Biotechnology Letters 25: 143–147, 2003. © 2003 Kluwer Academic Publishers. Printed in the Netherlands.
143
Arrest of cell cycle by inhibition of ribonucleotide reductase induces accumulation of NAD+ by Mn2+ -supplemented growth of Corynebacterium ammoniagenes Bouziane Abbouni, Hesham M. Elhariry & Georg Auling∗ Institut für Mikrobiologie, Universität Hannover, Schneiderberg 50, D-30167 Hannover, Germany ∗ Author for correspondence (Fax:+49-511-7625287; E-mail:
[email protected]) Received 4 October 2002; Revisions requested 16 October 2002; Revisions received 8 November 2002; Accepted 9 November 2002
Key words: cell cycle, Corynebacterium ammoniagenes, elongation, hydroxyurea, manganese-ribonucleotide reductase, NAD+ , p-methoxyphenol, scanning electron microscopy
Abstract Cell division of the wild type strain Corynebacterium (formerly Brevibacterium) ammoniagenes ATCC 6872 which requires 1 µM Mn2+ for balanced growth was inhibited by addition of 20 mM hydroxyurea (HU) or 10 mM pmethoxyphenol (MP) to a Mn2+ -supplemented fermentation medium at an appropriate time. Scanning electron microscopy (SEM) showed a restricted elongation characteristic of arrest of the cell cycle in coryneform bacteria. The cultures treated with HU or MP had, respectively, a fourfold or sixfold enhanced accumulation of NAD+ by a salvage biosynthetic pathway. An assay of nucleotide-permeable cells for ribonucleotide reductase activity using [3 H-CDP] as substrate revealed a pre-early and complete decline of DNA precursor biosynthesis not found in the untreated control. Overproduction of NAD+ is an alternative to the conventional fermentation process using Mn2+ deficiency. A simple model is presented to discuss the metabolic regulation of the new process based on the presence of a manganese ribonucleotide reductase (Mn-RNR) in the producing strain. Abbreviations: Mn-RNR: manganese ribonucleotide reductase; R2F: small subunit (metallocofactor) of the MnRNR; MP: p-methoxyphenol; HU: hydroxyurea; IMP: 5 -inosine monophosphate; ADH: alcohol dehydrogenase. Introduction Coryneform bacteria are widely used for production of flavor enhancing and other nucleotides by direct fermentation of sugars into 5 -ribonucleotides (Teshiba & Furuya 1989, Nagodawithana 1994) while novel processes are available based on fermentation of sugars into nucleosides, with subsequent phosphorylation to the corresponding 5 -ribonucleotides (Mori et al. 1997, Mihara et al. 2000). Production of the coenzyme, NAD+ by a salvage biosynthetic pathway using C. ammoniagenes ATCC 6872 and a high-salt mineral medium was first shown by Nakayama et al. (1968). Microorganisms synthesize the metabolites necessary for their growth and sustenance by strong metabolic regulation and prevention of overproduction. Mn2+ -limitation is a feasible approach to
achieve perturbation of balanced growth of the Mn2+ requiring coryneform bacteria. However, it is difficult to adjust Mn2+ -concentrations in industrial fermentation media precisely below a threshold value of 1 µM. Mn2+ control of growth and DNA formation in C. ammoniagenes ATCC 6872 has been traced back to DNA precursor biosynthesis, i.e., manganese-dependent ribonucleotide reduction (Auling & Follmann 1994). Blocking this step in the biosynthesis of DNA will inevitably arrest the cell cycle. Here, we report on the production of NAD+ with C. ammoniagenes ATCC 6872 in the presence of surplus manganese by proper inhibition of DNA precursor biosynthesis with the radical scavengers HU and MP, which were shown earlier to inhibit the manganese ribonucleotide reductase (Griepenburg et al. 1998).
144 Materials and methods Media and growth conditions Inoculum (10% v/v) was prepared in seed medium which consisted of 20 g D-glucose, 10 g yeast extract, 10 g peptone, and 2.5 g NaCl (adjusted to pH 7.2– 7.4 with 2 M NaOH before sterilizing). Cultures (50 ml) of Corynebacterium ammoniagenes ATCC 6872 were grown in 500 ml baffled Erlenmeyer flasks at 27 ◦ C and 120 rpm using the fermentation medium of Nakayama et al. (1968) which contains 100 g glucose, 10 g yeast extract, 10 g KH2 PO4 , 10 g KHPO4 , 10 g MgSO4 · 7H2 O, and 6 g urea, sterilized separately (initial pH 7.4). Likewise, 12 mg calcium-pantothenate, 10 mg thiamine, and 30 µg biotin were added per liter after filtration. The protocol of Nakayama et al. (1968) was modified by supplementing the synthetic medium with 10 µM MnCl2 in order to satisfy the manganese requirement of C. ammoniagenes. The radical scavengers were added 7.5 h after inoculation to give 20 mM HU or 10 mM MP. Adenine and nicotinic acid were added at 100 mg as precursors in the salvage biosynthetic pathway per 50 ml at 7.5 h after the onset of the fermentation. For the experiments shown in Figures 1b and 2b the pH was manually adjusted when necessary.
designated as E1 and the blank value (E0 ) was determined without addition of ADH. The concentration of NAD+ (mg ml−1 ) was calculated by the equation Concentration of NAD+ = 4.11 (E1 − E0 ).
Assay of Mn-RNR activity Mn-RNR activity with nucleotide-permeable cells was assayed according to Luo et al. (1997). For the 100 µl standard RNR assay, 1 × 108 cells were added to 6 mM DTT, 1 mM MgCl2 , and 0.25 µCi [3 H-CDP] (10– 30 Ci/mmol) and 50 µM dATP as allosteric effector. After 20 min at 30 ◦ C, the reaction was stopped by boiling for 3 min. Further treatment of the reaction mixture and automated nucleoside analysis and determination of radioactivity was carried out as described by Griepenburg et al. (1998). Blank values were obtained with heat-inactivated permeabilized cells (3 min boiling). Scanning electron microscopy After dissolving of crystal salts as described above, microphotographs were taken according to Zellner et al. (1991), at 15 kV with a scanning electron microscope.
Assay of cell growth To dissolve precipitated crystals arising in the highsalt fermentation medium 1 ml of the culture was diluted with 3 ml distilled water, and 100 µl 2 M HCl were added. As the cells of C. ammoniagenes ATCC 6872 tend to adhere together 5 µl Tween 80 (20% w/v) were added and the mixture was centrifuged at lowspeed. The pellet was resuspended in distilled water, the turbidity determined, and re-calculated to cell dry weight using a standard curve. Assay of NAD+ concentration The concentration of NAD+ was enzymatically determined. Aliquots (1 ml) were withdrawn as above and the supernatant was examined. The reaction mixture contained 0.4 ml supernatant, 0.3 ml alcoholsemicarbazide-hydrochloride solution, 1.3 ml 0.1 M potassium phosphate buffer, pH 7.2. The reaction was started by dropwise addition of a yeast alcohol dehydrogenase suspension and extinction was determined at 366 nm. The stable A366nm – value obtained was
Results Arrest of cell cycle in the presence of 10 µM Mn2+ upon addition of HU or MP and concomitant overproduction of NAD+ The ability of the radical scavengers HU and MP to arrest the cell cycle of C. ammoniagenes under conditions of balanced growth was examined. In order to satisfy the manganese requirement of C. ammoniagenes the fermentation medium of Nakayama et al. (1968) was supplemented with 10 µM Mn2+ . Controls (balanced growth) had short rods (1 µm) as expected for 48 h old stationary cells with completed cell division. However, 48 h old cultures which received 20 mM HU had elongated rods (1.5–1.7 µm). Elongated rods (1.2–1.5 µm) were likewise observed when 10 mM MP was added although the effect was less distinct. Both radical scavengers were added in the exponential phase of growth, 7.5 h after the onset of the fermentation. The elongation indicates that both treatments did not allow the cells to complete division.
145
Fig. 1. Accumulation of NAD+ () due to perturbation with 20 mM HU during growth () of C. ammoniagenes ATCC 6872 without (a) and with pH adjustment (b) in the manganese-supplied fermentation medium of Nakayama et al. (1968). Growth was followed turbidometrically at 578 nm and converted to cell dry weight using a standard curve. HU was added 7.5 h after the onset of the fermentation. Open symbols refer to the unproductive, non-perturbated controls.
Remarkably, C. ammoniagenes did not display any signs of initiated septum formation upon arrest of the cell cycle with HU or MP. On the other hand, divisioninhibited cells which have initiated septum formation but cannot complete it, are generated by cephalexin (H.M. Elhariry, unpublished work, not shown). Next, the ability of C. ammoniagenes ATCC 6872 to accumulate NAD+ upon deliberate cell cycle arrest was examined. Proper addition of the radical scavengers (neither too early nor too late in the growth curve) is important for their efficacy (Auling & Follmann 1994). When 20 mM HU (Figure 1a) or 10 mM MP (Figure 2a) were added 7.5 h after the onset of fermentation 0.6 or 1 g NAD+ per l were accumulated extracellulary after 5–7 d. The accumulation of NAD+ in the culture supernatant of the controls never exceeded background values of 0.16 g per l. The treatments with both radical scavengers induced unbalanced growth (Auling & Follmann 1994), noticeable from an increase of biomass similar to the untreated control. HU merely caused a transient slight reduction compared with the control (Figure 1a), whereas addition of MP even stimulated growth in the late stages of the fermentation (Figure 2a). In additional experiments with C. ammoniagenes ATCC 6872, the 10 µM Mn2+ -supplemented cultures,
Fig. 2. Accumulation of NAD+ () due to perturbation with 10 mM MP during growth () of C. ammoniagenes ATCC 6872 without (a) and with pH adjustment (b). Same conditions as described in the legend to Figure 1.
Fig. 3. Regulated shut-down (‘peak’ behaviour) of ribonucleotide reduction () in C. ammoniagenes ATCC 6872 versus deliberate inactivation of RNR-activity due to radical scavenging with 20 mM HU (open symbols). Same conditions as described in the legend to Figure 1.
perturbated by addition of the radical scavengers as above, were adjusted at a range of pH between 6.5 and 7 after one day. Adjustment of pH led to a further enhanced (twofold) accumulation of NAD+ with HU (Figure 1b). When MP was added, NAD+ was accumulated earlier (Figure 2b) but no additional increase was observed. Growth was not stimulated in the late stages of the fermentation (cf. Figure 2a).
146
Fig. 4. NAD+ accumulation of C. ammoniagenes ATCC 6872 after addition of adenine and nicotinic acid as precursors, triggered by a deliberate arrest of the cell cycle in order to perturbate non-productive balanced growth. The simple model is based on the large information available on the cell cycle in Gram-negative bacteria (Satta et al. 1994, White 2000) and a few data on this regulatory network in coryneform bacteria (Auling & Follmann 1994, Kijima et al. 1998). Whether treatment with penicillin, cycloserine and surfactants leading to accumulation of IMP (Nara et al. 1969) will also generate NAD+ is indicated by a question mark.
Deliberate shut-down of DNA precursor biosynthesis using radical scavengers In this study, HU and MP were successfully used for arrest of cell cycle and concomitant overproduction of NAD+ with C. ammoniagenes ATCC 6872. In order to confirm our hypothesis that inhibition of DNA+ precursor biosynthesis generates the overproduction, it was necessary to assay the levels of ribonucleotide reduction with and without perturbation by HU or MP. For this purpose, RNR activity was determined in nucleotide-permeable cells periodically before and after addition of radical scavengers (Figure 3). When the balanced growth was not perturbated by any addition the maximum of RNR activity was observed in the late exponential phase. The steep increase towards and the sharp decline after the maximum of RNR activity, characteristic of a ‘peak’ enzyme, displays a wellknown behaviour of the Mn-RNR of C. ammoniagenes
(Auling & Follmann 1994). Here, an immediate decline of RNR activity was measured upon addition of 20 mM HU in the phase of exponential growth (Figure 3). Similar results were obtained by addition of 10 mM MP (not shown). Discussion A limited elongation as observed here with C. ammoniagenes was already reported for Corynebacterium glutamicum (Kijima et al. 1998). These authors used sparfloxacin as an inhibitor of DNA-gyrase and cephalexin as an inhibitor of septum peptidoglycan synthesis as a first step towards characterizing the cell division mechanism of coryneform bacteria. Their unique regulation is depicted in a simple model (Figure 4). The absence of initiated septum formation upon arrest of cell cycle may indicate the induction
147 of a response in C. ammoniagenes comparable to the SOS system in Escherichia coli. In E. coli singlestranded DNA generated from interruption of DNA replication results in the induction of the SOS system; the increasing level of SulA, a gene product of the SOS regulon, then inhibits FtsZ-ring formation and thus cell division (White 2000). Members of the SulA-family occur in C. glutamicum (Nakagawa S., http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd= Retrieve&db=nucleotide&list_uids=19551250&dopt = GenBank, unpublished). A ftsZ-homologue was identified in C. glutamicum and division-inhibited cells of this species do not show filamentation but limited elongation (Kijima et al. 1998) as observed here with C. ammoniagenes ATCC 6872. This may explain why the latter species is able to accumulate the flavorenhancing nucleotide IMP upon addition of penicillin, cycloserine or surfactants (Nara et al. 1969). It remains to be studied whether such treatments induce accumulation of NAD+ as well. Here, we have demonstrated that early events in the cell cycle allow overproduction of NAD+ by deliberate inactivation of ribonucleotide reduction causing shortage of DNA precursors in rapidly growing cultures (Figure 4). The Mn-RNR of C. ammoniagenes consists of a large catalytic subunit and a small radicalbearing metallocofactor named CA2 or R2F (Griepenburg et al. 1998) and is expected to be controlled with respect to its overall activity by the regulated stability of the limiting R2 protein as described for the Fe-RNR (Chabes & Thelander 2000). Quenching of the organic radical of the Mn-RNR inhibits DNA precursor biosynthesis and generates pre-early (Figure 3) versus regulated shut-down (‘peak’ behaviour) of ribonucleotide reduction. The stimulatory effect of the radical scavenger MP may be explained by other properties of this phenolic compound considering a study of Liebl et al. (1989) on the growth behaviour of the related species C. glutamicum, i.e., coryneform bacteria display merely a limited biosynthesis of siderophores, and their iron availability is improved by adding phenols as growth factor for rapid and abundant growth in mineral medium.
Acknowledgement This work was supported in part by Ajinomoto, Co., Inc., Japan.
References Auling G, Follmann H (1994) Manganese-dependent ribonucleotide reduction and overproduction of nucleotides in coryneform bacteria. In: Sigel H, Sigel A, eds., Metal Ions in Biological Systems, Vol. 30. Metalloenzymes Involving Amino Acid-Residues and Related Radicals. New York: Marcel Dekker Inc, pp. 131–164. Chabes A, Thelander L (2000) Controlled protein degradation regulates ribonucleotide reductase activity in proliferating mammalian cells during the normal cell cycle and in response to DNA damage and replication blocks. J. Biol. Chem. 275: 17747– 17753. Griepenburg U, Blasczyk K, Kappl R, Hüttermann J, Auling G (1998) A divalent metal site in the small subunit of manganesedependent ribonucleotide reductase of Corynebacterium ammoniagenes. Biochemistry 37: 7992–7996. Kijima N, Goyal D, Takada A, Wachi M (1998) Induction of only limited elongation instead of filamentation by inhibition of cell division in Corynebacterium glutamicum. Appl. Microbiol. Biotechnol. 50: 227–232. Liebl W, Klamer R, Schleifer K (1989) Requirement of chelating compounds for the growth of Corynebacterium glutamicum in synthetic media. Appl. Microbiol. Biotechnol. 32: 205–210. Luo C-H, Hansen J, Auling G (1997) Temperature-sensitive mutants of Corynebacterium ammoniagenes ATCC 6872 with a defective large subunit of the manganese-containing ribonucleotide reductase. Arch. Microbiol. 167: 317–324. Mihara Y, Utagawa T, Yamada H, Asano Y (2000) Phosphorylation of nucleosides by a mutated acid phosphatase from Morganella morganii. Appl. Environ. Microbiol. 66: 2811–2816. Mori H, Iida A, Fujio T, Teshiba S (1997) A novel process of inosine 5 -monophosphate production using overexpressed guanosine/inosine kinase. Appl. Microbiol. Biotechnol. 48: 693–698. Nagodawithana TW (1994) Savory flavors. In: Gabelman A, ed., Bioprocess Production of Flavor, Fragrance, and Color Ingredients. New York: John Wiley & Sons Inc., pp. 135–168. Nakayama K, Sato Z, Tanaka H, Kinoshita S (1968) Production of nucleic acid-related substances by fermentative processes. Part XVII. Production of NAD and nicotinic acid mononucleotide with Brevibacterium ammoniagenes. Agric. Biol. Chem. 32: 1331–1336. Nara T, Misawa M, Komuro T, Kinoshita S (1969) Production of nucleic acid related substances by fermentative processes. Part XXX. Effect of antibiotics and surface active agents on 5 purine-nucleotide production by Brevibacterium ammoniagenes. Agric. Biol. Chem. 33: 1198–1204. Satta G, Fontana R, Canepari P (1994) The two-competing site (TCS) model for cell shape regulation in bacteria: the envelope as an integration point for the regulatory circuits of essential physiological events. Adv. Microb. Physiol. 36: 181–245. Teshiba S, Furuya A (1989) Production of Nucleotides and Nucleosides by Fermentation. Amsterdam: Gordon and Breach Science Publishers, OPA BV. White D (2000) The Physiology and Biochemistry of Prokaryotes, 2nd edn. New York: Oxford University Press. Zellner G, Geveke M, De Macario EC, Diekmann H (1991) Population dynamics of biofilm development during start-up of a butyrate-degrading fluidized-bed reactor. Appl. Microb. Biotechnol. 36: 404–409.
