Growth-dependent Phosphorylation of the PtsN (EIINtr) Protein of Pseudomonas putida*

The nitrogen-related branch of the phosphoenolpyruvate: carbohydrate phosphotransferase system (PTS) of Pseudomonas putida includes the ptsN gene encoding the EIINtr (PtsN) enzyme. Although the implication of this protein in a variety of cellular functions has been observed in diverse bacteria, the physiological signals that bring about phosphorylation/dephosphorylation of the PtsN protein are not understood. This work documents the phosphorylation status of the EIINtr enzyme of P. putida at various growth stages in distinct media. Culture conditions were chosen to include fructose (the uptake of which is controlled by the PTS) or glucose (a non-PTS sugar in P. putida) in minimal medium with casamino acids, ammonia, or nitrate as alternative nitrogen sources. To quantify the relative ratio of PtsN/PtsN ∼ P in live cells, we resorted to the in situ electrophoresis of whole bacteria expressing an E-epitope-tagged EIINtr followed by the fractionation of the thereby released native proteome in a non-denaturing gel. Although the PtsN species phosphorylated in amino acid His68 was detected under virtually all growth scenarios, the relative levels of the non-phosphorylated form varied dramatically depending on the growth phase and the nutrients available in the medium. The share of phosphorylated PtsN increased along growth in a fashion apparently independent of any trafficking of sugars. The large variations of non-phosphorylated PtsN in different growth conditions, in contrast to the systematic excess of the phosphorylated PtsN form, suggested that the P-free PtsN is the predominant signaling species of the protein.

The phosphoenolpyruvate:carbohydrate phosphotransferase systems (PTS) 2 are among the most widespread protein phosphorylation setups in both Gram-positive and Gram-negative bacteria. The PTS systems generally participate in the uptake of sugars into the cells by means of a cascade of high energy phosphate transfer (1,2). As shown schematically in Fig.  1A, the archetypal PTS system consists of enzyme I (EI), HPr, and enzyme II (EII). Due to their ability to interact with other proteins in the cytoplasm, the set of phosphorylated and nonphosphorylated PTS proteins (more often HPr and EIIA) fulfills regulatory functions as diverse as catabolite repression, chemotaxis, the regulation of cyclic AMP synthesis, or nitrogen assimilation (2)(3)(4). Apart from these PTS components involved in sugar transport, many prokaryotes also have PTS branches that are not involved in carbohydrate traffic. This is because their EII component fails to have the permease partners (the so-called EIIB and EIIC domains) needed for sugar intake. However, such proteins still participate in regulation of some processes in a fashion dependent on their phosphorylation state. The PtsN protein of Pseudomonas putida (Fig. 1B) is one example of this case (5,6). This polypeptide is orthologous to the so-called IIA Ntr protein of Escherichia coli. This product is hypothesized to play a role in N-metabolism because of the clustering of the ptsN gene with rpoN, which encodes the alternative, nitrogen-related sigma factor 54 (3). Sequence comparison with other IIA Ntr proteins revealed that PtsN can most likely be phosphorylated on the His 68 residue (6). Indeed, the IIA Ntr protein of E. coli can be phosphorylated in vitro both by HPr and by the homologous protein NPr (3), although this has never been demonstrated to occur in vivo.
The question of visualizing the phosphorylation state of PtsN in vivo is of essence for understanding the physiological control of biodegradative gene expression in P. putida and other soil bacteria. We (6) and others (8) have previously identified the ptsN gene as one of the elements that mediates the C-source repression of m-xylene catabolism determined by the pWW0 plasmid of P. putida mt-2. The principal m-xylene-responsive 54 -dependent promotor of this system (called Pu) is regulated not only by the presence of pathway substrates but also subject to different physiological inputs that cause its down-regulation in vivo (6, 9 -13). Specifically, glucose and other carbohydrates of the Entner-Doudoroff pathway repress transcription from Pu (14) through a mechanism that involves the phosphorylation of the His 68 residue of PtsN. The course of the phosphotransfer in the process or the origin and fate of the high energy phosphate that passes through PtsN in vivo is, however, difficult to assess. This is because phospho-histidines are labile in any of the procedures that allow detection of other phospho-amino acids (15,16). To overcome this difficulty, we have resorted to the release of much of the intact proteome of live P. putida cells directly into a native polyacrylamide gel 3 by means of the in situ electrophoresis of whole bacteria. With this experimental setup * This work was supported in part by European Union grants of the Sixth Framework Program. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed. in hand, we were able to examine the phosphorylation state of PtsN of P. putida through various growth stages and distinct nutrient conditions. These included media with fructose (a PTS sugar in P. putida (18)) or glucose (a non-PTS sugar (18)), along with casamino acids, ammonia, or nitrate as N-sources. The data presented below revealed that the phosphorylated form of PtsN systematically accumulates along growth regardless of the composition of the culture medium. Furthermore, our results suggest that the P-free form of PtsN is the one with a superior regulatory capacity.

MATERIALS AND METHODS
Strains, Plasmid, and Growth Conditions-All Pseudomonas strains used in this work were derived from strain P. putida MAD2, a derivative of the reference strain P. putida KT2440 (19). The P. putida MAD2 variants bearing directed chromosomal insertions of gene ptsN, with either a kanamycin (Km) resistance gene or the xylE marker, have been described before (6,9). Plasmid pVLTptsN tag, encoding a variant of the ptsN gene with a short E-tag epitope added in its C terminus, was constructed as follows. First, the ptsN sequence was amplified by PCR using the oligonucleotides F154SD (5Ј-CGGAATTCAGGAGATAG-AATGATCCGACTTGAAACCATC-3Ј) and RptsNtag-5Ј (5Ј-GCTCTAGAT CACGCGGCACGCGGTTCCAGCGG-ATCCGGATACGGCACCGGCGCACCTCCGTGCTCG-TTCTGTACGTCCAGG-3Ј). The product of such amplification bore an EcoRI restriction site (underlined) and a ribosome binding site (bold) at the 5Ј-end of the ptsN gene sequence, as well as a downstream sequence encoding the E-tag (bold) followed by an extra XbaI restriction site (underlined). The resulting DNA was subsequently ligated to the polylinker of pVLT31 (20) as an EcoRI-XbaI fragment, and the ligation rescued in E. coli DH5␣ cells to yield the plasmid of interest, pVLTptsN tag. Plasmid pVLTptsNHA_E, encoding an E-tagged ptsN variant bearing an H68A change in its primary amino acid sequence, was constructed in an analogous manner using plas-mid pJMT/154-H68A (6) as the PCR template. These plasmids were transferred by conjugation to the P. putida strains indicated, in each case using E. coli HB101 (RK600) as the helper of a triparental mating procedure (21). The P. putida strain bearing a chromosomally encoded E-tagged version of PtsN that otherwise keeps its native promoter and stoichiometry was constructed as follows. The 362-bp XmaI/XbaI fragment of pVLTptsN tag, encoding the C-terminal part of the protein fused to the E-tag, was recloned in the pir-dependent replication vector pJP5603, which has a Km resistance cassette and an oriT origin of transfer for RP4-mediated mobilization (22). The resulting plasmid pJPptsN⌬N was subsequently passed from E. coli CC118 pir into P. putida by conjugation to promote formation of co-integrates by homologous recombination between the chromosomal ptsN gene and the plasmid bearing the truncated ЈptsN E-tag segment (see "Results"). Km-resistant P. putida exconjugants were examined by PCR for integration of pJPptsN⌬N into the chromosome by a single DNA crossover event between the homologous sequences. One clone fulfilling all criteria was then kept for further analysis. Unless indicated otherwise, cells were grown at 30°C in either rich LB medium or synthetic mineral M9 medium (23) supplemented with 0.2% casamino acids and 5 g/ml tetracycline or 50 g/ml kanamycin, in the presence or absence of 0.2% glucose or fructose. To analyze the influence of the N-source, cells were grown in nitrogen-free M9 medium supplemented with either 10 mM NH 4 Cl or 2 mM NaNO 3 . For induction of ptsN expression born by plasmids pVLTptsN tag and pVLTptsNHA_E, isopropyl-1thio-␤-D-galactopyranoside was added to the cultures to a final concentration of 1 mM.
Sample Preparation and Processing-To obtain samples for analysis in the native PAGE system, 3 cells were grown in the media indicated, amended where required with suitable antibiotics and isopropyl-1-thio-␤-D-galactopyranoside. The cells were then harvested by centrifugation (2 min, 14,000 rpm, 4°C) at the time points indicated, and the pellets were resuspended in a non-denaturing loading buffer (10% glycerol, 40 mM glycine, 5 mM Tris, pH 8.9, and 0.005% w/v bromphenol blue) such that a cell mass equivalent of 1 ml of culture at an A 600 ϭ 1.0 was adjusted to disperse in 200 l of the loading buffer. 10 l of such intact cell suspensions were directly loaded into the wells of the non-denaturing gel system described below. Where required, protein specimens were stored at Ϫ20°C until use (never longer than 1 week).
Native and Denaturing PAGE and Western Blotting-The non-denaturing gel electrophoresis system employed throughout this work is explained elsewhere. 3 In brief, the gels consisted of a 10% polyacrylamide (10% (v/v) acrylamide/bis-acrylamide solution (29:1) polymerized with 0.05% (v/v)TEMED and 0.05% ammonium persulfate in 1ϫ running buffer (200 mM glycine, 25 mM Tris, pH 8.9) and were assembled in a mini-Protean gel box (Bio-Rad). 10 l of the protein samples prepared as described above and adjusted to contain equivalent cell numbers were loaded in each well, and electrophoresis was carried out with a fixed current of 12.5 mA/gel (8.5 ϫ 6.5 cm, width ϫ height) at 4°C for 35 min. The proteins were then transferred onto a polyvinylidene difluoride membrane (Immobilon-P, Millipore) using a semidry transfer apparatus (Bio-FIGURE 1. Organization of the phosphotransfer chain in the phosphoenolpyruvate:carbohydrate phosphotransferase systems. A, the core PTS involves a flow of high energy phosphate (ϳP) that originates in phosphoenolpyruvate (PEP) and runs through the EI, HPr, and EII enzymes (41). The last is commonly composed of three domains (which may also appear as separated proteins): IIA, which receives the phosphate from HPr; IIB; and IIC (17). The membrane-bound component EIIC is a permease that couples phosphorylation of a specific sugar to its internalization (1,3). This basic scheme has a large number of variations, including the fusion of PTS protein domains as a single polypeptide or the combination of PTS to non-PTS protein domains (2). B, the abridged N-related PTS system of P. putida includes three proteins of the EI, HPr, and EIIA type (named, respectively, PtsP/EI Ntr , PtsO/NPr, and PtsN/EIIA Ntr ), which, by similarity to the equivalent products in E. coli, are believed to circulate ϳP groups as shown in the scheme. However, the end point of the putative phosphotransfer chain is uncertain because of the lack of cognate EIIB and EIIC partners (2).

The PtsN (EII Ntr ) Protein of P. putida
Rad) as described (24). The anti-E-tag monoclonal antibodyperoxidase conjugate (Amersham Biosciences) was applied for luminescent detection of the PtsN-E-tag fusion proteins with the procedure described (24). The SuperSignal West Femto maximum sensitivity substrate (Pierce Biotechnology) system was employed for detection of chromosomally encoded E-tagged ptsN, according to the supplier's manual. The intensity of the PtsN bands was quantified in a molecular imager Versadoc 4000 (Bio-Rad) with the Quantity One software.

RESULTS
PtsN Occurs in Vivo as Two Protein Species Differing by Phosphorylation of His 68 -The first step for examining the effect of various physiological conditions on the phosphorylation state of PtsN was to authenticate the protein forms that are released from P. putida cells upon in situ electrophoresis. 3 To this end, we employed an epitope-containing variant of the PtsN protein fused to an E-tag sequence so that we could accurately follow its presence in various types of samples. As shown in Fig. 2, the PtsN protein freed from live P. putida ptsN::Km (pVLTptsN tag) cells grown in LB separated into two bands (lane 3) in the native gel system, possibly reflecting the phosphorylated and the non-phosphorylated forms. Unfortunately, neither are P ϳ His bonds good substrates for commercially available phosphatases that act well on P ϳ Ser and P ϳ Tyr, nor do reliable procedures exist yet for analyzing P ϳ His peptides by mass spectrometry. We thus chose a genetic procedure for recognizing phosphorylation of PtsN and for tracing it to a distinct position within the amino acid sequence of the protein. This consisted of examining the gel migration properties of a PtsN variant in which the conserved His 68 residue, that is, the substrate of phosphorylation by cognate kinases, has been exchanged by an Ala and is thus locked in a non-phosphorylated form. P. putida cells bearing pVLTptsNHA_E were subject to the same treatment as before and examined in the non-denaturing system. The data of Fig. 2 show that such an H68A variant produces one band that migrates to the same position as the slower-migrating protein of the wild-type PtsN but lacks the faster-migrating form (the extra E-tagged PtsN H68A band observed in lane 2 might be a complex of PtsN with another cytoplasmic protein). This unambiguously verified that PtsN of P. putida exists in vivo in two forms, which differ in the phosphorylation state of residue His 68 .
Visualizing PtsN ϳ P Evolution along Growth-Once we had in hand a reliable procedure for detecting the two forms of PtsN derived from live cells, we set out to examine the ratio between them under various growth conditions. As a first approach, we opted for monitoring the evolution of PtsN species in cells growing in rich LB medium, bearing an E-tagged version of the protein expressed from its native promoter in monocopy gene dosage. This was based on the use of one P. putida strain in which the E-tag had been added to the genomic sequence of ptsN by a single-crossover homologous recombination with plasmid pJPptsN⌬N (Fig. 3A). This strain was grown in LB, samples were taken at the transition between exponential and stationary phase (Fig. 3B), and cells were analyzed in the native PAGE system explained above. Production of the two PtsN forms could be clearly visualized (Fig. 3C, upper panel), thereby revealing two valuable details. First, it revealed that the onset of the stationary phase was accompanied by the prevalence of the phosphorylated form of PtsN. Second, it revealed that the total level of the PtsN product appeared to decline, in any case, as growth proceeded into the late stationary phase.
Due to the intrinsic production of PtsN at relatively low levels in vivo, the signals from the chromosomally encoded, E-tagged product in the Western blot were quite weak, almost at the detection limit of the procedure. A super-sensitive detection system and longer exposure times had to be used for identification of the bands (see "Materials and Methods"). To overcome this limitation, we resorted to expressing the same protein from plasmid pVLTptsN tag for both raising the level of intracellular protein and uncoupling expression from its native promoter. To ensure that this change still reflected the physiological scenario, we followed the evolution of PtsN in P. putida ptsN::Km (pVLTptsN tag) cells grown in LB under the same conditions of the chromosomally ptsN-tagged P. putida strain. The result (Fig. 3C, lower panel) reproduced dependably the behavior of the chromosomal ptsN-tagged strain (including the sharp decrease of PtsN signals at late stationary phase). Moreover, quantification of the products was facilitated by a higher expression of the proteins, and deviations between experiments decreased. All this endorsed P. putida ptsN::Km (pVLTptsN tag) as a strain of choice for examining the relative ratio of PtsN species in connection to the presence of diverse C-sources or N-sources in the medium.
PtsN ϳ P Accumulates with Growth Regardless of the Culture Medium-As the phosphorylation state of PtsN is an indicator of the passage of high energy phosphate between the components of the abridged PTS system sketched in Fig. 1B, the pro-  1 and 2). See "Results" for a comment on the abnormal migration of the sample in lane 2.

The PtsN (EII Ntr ) Protein of P. putida
cedure and the strain P. putida ptsN::Km (pVLTptsN tag) presented above were instrumental to follow the evolution of the relative PtsN/PtsN ϳ P ratio along growth in the following media: (i) control, LB, excess of C-source and N-source; (ii) mineral medium with casamino acids (excess N-source) with either glucose (a non-PTS sugar in P. putida) or fructose (a PTS sugar); and (iii) mineral medium with glucose as C-source and either ammonium (an easily metabolizable N-source) or nitrate (a suboptimal N-source). It should be emphasized that the information pursued with such experiments was the relative ratio between PtsN and PtsN ϳ P, not the absolute amount of the PtsN protein, which decreases in any case during late stationary phase. As the proportion between the two PtsN forms depends exclusively of the phosphorylation/dephosphorylation machinery, we assume that such a ratio grossly reflects the functioning of the corresponding PTS branch (see below).
Figs. 4 -6 show the accumulation of the two forms of PtsN under different growth settings. The first noticeable feature of all conditions is the systematic occurrence of a considerable level of the faster-migrating phosphorylated protein species, in contrast with the more variable non-phosphorylated form. Furthermore, in all cases, the predominant form of PtsN in the stationary phase is the phospho-containing species. On the contrary, the presence of the non-phosphorylated form varies significantly depending on the growth medium and the growth stage. For instance, Fig. 4 shows that there is a considerable amount of the non-phosphorylated PtsN in glucose-grown cells at a growth phase in which the same form has virtually disappeared from cells grown in the presence of fructose. In Fig. 5, the various ratios of PtsN/PtsN ϳ P species are shown as a function of growth phase. Although all of them shared the predominance of the phosphorylated form at late growth stages, the disappearance of the non-phosphorylated species seems to occur earlier in the fructose-grown cells. A different case is posed by varying the N-source from ammonia (NH 4 ϩ ) to nitrate (NO 3 Ϫ ) in the culture (Fig. 6). This anion cannot be used as a terminal electron acceptor by P. putida as this bacterium lacks the key enzymes respiratory nitrate and nitrite reductase that would be required for such a function (25,26). However, NO 3 Ϫ can be employed as a nitrogen source, through the action of assimilatory nitrate and nitrite reductases, which convert nitrate into nitrite and then into metabolizable NH 4 ϩ . The physiological change caused by NO 3 Ϫ is, therefore, the replacement of an easy N-source by a less direct counterpart. The result shown in Fig. 6 suggests that the phosphorylated form of PtsN is virtually the only product observed with NO 3 Ϫ , whereas both forms can be detected in NH 4 ϩ -grown bacteria. That the PtsN/ PtsN ϳ P ratios can be changed by growth phase, C-source, and N-source suggests that the phosphorylation state of this protein is more a reflection of the metabolic status of the cells than a component of any specific sugar-traffic system.

DISCUSSION
The PTS system (components of which are encoded in almost every bacterial genome known (2)) has evolved to provide a mechanism by which the activities of distinct sets of proteins are regulated epigenetically. This occurs by means of A, E-tagging the genomic ptsN sequence by homologous recombination. The sketch summarizes the steps followed to deliver a C-terminal E-tag to the chromosomally encoded ptsN gene. This involved the mobilization of suicide plasmid pJPptsN⌬N (which encodes a truncated, E-tagged ЈptsN sequence) and the formation of a co-integrate with the cognate chromosomal region (the coordinates of the enlarged genomic context are indicated). This leaves a single copy of the E-tagged ptsN expressed under the same promoter as the original ptsN gene. The reorganization of the resulting chromosomal region after cointegration was verified by PCR (not shown). B, growth of P. putida cells bearing the E-tagged ptsN gene. Bacteria were cultured at 30°C in LB with Km added. Samples were collected at the times indicated with arrows. C, evolution of PtsN forms along growth. Intact cells were loaded in the native PAGE system followed by Western blot analysis with an anti-E-tag antibody. The results with the strain bearing a monocopy (mono) E-tagged ptsN (upper panel) are compared with those of an equivalent experiment with P. putida ptsN::Km (pVLTptsN tag, multi) at the lower panel.

The PtsN (EII Ntr ) Protein of P. putida
their interactions with the phosphorylated and non-phosphorylated forms of each of the canonical EI, HPr, and EII components and their variants (4). Similarly, the intensity of the flow of high energy phosphate between phosphoenolpyruvate and the terminal phosphate acceptor (sugar or otherwise) is translated in a given ratio of phosphorylated versus non-phosphorylated PTS proteins. In turn, such ratios between each protein form coordinately enhance or inhibit groups of otherwise unre-lated functions. The intriguing side of what appears to be a separate branch of the PTS system (Fig. 1B) is that the protein encoded by ptsN (EIIA Ntr ) lacks the membrane-associated permease moieties EIIB and EIIC that typically tie PTS proteins to sugar transport (4). An early hint about the functions of such proteins was derived from the genetic association of ptsN (and the adjacent ptsO gene) to rpoN in several bacteria (3). rpoN encodes a major sigma factor ( 54 ) involved among many other functions (27)(28)(29) in nitrogen metabolism. The proposition thus was that these PTS genes could be related to sensing the N versus C balance (i.e. the excess of one nutrient source versus the other), through a thus far unknown mechanism. One useful clue in this direction is the recent observation that the P. putida EI Ntr , NPr, and EII Ntr proteins act in concert to control the intracellular accumulation of polyhydroxylalkanoates, typical products of carbon overflow in respect to other essential nutrients, e.g. nitrogen (18).
The whole set of EI Ntr , NPr, and EIIA Ntr proteins of E. coli have been purified and shown in vitro to sustain a typical flow of high energy phosphate: phosphoenolpyruvate 3 EI Ntr 3 NPr 3 EIIA Ntr (3,30). The three-dimensional structure of PtsN (EIIA Ntr ) has been determined (31,32), and its interactions in solution with NPr have been studied (33). ptsN mutants of E. coli have been generated and subjected to a variety of phenotypic analyses. For instance, ptsN mutants are extremely sensitive to leucine-containing peptides (LCPs (34)), a phenomenon that can be traced to the interaction of EIIA Ntr with the K ϩ transporter TrkA (35). In addition, growth of the ptsN mutant of E. coli is inhibited by several sugars and tricarboxylic acid cycle intermediates in a medium containing an amino acid or nucleoside base as a combined source of nitrogen and carbon (3). Such an inhibition can be reverted by supplying ammonium salts, an indication that ptsN is indeed related to N-metabolism. ptsN mutants of P. putida change the sensitivity of the m-xylene-responsive 54 promoter Pu to the presence of glucose in the medium (6, 8, 36 -38) and alter the expression pattern of a large number of proteins of the cell proteome (36). ptsN is also involved in the response of Pasteurella multocida to iron starvation (39). Finally, inactivation of ptsN in Rhizobium etli reduces growth on medium containing succinate, lessens production of melanin, and inhibits expression of the N-fixation gene nifH (40). Despite these multiple observations, the phosphorylation state of PtsN in vivo or the physiological circumstances that alter such a state have not been examined before.
Perusal of the results shown in Figs. 3-6 clearly indicated that, regardless of the culture conditions: (i) the phosphorylated form of PtsN is present in all growth stages; (ii) PtsN ϳ P accumulates in the later stationary phase; and (iii) the non-phosphorylated PtsN protein appears associated to rapid growth. Although the expression of the ptsN gene from a heterologous promoter of a plasmid does unbalance the native stoichiometry of the system, we have shown that our experimental setup reasonably reflects the physiological scenario, probably because equilibrium is reached in the phosphorylation of all the proteins of the PTS system. Although we detect phosphorylated PtsN in virtually all growth conditions, the non-phosphorylated form seems to be subject to growth-dependent variations. If so, the protein form with more regulatory significance should be the  Km (pVLTptsN tag). Cells were grown in the media indicated, and samples were collected at the times indicated by arrows. The relative contents of the two PtsN forms were then determined in live cells as before. The bar diagrams represent the average of at least three separate experiments. Note the important presence of non-phosphorylated PtsN in cells grown in casamino acids (CAA) and glucose as compared with bacteria grown in fructose. In either case, the non-phosphorylated PtsN forms virtually disappear during the late stationary phase.  Km (pVLTptsN tag). Cells were grown at 30°C in N-free M9 minimal medium and glucose, supplemented with either 10 mM ammonia (NH 4 ϩ ) or 2 mM nitrate (NO 3 Ϫ ). Culture samples were collected at the late exponential (late exp, lanes 1 and 2, A 600 of 0.8 for NH 4 ϩ or 0.3 for NO 3 Ϫ ) and early stationary (early stat, lanes 3 and 4, A 600 of 1.5 for NH 4 ϩ or 0.7 for NO 3 Ϫ ) phase, and cells were subject to the standard analysis of the PtsN forms. Note that PtsN ϳ P is the only species that appears in cells growing in NO 3 Ϫ as N-source.
non-phosphorylated PtsN species, the one that clearly varies with the carbon source at the onset of stationary phase. This notion is consistent with the appearance of a slower-migrating band in non-denaturing gel analysis of an extract expressing the H68A variant of PtsN (Fig. 2, lane 2), which may consist of complexes between PtsN and other proteins. Moreover, our results show that the disappearance of the non-phosphorylated PtsN does not occur simultaneously in all cases. Instead, the decline of the P-less species seems to happen earlier in the presence of fructose than with glucose (Fig. 5) and earlier in the presence of nitrate (NO 3 Ϫ ) than with ammonia (NH 4 ϩ , Fig. 6). The regulatory significance of these differences is still unclear and deserves further studies. However, they do suggest that it is the physiological conditions of the bacteria and not so much the traffic of specific sugars (as it happens with the canonical PTS) that dictates the relative phosphorylation states of the PtsN protein and possibly of the other enzymes of the N-related PTS of P. putida.