularly with the cytosolic fraction (CyF), membrane-associated proteins particularly with the enriched membrane fraction (EMF), integral membrane proteins particularly with the membrane shaving fraction (MSF), and extracellular proteins particularly with the extracellular fraction (ExF). The most important steps in the preparation of the proteome fractions are illustrated in Fig. 1. The analysis of subcellular fractions of maltose- and glucose-grown Actinoplanes sp. SE50/110 cultures resulted in an identification of 1537 proteins in the CyF (Table S1), 1884 proteins in the EMF (Table S2), 372 proteins in the MSF (Table S3), 577 proteins in the ExF (Table S4) and a combined total of 2563 different proteins (Table S5 & Fig. 2). 14 N/15N quantitative data could be obtained for about 97% or 2497 of

the identified proteins, which represent about 30% of the approximate 8270 theoretical proteins of Actinoplanes sp. SE50/110 (summarizing list in Table S5). In Fig. 2, the composition according to the predicted subcellular locations (SCL) of the proteins is indicated for the identified proteins and their corresponding normalized spectral abundance factors (NSAF). While the composition of the CyF and MSF was similar according to the predicted SCLs for maltose- and glucose-grown cultures, differences between the conditions were observed in the EMF. Here, predicted membrane-associated proteins had a higher share in the NSAFs of maltose-grown cells. Moreover, the composition of the ExF also differed considerably when comparing the identified proteins or the NSAFs. In the ExF, the group of predicted intracellular proteins was reproducibly in all three replicates the major group in glucosegrown cultures. This indicates cell leakage, which might have been caused by a higher osmotic stress due to monosaccharide feeding with glucose in comparison to the disaccharide feeding with maltose as both had been applied in C-equimolar amounts. The unexpectedly high share of integral membrane proteins observed in maltose-grown cultures is mainly due to a miss-prediction of the protein Cgt as an integral membrane protein instead of a truly secreted protein [14].

3.2. Significantly different quantities were determined for 396 proteins After obtaining a large set of quantitative proteome data, the next objective was to extract biologically meaningful information. In a first screening, we determined proteins that had significantly different amounts under the two conditions by applying a standard pair-wise Student's t-test with a p-value b 0.01 (Tables S1–S5). Altogether, 396 proteins with significantly different quantities were determined, of which 134 belonged to the CyF, 184 to the EMF, 19 to the MSF and 59 to the ExF. In order to obtain biologically more meaningful and relevant differences, we used a fold change of 1.5 and a NSAF N0.02% as more stringent thresholds. By this, we could determine 81 proteins in the CyF (Table S6), 113 proteins in the EMF (Table S7), 18 proteins in the MSF (Table S8) and 49 proteins in the ExF (Table S9) with substantially different quantities. The differences in the four fractions of maltose- and glucose-grown cultures are presented with respect to the predicted subcellular locations in Fig. 3. Starting with the cytosolic fraction, most proteins with substantially different quantities, namely 72 of the 81, were predicted intracellular proteins, which was also in agreement with expectations.

Fig. 2. Proteins identified in the four subcellular proteome fractions of Actinoplanes sp. SE50/110 grown in minimal medium containing either maltose (Mal) or glucose (Glc). The figure shows the number of identified proteins as well as the normalized spectral abundance factors [1] (NSAFs) according to their predicted subcellular locations (SCLs) of the cytosolic (CyF), enriched membrane (EMF), membrane shaving (MSF) and extracellular fraction (ExF). The subcellular locations of gene products were predicted with LocateP [2] and are differentiated into an intracellular (In), a membrane-associated (MA), an integral membrane (IM) and a secreted (Se) category. Colors of the different proteome fractions match subcellular location colors to which they are expected to correspond.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

5

Fig. 3. Changes in subcellular proteome fractions of Actinoplanes sp. SE50/110 grown in minimal medium containing either maltose (Mal) or glucose (Glc). The figure shows the numbers of proteins determined to have substantially different quantities. Proteins are categorized by predicted subcellular locations (SCLs) of the cytosolic (CyF), enriched membrane (EMF), membrane shaving (MSF) and extracellular fraction (ExF). The filter criteria were set to a p-value b0.01, a normalized spectral abundance factor N 0.02% and a fold change cut-off N1.5. The subcellular locations of gene products were predicted with LocateP [2] and are differentiated into an intracellular (In), a membrane-associated (MA), an integral membrane (IM) and a secreted (Se) category. Colors of the different proteome fractions match subcellular location colors to which they are expected to correspond.

In the enriched membrane fraction on the other hand, 58 and therefore the vast majority of the 61 proteins that were found in higher quantities in cultures grown on glucose belonged to the predicted intracellular proteins and had no obvious relation to the membrane. In the counterpart, the proteins with higher quantities in maltose-grown cultures, 47 of 52 were related to the membrane (27 predicted membraneassociated and 20 predicted integral membrane proteins). In the membrane shaving fraction, all 18 proteins with substantially different quantities were either predicted membrane-associated (2) or integral membrane (16) proteins. In this fraction only 2 proteins had a higher quantity in the presence of glucose while 16 had a higher quantity in the presence of maltose. In the extracellular fraction, 34 proteins were found in higher quantities in cultures grown on glucose. Of these, only 1 was predicted to be a secreted protein while most were predicted intracellular proteins. Of the 15 proteins with a higher quantity on maltose on the other hand, 10 were predicted as secreted proteins. Taken together, it was evident that the composition of proteins having a higher relative quantity in glucose- and maltose-grown cultures was different. In the EMF, MSF and ExF for example very few proteins with a higher quantity in glucose-grown cultures were inherent to these fractions. The opposite was the case for maltose-grown cultures since the predicted locations for these proteins, which are based on sequence motifs, was in accordance with the fraction where these proteins were detected. The above mentioned cell leakage of glucosegrown cells could explain the high number of predicted intracellular proteins found with higher amounts in the EMF and ExF. In case of the EMF, proteins which had been previously released into the extracellular space could for example have been adsorbed and/or (co-) precipitated at the membrane [23–27]. In the next step, the initial research question was addressed and proteins encoded in the acb gene cluster as well as saccharide transporters were analyzed in more detail.

been mentioned is the statistical analysis and the filter criteria (p b 0.01, NSAF N 0.02%). The other point is that the subcellular locations of proteins and therefore an analysis of the correct proteome fractions are important. For the acarbose biosynthesis gene cluster, the suggested membrane biosynthesis implicates that for the intracellular and/or biosynthetic proteins the enriched membrane fraction needs to be included in the analysis in addition to the cytosolic fraction. For the other proteins only the major locations (fractions) of the proteins were considered. The results of this analysis are shown in Fig. 4. Of the 22 proteins, quantitative differences were only determined for AcbK and AcbN. The acarbose-7-kinase AcbK was found to be more abundant in the CyF of maltose-grown cultures. AcbK was postulated to protect the host against the inhibition by acarviose metabolites [2,28,29]. This might be relevant because acarviosyl-maltose had been shown to be a stronger α-glucosidase inhibitor than acarviosyl-glucose [1]. The cyclitol oxidoreductase AcbN was found in higher quantity in cultures that were grown with glucose in the EMF. The enzyme was proposed to catalyze the fifth of the seven steps of the cyclitol subunit biosynthesis [2]. It is not obvious at this point, why only one enzyme which catalyzes an intermediate biosynthesis step had a differing quantity. In contrast to the quantitative proteome data for AcbK, a tendency of a higher expression level of AcbN in the presence of glucose was also not supported by the microarray that has been carried out under the same conditions (unpublished data). Taken together, differences were only observed for a few single Acb proteins, but not for the whole acb cluster and do not provide an explanation why different acarviose metabolites were formed when cultures were grown either with maltose or glucose. The study at hand has been carried out with the Actinoplanes sp. SE50/110 wild type strain. It would certainly be interesting to see whether a comparative proteome analysis with industrial Actinoplanes strains that are able to produce much higher amounts of acarbose would come to the same results.

3.3. No differences were observed for Acb proteins that could provide an explanation for the formation of different acarviose metabolites

3.4. Quantitative data obtained for saccharide transporters could provide an explanation how different acarviose metabolites are synthesized

In a quantitative analysis of gene cluster proteins some aspects have to be considered to obtain meaningful results. The part that has already

In contrast to the acarbose biosynthesis gene cluster, some proteins of the carbohydrate transport and metabolism were found to be

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

6

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

Fig. 4. Expression of the acarbose biosynthesis gene cluster of Actinoplanes sp. SE50/110 grown in minimal medium containing either maltose or glucose. Positive Δlog2 values indicate a higher protein quantity in maltose-grown cultures, while negative-values indicate a higher protein quantity in glucose-grown cultures. Asterisks indicate a p-value below 0.01. The subcellular locations of gene products were predicted with LocateP [2]. Colors of the different proteome fractions match subcellular location colors to which they are expected to correspond.

differentially expressed. The proteins concerned were in particular the maltose-inducible α-glucosidase/maltase MalL and the three ABC-type saccharide transporters PulA/MalEFG/MalR, AglEFG and MstEAF/MstR.

Differences in relative quantities as well as the abundance ranks for proteins of the corresponding gene clusters and the four proteome fractions are summarized in Fig. 5.

Fig. 5. Expression of carbohydrate metabolism and transport genes of Actinoplanes sp. SE50/110 grown in minimal medium containing either maltose or glucose. Positive Δlog2 values indicate a higher protein quantity in maltose-grown cultures, while negative-values indicate a higher protein quantity in glucose-grown cultures. Asterisks indicate a p-value below 0.01. The abundance ranks were deduced from normalized spectral abundance factors in each fraction. The subcellular locations of gene products were predicted with LocateP [2]. Colors of the different proteome fractions match subcellular location colors to which they are expected to correspond.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

3.4.1. The maltose-inducible α-glucosidase/maltase MalL has a higher quantity on maltose and is proposed to cleave maltose to glucose The maltose-inducible α-glucosidase/maltase MalL is encoded in the genome as a single gene. It was found in the cytosolic fraction as one of the proteins with the highest fold change (15.9) (Fig. 5). The protein most closely related and experimentally characterized is the maltoseinducible alpha-glucosidase (sucrase-isomaltase-maltase) MalL of Bacillus subtilis (blastp [45] aa identity 31%, E 8e-78). In B. subtilis, MalL was described to be induced by maltose, to be repressed by glucose and to catalyze the cleavage of maltose to glucose [30,31]. Regulatory proteins of the MalL expression in B. subtilis have not yet been determined. In the genetic proximity of the Actinoplanes sp. SE50/110 malL locus four genes annotated as putative transcriptional regulators are present (acpl_2532, acpl_2548, acpl_2552 & acpl_2554). However, none of these proteins were identified in the proteome, nor showed noticeable transcription profiles in the previous RNA-sequencing [12] or microarray experiments (unpublished data). 3.4.2. Proteins of the multiple α-glucoside transporter AglEFG are present in higher quantities on maltose The genes of the ABC-type multiple α-glucoside transporter aglEFG are organized in a cluster of three consecutive genes. According to the annotation of the encoded proteins, AglE is a solute-binding protein while AglF and AglG are permease components. AglE, AglF and AglG were all present in higher quantities in maltose- compared to glucose-grown cultures in the corresponding fractions with fold changes of 10.8, 3.7 and 3.4, respectively (Fig. 5). Agl-Proteins have previously been reported as dominant proteins of the extracellular proteome of Actinoplanes sp. SE50/110 [13,14]. The first representative of the multiple α-glucoside ABC-type transporter AglEFG was found in Sinorhizobium meliloti (blastp [45] aa identity 37%, E 2e-88 for AglE; 45%, E 1e-70 for AglF; 50%, E 9e-70 for AglG) [32]. In addition to the aglEFG genes, the identified S. meliloti cluster also contained the presumed regulatory gene aglR, which as such is not present in Actinoplanes sp. SE50/110. The S. meliloti AglEFG system was reported to be induced by sucrose and to transport maltose, trehalose, sucrose and trisaccharides [33]. Prior to the sequencing of the genome of Actinoplanes sp. SE50/110 in 2012 [10], Brunkhorst & Schneider [34] investigated the transport of saccharides and concluded that Actinoplanes sp. possesses an acarbose-insensitive maltose/sucrose/trehalose transporter similar to AglEFG of S. meliloti [33]. 3.4.3. Proteins of the maltose/maltodextrin transporter MalEFG are present in higher quantities on maltose The ABC-type maltose/maltodextrin transport gene cluster consists of the five genes pulA (pullulanase & α-amylase), malG (permease), malF (permease), malE (solute-binding protein) and malR (transcriptional regulator), which are organized in at least two transcription units (Fig. 5). In this work, all five corresponding proteins were identified. The three proteins of the ABC-type transporter MalE, MalF and MalG were present in higher abundance in maltose- compared to glucose-grown cultures in the corresponding fractions with fold changes of 5.2, 2.4 and 2.3, respectively (Fig. 5). For PulA, a higher quantity (fold change of 2.1) was observed in the ExF, but only with a p-value of 0.03. In the case of the transcriptional regulator MalR, a fold change could not be calculated, because it was detected only in one replicate of the maltose-grown cultures. It is common for transcriptional regulators to be present in low abundance [35–38], which imposes a challenge for detection in proteomics [39–41]. Proteins of the mal cluster have previously been detected as abundant proteins of the extracellular proteome of Actinoplanes sp. SE50/110 [13,14]. The MalEFG maltose/maltodextrin ABC-type transporters were first characterized for E. coli (reviewed in [42]) and subsequently also detected in bacteria more closely related to Actinoplanes sp. SE50/110 such as Streptomyces coelicolor A3(2) (blastp [45] aa identity 34%, E 2e-44 for MalE; 29%, E 9e-30 for MalF; 34%, E 2e44 for MalG; 62%, E 2e-149 for MalR) [43,44]. In S. coelicolor A3(2), the transcription of malE was induced by maltose and repressed by glucose,

7

while disruption of malR resulted in constitutive, glucose-insensitive transcription of malE [44]. A rigorous glucose repression seems not to apply to the malEFG system of Actinoplanes sp. SE50/110 since proteins are still among the more abundant proteins in the presence of glucose (Fig. 5). In their investigation of the transport of saccharides, Brunkhorst & Schneider [34] concluded that Actinoplanes sp. possesses also an acarbose-sensitive maltose/maltodextrin transporter similar to MalEFG of E. coli or S. coelicolor A3(2). The maltose/maltodextrin transporter (MalEFG) was also proposed as a possible reimporter for acarbose or thereof derived metabolites [34] and is currently believed to be the most likely candidate for this function [13]. MalEFG is not to be confused with the putative galactose transporter GalEFG (previously referred to as AcbFGH), which has not been detected in this study and is not any more considered to be part of the acarbose gene cluster nor involved in acarviose metabolite uptake [13,51]. 3.4.4. The monosaccharide transporter MstEAF is proposed as the glucose transporter as it is the only highly abundant saccharide transporter on glucose The ABC-type monosaccharide transport (mst) gene cluster consists of the four genes mstE (solute-binding protein), mstA (ATP-binding protein), mstF (permease) and mstR (transcriptional regulator), which are oriented consecutively on one strand (Fig. 5). In this work, the three subunits ABC-type transporter subunits MstE, MstA and MstF were detected. All three ABC-type transporter proteins were present in higher quantity in maltose- compared to glucose-grown cultures. MstE, MstA and MstF had fold changes of 3.8, 1.5 and 1.6, respectively (Fig. 5). However, the higher quantity of MstA was only determined with a p-value of 0.02. The abundance ranks showed that the MstEAF transporter was the only highly-abundant saccharide transporter in glucose-grown cultures, but only one of three in maltose-grown cultures. MalL, MalEFG and AglEFG in contrast, were only highly abundant in the presence of maltose. MstEAF is considered to be a transporter of monosaccharides since this was recognized as the common feature of the most closely related and experimentally characterized ABC-type transporters [14] XylFGH from Escherichia coli K12 (blastp [45] aa identity 39%, E 2e-58 for MstE; 34%, E 3e-44 for MstA; 36%, E 8e-50 for MstF) [46] and ChvE from Agrobacterium tumefaciens (blastp [45] aa identity 36%, E 9e-42) [47,48]. The latter protein was also reported to be induced by the monosaccharides Dgalactose, L-arabinose and D-fucose [49] and described as relevant for chemotaxis towards sugars [50]. To our knowledge an involvement in the transport of di- or oligosaccharides by systems homologous to MstEAF has not been reported so far. 4. Conclusions The present study was carried out to obtain new knowledge on the question why Actinoplanes sp. SE50/110 produces acarviosyl-maltose (acarbose) when grown in medium with maltose and acarviosylglucose when grown in medium with glucose [5]. The results showed that the different products cannot be explained by differences in protein quantities of acarbose gene cluster proteins. However, the quantitative data obtained for saccharide transporters on the other hand provides an explanation how the different acarviose metabolites can be synthesized. The data suggests that maltose is most probably taken up by AglEFG and MalEFG, upon which it can be used for the synthesis of acarviosyl-maltose (acarbose) or can be hydrolyzed by the maltoseinducible α-glucosidase/maltase MalL to glucose. Glucose on the other hand is most probably taken up by MstEAF after which it can be used as a carbon source or for the synthesis of acarviosyl-glucose. Moreover, the supposed maltose transporters AglEFG and MalEFG are not repressed by glucose and the postulated glucose transporter MstEAF is not repressed by maltose. In future, a targeted mutagenesis of gene regions of malL, aglEFG, malEFG and mstEAF should be conducted to test the proposed functions of corresponding proteins.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

8

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jprot.2015.10.023. Transparency document The Transparency document associated with this article can be found in online version. Acknowledgments SW acknowledges financial support from German National Academic Foundation. AN, FW, TW and VO acknowledge financial support from the Graduate Cluster Industrial Biotechnology (CLIB2021) at Bielefeld University, Germany. The Graduate Cluster is supported from Bielefeld University and the Ministry of Innovation, Science and Research (MIWF) of the federal state North Rhine-Westphalia, Germany. AP and JK also acknowledge a grant from the MIWF entitled “Technologieplattform PolyOmics”. We also acknowledge financial support from Bayer AG (Leverkusen, Germany). The data deposition to the ProteomeXchange Consortium was supported by the PRIDE Team. References [1] E. Truscheit, W. Frommer, B. Junge, L. Müller, D.D. Schmidt, W. Wingender, Chemistry and biochemistry of microbial α-glucosidase inhibitors, Angew. Chem. Int. Ed. Engl. 20 (1981) 744–761. [2] U.F. Wehmeier, W. Piepersberg, Biotechnology and molecular biology of the alphaglucosidase inhibitor acarbose, Appl. Microbiol. Biotechnol. 63 (2004) 613–625. [3] H. Bischoff, H.J. Ahr, D. Schmidt, J. Stoltefuß, Acarbose - ein neues Wirkprinzip in der Diabetestherapie, Nachr. Chem. Tech. Lab. 42 (1994) 1119–1128. [4] F. Heiker, H. Böshagen, B. Junge, L. Müller, J. Stoltefuß, Studies designed to localize the essential structural unit of glycoside-hydrolase inhibitors of the acarbose type, in: W. Creutzfeld (Ed.), First Int. Symp. Acarbose, Amsterdam: Excerpta Medica 1981, pp. 137–141. [5] S. Wendler, V. Ortseifen, M. Persicke, A. Klein, A. Neshat, K. Niehaus, et al., Carbon source dependent biosynthesis of acarviose metabolites in actinoplanes sp. SE50/ 110, J. Biotechnol. 191 (2014) 113–120. [6] D.D. Schmidt, W. Frommer, B. Junge, L. Müller, W. Wingender, E. Truscheit, et al., alpha-Glucosidase inhibitors. New complex oligosaccharides of microbial origin, Naturwissenschaften 64 (1977) 535–536. [7] W. Frommer, B. Junge, L. Müller, D. Schmidt, E. Truscheit, New enzyme inhibitors from microorganisms (author's transl), Planta Med. 35 (1979) 195–217, http://dx. doi.org/10.1055/s-0028-1097207. [8] C. Brunkhorst, C. Andersen, E. Schneider, Acarbose, a pseudooligosaccharide, is transported but not metabolized by the maltose-maltodextrin system of Escherichia coli, J. Bacteriol. 181 (1999) 2612–2619. [9] U.F. Wehmeier, The biosynthesis and metabolism of acarbose in actinoplanes sp. SE 50/110: a progress report, Biocatal. Biotransform. 21 (2003) 279–284. [10] P. Schwientek, R. Szczepanowski, C. Rückert, J. Kalinowski, A. Klein, K. Selber, et al., The complete genome sequence of the acarbose producer actinoplanes sp. SE50/110, BMC Genomics 13 (2012) 112. [11] P. Schwientek, A. Neshat, J. Kalinowski, A. Klein, C. Rückert, S. Schneiker-Bekel, et al., Improving the genome annotation of the acarbose producer actinoplanes sp. SE50/ 110 by sequencing enriched 5′-ends of primary transcripts, J. Biotechnol. 190 (2014) 85–95. [12] P. Schwientek, S. Wendler, A. Neshat, C. Eirich, C. Rückert, A. Klein, et al., Comparative RNA-sequencing of the acarbose producer actinoplanes sp. SE50/110 cultivated in different growth media, J. Biotechnol. 167 (2013) 166–177. [13] S. Wendler, D. Hürtgen, J. Kalinowski, A. Klein, K. Niehaus, F. Schulte, et al., The cytosolic and extracellular proteomes of actinoplanes sp. SE50/110 led to the identification of gene products involved in acarbose metabolism, J. Biotechnol. 167 (2013) 178–189. [14] S. Wendler, A. Otto, V. Ortseifen, F. Bonn, A. Neshat, S. Schneiker-Bekel, et al., Comprehensive proteome analysis of actinoplanes sp. SE50/110 highlighting the location of proteins encoded by the acarbose and the pyochelin biosynthesis gene cluster, J. Proteome 125 (2015) 1–16. [15] A. Otto, J. Bernhardt, H. Meyer, M. Schaffer, F.-A. Herbst, J. Siebourg, et al., Systems-wide temporal proteomic profiling in glucose-starved Bacillus subtilis, Nat. Commun. 1 (2010) 137. [16] B. Hessling, F. Bonn, A. Otto, F.-A. Herbst, G.-M. Rappen, J. Bernhardt, et al., Global proteome analysis of vancomycin stress in Staphylococcus aureus, Int. J. Med. Microbiol. 303 (2013) 624–634. [17] M.J. MacCoss, C.C. Wu, H. Liu, R. Sadygov, J.R. Yates, A correlation algorithm for the automated quantitative analysis of shotgun proteomics data, Anal. Chem. 75 (2003) 6912–6921. [18] F. Bonn, J. Bartel, K. Büttner, M. Hecker, A. Otto, D. Becher, Picking vanished proteins from the void: how to collect and ship/share extremely dilute proteins in a reproducible and highly efficient manner, Anal. Chem. 86 (2014) 7421–7427.

[19] J.A. Vizcaíno, R.G. Côté, A. Csordas, J.A. Dianes, A. Fabregat, J.M. Foster, et al., The PRoteomics IDEntifications (PRIDE) database and associated tools: status in 2013, Nucleic Acids Res. 41 (2013) D1063–D1069. [20] S.K. Park, J.D. Venable, T. Xu, J.R. Yates, A quantitative analysis software tool for mass spectrometry-based proteomics, Nat. Methods 5 (2008) 319–322, http://dx.doi.org/ 10.1038/nmeth.1195. [21] B. Zybailov, A.L. Mosley, M.E. Sardiu, M.K. Coleman, L. Florens, M.P. Washburn, Statistical analysis of membrane proteome expression changes in Saccharomyces cerevisiae, J. Proteome Res. 5 (2006) 2339–2347. [22] M. Zhou, J. Boekhorst, C. Francke, R.J. Siezen, LocateP: genome-scale subcellularlocation predictor for bacterial proteins, BMC Bioinf. 9 (2008) 173, http://dx.doi.org/ 10.1186/1471-2105-9-173. [23] S.H. Phadnis, M.H. Parlow, M. Levy, D. Ilver, C.M. Caulkins, J.B. Connors, et al., Surface localization of Helicobacter pylori urease and a heat shock protein homolog requires bacterial autolysis, Infect. Immun. 64 (1996) 905–912. [24] A. Frisk, C.A. Ison, T. Lagergård, GroEL heat shock protein of haemophilus ducreyi: association with cell surface and capacity to bind to eukaryotic cells, Infect. Immun. 66 (1998) 1252–1257. [25] B. Bukau, P. Reilly, J. McCarty, G.C. Walker, Immunogold localization of the DnaK heat shock protein in Escherichia coli cells, J. Gen. Microbiol. 139 (1993) 95–99. [26] C. Hennequin, F. Porcheray, A. Waligora-Dupriet, A. Collignon, M. Barc, P. Bourlioux, et al., GroEL (Hsp60) of Clostridium difficile is involved in cell adherence, Microbiology 147 (2001) 87–96. [27] A.O. Helbig, A.J.R. Heck, M. Slijper, Exploring the membrane proteome–challenges and analytical strategies, J. Proteome 73 (2010) 868–878. [28] K. Goeke, A. Drepper, H. Pape, Formation of acarbose phosphate by a cell-free extract from the acarbose producer actinoplanes sp, J. Antibiot. 49 (1996) 661–663 (Tokyo). [29] A. Drepper, H. Pape, Acarbose 7-phosphotransferase from actinoplanes sp.: purification, properties, and possible physiological function, J. Antibiot. 49 (1996) 664–668 (Tokyo). [30] S. Schönert, T. Buder, M.K. Dahl, Identification and enzymatic characterization of the maltose-inducible alpha-glucosidase MalL (sucrase-isomaltase-maltase) of Bacillus subtilis, J. Bacteriol. 180 (1998) 2574–2578. [31] S. Schönert, T. Buder, M.K. Dahl, Properties of maltose-inducible alpha-glucosidase MalL (sucrase-isomaltase-maltase) in Bacillus subtilis: evidence for its contribution to maltodextrin utilization, Res. Microbiol. 150 (1999) 167–177. [32] L.B. Willis, G.C. Walker, A novel Sinorhizobium meliloti operon encodes an alphaglucosidase and a periplasmic-binding-protein-dependent transport system for alpha-glucosides, J. Bacteriol. 181 (1999) 4176–4184. [33] J.B. Jensen, N.K. Peters, T.V. Bhuvaneswari, Redundancy in periplasmic binding proteindependent transport systems for trehalose, sucrose, and maltose in Sinorhizobium meliloti, J. Bacteriol. 184 (2002) 2978–2986. [34] C. Brunkhorst, E. Schneider, Characterization of maltose and maltotriose transport in the acarbose-producing bacterium actinoplanes sp, Res. Microbiol. 156 (2005) 851–857. [35] R. Jansen, M. Gerstein, Analysis of the yeast transcriptome with structural and functional categories: characterizing highly expressed proteins, Nucleic Acids Res. 28 (2000) 1481–1488. [36] D. Greenbaum, R. Jansen, M. Gerstein, Analysis of mRNA expression and protein abundance data: an approach for the comparison of the enrichment of features in the cellular population of proteins and transcripts, Bioinformatics 18 (2002) 585–596. [37] Y. Ishihama, T. Schmidt, J. Rappsilber, M. Mann, F.U. Hartl, M.J. Kerner, et al., Protein abundance profiling of the Escherichia coli cytosol, BMC Genomics 9 (2008) 102. [38] S. Karlin, J. Mrazek, A. Campbell, D. Kaiser, Characterizations of highly expressed genes of four fast-growing bacteria, J. Bacteriol. 183 (2001) 5025–5040. [39] D. Becher, K. Hempel, S. Sievers, D. Zühlke, J. Pané-Farré, A. Otto, et al., A proteomic view of an important human pathogen–towards the quantification of the entire Staphylococcus aureus proteome, PLoS One 4 (2009), e8176. [40] S. Wolff, H. Antelmann, D. Albrecht, D. Becher, J. Bernhardt, S. Bron, et al., Towards the entire proteome of the model bacterium Bacillus subtilis by gel-based and gel-free approaches, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 849 (2007) 129–140. [41] U. Haussmann, A. Poetsch, Global proteome survey of protocatechuate- and glucosegrown Corynebacterium glutamicum reveals multiple physiological differences, J. Proteome 75 (2012) 2649–2659. [42] E. Bordignon, M. Grote, E. Schneider, The maltose ATP-binding cassette transporter in the 21st century—towards a structural dynamic perspective on its mode of action, Mol. Microbiol. 77 (2010) 1354–1366. [43] G.P. Van Wezel, J. White, M.J. Bibb, P.W. Postma, The malEFG gene cluster of Streptomyces coelicolor A3(2): characterization, disruption and transcriptional analysis, Mol. Gen. Genet. 254 (1997) 604–608. [44] G.P. Van Wezel, J. White, P. Young, P.W. Postma, M.J. Bibb, 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 (1997) 537–549. [45] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment search tool, J. Mol. Biol. 215 (1990) 403–410. [46] M. Sumiya, E.O. Davis, L.C. Packman, T.P. McDonald, P.J. Henderson, Molecular genetics of a receptor protein for D-xylose, encoded by the gene xylF, in Escherichia coli, Recept. Channels 3 (1995) 117–128. [47] R.G. Ankenbauer, E.W. Nester, Sugar-mediated induction of Agrobacterium tumefaciens virulence genes: structural specificity and activities of monosaccharides, J. Bacteriol. 172 (1990) 6442–6446. [48] F. He, G.R. Nair, C.S. Soto, Y. Chang, L. Hsu, E. Ronzone, et al., Molecular basis of ChvE function in sugar binding, sugar utilization, and virulence in Agrobacterium tumefaciens, J. Bacteriol. 191 (2009) 5802–5813.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023

S. Wendler et al. / Journal of Proteomics xxx (2015) xxx–xxx [49] S.L. Doty, M. Chang, E.W. Nester, The chromosomal virulence gene, chvE, of Agrobacterium tumefaciens is regulated by a LysR family member, J. Bacteriol. 175 (1993) 7880–7886. [50] G.A. Cangelosi, R.G. Ankenbauer, E.W. Nester, Sugars induce the Agrobacterium virulence genes through a periplasmic binding protein and a transmembrane signal protein, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 6708–6712.

9

[51] A. Licht, H. Bulut, F. Scheffel, O. Daumke, U.F. Wehmeier, W. Saenger, E. Schneider, A. Vahedi-Faridi, Crystal structures of the bacterial solute receptor AcbH displaying an exclusive substrate preference for β-d-galactopyranose, J. Mol. Biol. 406 (1) (2011 Feb 11) 92–105.

Please cite this article as: S. Wendler, et al., Comparative proteome analysis of the Actinoplanes sp. SE50/110 grown with maltose or glucose shows minor differences for acarbose biosynthesis proteins..., J Prot (2015), http://dx.doi.org/10.1016/j.jprot.2015.10.023