protein phosphorylation on serine, threonine, and tyrosine...

Protein phosphorylation on serine, threonine, and tyrosine residues modulates membrane-protein interactions and transcriptional regulation in Salmonella typhlmurlum Paula Cristina Ostrovsky and Stanley Maloy 1

Department of Microbiology, University of Illinois, Urbana, Illinois 61801 USA

There exists a plethora of tyrosine kinases that play essential roles in regulation of eukaryotic proteins. Several dual specificity kinases that phosphorylate proteins on threonine, serine, and tyrosine residues also play critical roles in eukaryotic phosphorylation cascades. In contrast, very few prokaryotic proteins have been shown to be phosphorylated on tyrosine residues, and the functions of the rare examples remain obscure. Furthermore, no dual specificity kinases have been described in prokaryotes. Our results indicate that PutA protein from the bacterium Salmonella typhimurium autophosphorylates on several threonine, serine, and tyrosine residues. PutA protein both represses the proline utilization ~ut) operon and degrades proline to glutamate. These two opposing functions are regulated by the availability of proline and the membrane sites needed for the proline dehydrogenase activity of PutA protein. In addition, these functions are modulated by phosphorylation of PutA protein. The rate of dephosphorylation of PutA protein is determined by the availability of proline and membranes. Dephosphorylated PutA protein has a higher DNA binding affinity than the phosphorylated protein and thus may prevent toxic overexpression of PutA protein in the absence of available membrane sites.

[Key Words: Dual specificity kinase/phophatase; transcriptional regulation; protein phosphorylation; membrane association; proline metabolism]

Received April 17, 1995; revised version accepted July 11, 1995.

Proline oxidation is widespread in prokaryotes (Costilow and Cooper 1978; Chen and Maloy 1991) and in the mi- tochondria of eukaryotic cells (Sacktor and Childress 1967; Stewart and Lai 1974; Sylvester et al. 1974; Down- ing et al. 1976; Brandriss and Magasanik 1979), allowing proline to be used as a source of carbon, nitrogen, and energy. In enteric bacteria, proline utilization requires two genes: The putP gene encodes the major proline per- mease, and the putA gene encodes a multifunctional protein (Maloy 1987). In the presence of proline, the PutA protein catalyzes the two enzymatic steps required to oxidize proline to glutamate, and in the absence of proline it acts as the transcriptional repressor of both put genes (Menzel and Roth 1981a; Maloy and Roth 1983; Ostrovsky de Spicer et al. 1991 ). These functions of PutA protein occur in different cellular compartments: Oxida- tion of proline occurs at the cytoplasmic membrane where the tightly bound FAD cofactor donates electrons directly to the electron transport chain; repression of put

1Corresponding author.

transcription occurs in the cytoplasm (Abrahamson et al. 1983; Wood 1987; Wood et al. 1987; A.M. Muro-Pastor, P.C. Ostrovsky, and S. Maloy, in prep.).

The PutA protein senses the intracellular proline concentration and the physiological state of the membranes to mediate the decision between these two opposing roles. Reduction of the FAD cofactor causes an increase in the relative hydrophobicity of PutA protein, favoring its interaction with the cytoplasmic membrane (Ostrov- sky de Spicer and Maloy 1993). Thus, when both proline and membranes are available, PutA protein interacts with the membrane and is released from its specific binding sites in the put control region DNA (Muro-Pas- tor and Maloy 1995; Ostrovsky de Spicer and Maloy 1993; A.M. Muro-Pastor, P.C. Ostrovsky, and S. Maloy, in prep.). These results suggested that the presence of proline causes PutA protein to be sequestered in the membrane where it can function as a dehydrogenase but where it lacks access to DNA.

The results presented here indicate that the mechanism for deciding between these two opposing roles also involves autophosphorylation. Protein phosphorylation

2034 GENES & DEVELOPMENT 9:2034-2041 © 1995 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/95 $5.00

Cold Spring Harbor Laboratory Press on May 28, 2020 - Published by genesdev.cshlp.orgDownloaded from

http://genesdev.cshlp.org/

http://www.cshlpress.com

Thr, Set, and Tyr phosphorylation in bacteria

is widespread in eukaryotes and prokaryotes and often it is involved in the regulation of membrane receptors or enzymes (Cozzone 1993; Edelman et al. 1987; Hunter and Cooper 1985). As with many well-characterized examples of protein phosphorylation, phosphorylation of PutA protein seems to affect its regulatory function. The phosphorylation pattern observed with PutA protein, however, has previously only been observed in eukaryotes.

Results

Purified PutA protein is phosphorylated in vitro

Several lines of evidence indicate that purified PutA protein is autophosphorylated in vitro (Fig. 1). (1) PutA protein becomes radioactively labeled when incubated with [7-32p]ATP. (2) PutA protein is not radioactively labeled when incubated with [a-32P]ATP, indicating that label- ing observed with [7-32P]ATP is not simply due to non- covalent binding of ATP or covalent adenylation (Foster et al. 1989). (3) PutA protein retains the 32p label when subjected to denaturing polyacrylamide gel electrophoresis, indicating that PutA is covalently phosphorylated. (4) Most of the 32p label on PutA protein can be removed with nonspecific alkaline phosphatases (Figs. 1A and 2), confirming that PutA is phosphorylated not simply ade- nylated, and suggesting that the phosphorylated residues are located on the surface of the protein.

PutA protein is phosphorylated on serine, threonine, and tyrosine residues

T h e 32p label is acid resistant and somewhat base labile {data not shown), indicating that PutA may be phosphorylated on multiple amino acid residues. However, inter-

A B C

1 2 3 1 2 3

Alka~ phosl

Figure 1. Phosphorylation of Purified PutA protein in vitro. (A) PutA protein is phosphorylated in the presence of [732p]ATP, and dephosphorylation is stimulated by nonspecific phosphatases or proline. (Lane 1), Labeled PutA with 1 unit of calf thymus alkaline phosphatase; (lane 2) labeled PutA with 100 mM proline; (lane 3) labeled PutA. (B) Purified PutA protein is homogeneous. Same as A, lane 3, but stained for protein with Fastain. (C) Membranes counteract the proline-induced dephosphorylation. Labeled PutA protein as in A but greatly overex- posed to illustrate the difference in label intensities. (Lane 1) Labeled PutA with 50 ng of E. coli membranes; (lane 2) labeled PutA with 50 ng of E. coli membranes and 100 mM proline; (lane 3) labeled PutA with 100 mM proline. The radiolabel was quantitated with a PhosphorImager and the results are shown in Table 1.

pretation of these results is complicated because the sta- bility of phosphoamino acids can vary greatly depending on the nearby residues in the protein (Buss and Stull 1983). Therefore, to determine which types of amino acid residues were phosphorylated, we analyzed PutA protein by two-dimensional phosphoamino acid analysis and 3Xp-nuclear magnetic resonance (NMR). The two- dimensional phosphoamino acid analysis revealed phosphoserine, phosphothreonine, and phosphotyrosine residues (Fig. 2). 31p-NMR analysis also demonstrates that PutA protein is phosphorylated. The 31P-NMR results clearly showed peaks with chemical shifts corresponding to phosphoserine and phosphothreonine at pH 8.0 but did not reveal phosphotyrosine (Fig. 3A). The apparent discrepancy between the results obtained from these two different methods could be due to several factors: (1) The expected concentration of phosphotyrosine is at the threshold of the range of sensitivity of 31P-NMR measurements; (2} two-dimensional phosphoamino acid analysis of PutA protein treated with phosphatase or proline (Fig. 2) indicated that the phosphotyrosine may not be surface-exposed and therefore not detectable by the 3Xp-NMR measurements; and (3) the phosphate moiety of phosphotyrosine often has a rapid turnover (Hunter and Cooper 1985), so the phosphotyrosine in PutA protein might be spontaneously dephosphorylated during the extended time course of the NMR experiment.

PutA protein is also phosphorylated in vivo

The experiments described above were done with purified PutA protein in vitro. To determine if the phosphorylation of PutA protein is simply an in vitro artifact, we directly examined the phosphorylation of PutA protein expressed in vivo. PutA protein from crude cell extracts was recognized by a polyclonal anti-phosphotyrosine antibody in ELISA assays (Fig. 4A) and by monoclonal anti- phosphotyrosine, anti-phosphothreonine (Fig. 5), and anti-phosphoserine (data not shown) antibodies in West- ern blot immunoassays. These results indicate that PutA protein is also phosphorylated on serine, threonine, and tyrosine residues in vivo.

PutA protein contains protein kinase-like subdomain motifs

The deduced sequence of the PutA protein {EMBL acces- sion code X70843) carboxyl terminus displays moderate similarity with several of the subdomains conserved among protein kinases (Hanks et al. 1988). A sequence similar to subdomain I, containing the ATP-binding motif, is present in PutA protein. Preliminary data suggest that mutation of the conserved glycines in this motif affects phosphorylation of PutA protein (S. Allen and S. Maloy, unpubl.). A motif similar to subdomain VI is also present, and it contains the sequence GXgLX6HRDLAXR usually indicative of a kinase with specificity for tyrosine (Hanks et al. 1988). In general, these domains in PutA protein show a higher similarity to protein-tyrosine kinase family members and in par-

GENES & DEVELOPMENT 2035




Ostrovsky and Maloy

III

" ~

c

q

C

400000-

300000-

200000-

100000 -

o-

P-ser (a)

6 0 ( 0 ) 0 -

5(XX~ -

4OX)(K)

3(X)(K)O

21XI(K)O

100000

0

P-thr (b) P-tyr (c)

75000 -

50030-

25000- ~ 1

P-ser (I)

P-ser + pro (II)

[~! P-ser + Pase (III)

[~ P-thr (I)

P-thr + pro (II)

P-thr + Pase (III)

D P-tyr (I)

P-tyr + pro (II)

E! P-tyr + Pase (III)

Figure 2. Phosphoamino acid analysis of PutA protein reveals phosphoserine, phosphothreonine, and phosphotyrosine, but only phosphoserine, and phosphothreonine are sensitive to phosphatase and proline. (Top} The phosphoamino acids are phosphoserine (a), phosphothreonine (b), and phosphotyrosine (c) in the following conditions: (I1 Untreated PutA protein, (II) PutA protein in 100 mM proline, (III) PutA protein treated with calf thymus alkaline phosphatase. (Bottom)Quantitation of each phosphoamino acid (a, b, and c) label in conditions I, II, and III, as described above.

ticular to those that do not belong to the vertebrate members of the Src subfamily (Hunter and Cooper 1985). For example, the TRK protein surfaced during GenBank searches for proteins similar to the carboxyl terminus of PutA protein. The TRK protein is an oncogene product with tyrosine kinase activity, which is expressed in some human tumor cells (Martin-Zanca et al. 1986). The presence of these motifs and similarities reinforces the experimental data, which indicate that PutA protein can carry out the transfer of phosphate from ATP to amino acid residues on a protein. Not all protein kinases auto- phosphorylate, but many of those that share similarity with PutA protein do. Although some proteins are ade- nylated on tyrosine residues, despite careful searches (Cortay et al. 1986; Foster et al. 1989)no other proteins that contain phosphotyrosine have yet been identified in Salmonella typhimurium or Escherichia coli.

Proline stimulates PutA dephosphorylation

The phosphorylation state of PutA protein changes in response to the physiological signals required for proline utilization. Addition of proline resulted in the rapid loss of ---60% of the total 3Zp-radioactive label from PutA protein in vitro (Fig. 1A, C; Table 1). This proline-stimulated dephosphorylation coincided with an increase in the inorganic phosphate peak in 31p-NMR experiments, but PutA protein was rapidly rephosphorylated in the presence of excess ATP (Fig. 3B). Two-dimensional phos-

phoamino acid analysis of PutA protein treated with proline or phosphatase revealed that both treatments resulted in dephosphorylation of the phosphoserine and phosphothreonine residues but not the phosphotyrosine residues (Fig. 2). The proline-induced dephosphorylation was prevented by addition of membranes isolated from an E. coli strain deleted for the put genes (Fig. 1 C; Table 1).

Dephosphorylation increases the affinity of PutA protein for DNA

To determine whether the phosphorylation of PutA protein has a physiological role, we tested the effects of phosphorylation on each of its known functions. Phos- phorylation did not significantly affect the enzymatic activities of PutA protein (data not shown). However, phosphorylation did affect the apparent K d for DNA binding as measured by gel mobility retardation experiments (Fried and Crothers 1981; Carey 1991). Preincu- bation of PutA protein with ATP, which increases PutA phosphorylation in vitro, decreased the apparent affinity of PutA for DNA about eightfold {Fig. 4C). In addition, preincubation of PutA protein with proline, which induces put gene expression in vivo, decreased the apparent affinity of PutA for DNA about twofold (Fig. 4C). Addition of both proline and ATP further decreased the apparent affinity of PutA for DNA (Fig. 4C). Although proline stimulates dephosphorylation of PutA protein, in the presence of excess ATP, it is rapidly rephosphory-

2036 GENES & DEVELOPMENT




Thr, Set, and Tyt phosphorylation in bacteria

B

o e0

o

J

r - r . -

o o o

' i " ' ' I ' ' ' i ' ' '- 1 ' ' ' I ' ' ~ l ~ " l ' ' ' I ' ' '- I ' '

6 4 2 0 - 2 - 4 - 6 - 8 - 1 0 P P M

l i "

Q a

Q O

a o

f o

T ~ I ' ' ' 1 ' ' ' 1 ' ' ~ i ' ' ' 1 ' ~ ' I ' ' ' I ' ' ' 1 ' ' t l ' ~

6 4 2 0 - 2 - 4 - 6 - 8 - 1 0 PPM

Figure 3. 3ap-NMR spectra of PutA protein confirms the presence of phosphoserine and phosphothreonine. (A) Spectra in the absence of proline. The assigned peaks correspond to phosphoserine {+ 4.66 ppml, phosphothreonine { + 4.18 ppm}, inorganic phosphate l + 2.9 ppml, the -c-phosphate of ATP {-5.35 ppm) and the a-phosphate of ATP I-9.43 ppm3 {Burt 1987). (B) Same as A but with 100 mM proline. The chemical shifts are the same as in A but the integrated area of the inorganic phosphate peak has increased by 68%.





Ostrovsky and Maloy

A B C

0.5[

0.4

0.3 <

0.2

0.1

0.5

0.4

~ 0.3 <

0.2

0.1

0 0 0 + PutA +PutA - PutA P-AA + PutA BSA P-AA + Pro - Pro + Pro - Pro +pro -pro -Pro +ATP +ATP +ATP -ATP -ATP

p..

O

< 4 z c~

M 2

Figure 4. PutA protein from crude cell extracts is recognized by a polyclonal anti-phosphotyrosine antibody. (A} Immunoassays of crude extracts of S. typhimurium cells overexpressing PutA protein (+ PutA) grown with and without proline (+ Pro and -Pro, respectively), crude extracts of S. typhimurium cells that do not express PutA protein ( - PutA), or pure phosphoamino acids (P-AA). (B) Immunoassays of purified PutA protein, bovine serum albumin with ATP (BSA + ATP}, or pure phosphoamino acids (P-AA). (C) KdDNA of purified PutA protein under different conditions as indicated.

la ted (Fig. 3B). Thus, these resul ts suggest tha t in addi- t ion to s t imu la t i ng dephosphory la t ion of PutA protein, prol ine also has an independen t effect on the P u t A - D N A in te rac t ion as well.

D i s c u s s i o n

These resul ts indicate tha t the S. t y p h i m u r i u m PutA prote in is cova len t ly phosphory la ted both in vitro and in vivo. PutA prote in is phosphory la ted on serine, threonine, and tyros ine residues, a phosphory la t ion pa t te rn c o m m o n to dua l - func t ion kinases in eukaryotes (Nishida and G o t o h 1993) but not previous ly observed in bacteria.

Does phosphory la t ion of PutA prote in play an impor- tant physiological role? Expression of PutA protein is necessary for cells to use prol ine as a carbon, nitrogen, or energy source. However , l ike m a n y other m a n y mem-

A B

I 2 3 4 5 6 7 I 2 3 4 5 6 7

....... ~:~ ~ g

Figure 5. PutA protein from crude cell extracts is recognized by monoclonal phosphothreonine and phosphotyrosine antibodies in Western immunoblots. Lanes 1 and 4 contain extracts of S. typhimurium cells that do not express PutA protein (MST2489). Lanes 2, 3, 5, and 6 contain extracts of S. typhimurium cells that express PutA protein under the control of the Ptac promoter (MST2830}; in lanes 2 and 5, the cells were not induced, but in lanes 3 and 6, the cells were induced with 0.1 mM IPTG. Lane 7 contains 4 txg of purified PutA protein. The samples in lanes 1-3 were sonicated after induction and ATP was added to 10 mM; the samples in lanes 4-6 were directly solubilized in SDS sample buffer, and no ATP was added. (A) Blot probed with anti-phosphothreonine monoclonal antibody, (B} Blot probed with anti-phosphotyrosine monclonal antibody.

brane proteins, overexpress ion of PutA pro te in is toxic. Thus, ma in t a in ing the op t imal express ion of the put genes requires a way of responding to both the availability of prol ine and the avai labi l i ty of func t iona l m e m - brane sites. Previous s tudies have d e m o n s t r a t e d tha t re- duc t ion of the PutA prote in during prol ine ca tabo l i sm increases its hydrophobic i ty , favoring the d issocia t ion of PutA prote in from the D N A and its par t i t ion ing into the m e m b r a n e (Ostrovsky de Spicer and Maloy 1993; Muro- Pastor and Maloy 1995; A.M. Muro-Pastor , P.C. Ostrov- sky, and S. Maloy, in prep.). These resul ts explain how proline induces the expression of the put genes, but not how toxic overexpress ion of PutA prote in is avoided when prol ine is present and the func t iona l m e m b r a n e sites are saturated.

Phosphory la t ion may serve as a m e c h a n i s m to avoid the toxic overexpress ion of PutA prote in in the absence of funct ional m e m b r a n e sites. In the absence of m e m - brane-binding sites, prol ine s t imu la t e s dephosphoryla- t ion of PutA protein, increasing the affinity of PutA protein for D N A and repressing the put operon. In contrast , when both membrane -b ind ing sites and prol ine are available, PutA prote in remains phosphoryla ted , decreasing the affinity of PutA prote in for D N A and derepress ing the put operon.

Table 1. Effect of proline and membranes on PutA autophosphorylation

Membranes a Proline ~ Label intensity b Percent label

+ - 790 x 10 "~ 100.0 + + 5 1 0 X 103 64.6 - + 317 x 103 40.1

aThe conditions indicated here correspond to lanes 1, 2, and 3 of Fig. 1C. bThe label intensity was quantitated by use of a PhosphorIm- ager.





Thr, Set, and Tyr phosphorylation in bacteria

How does the availability of proline and membrane sites control PutA phosphorylation.~ Proline increases the relative hydrophobicity of PutA protein {Ostrovsky de Spicer and Maloy 1993}, thus promoting either membrane binding or dimerization of PutA protein [possibly via interactions between hydrophobic surfaces on the monomers {Menzel and Roth 1981b}]. Dimerization plays a critical role in controlling the phosphorylation of many other proteins (Ninfa et al. 19931 Resh 1993). By analogy, proline-induced dephosphorylation may occur between monomeric subunits of PutA dimers. An experimental observation that supports this idea is that although both dimers and monomers of PutA protein can be observed on native polyacrylamide gels, only the monomeric form of PutA protein seems to be labeled by [a2P]ATP {data not shown}.

A model that accounts for these results is shown in Figure 6. The proline-dependent increase in hydrophobicity would direct PutA protein toward two altemative routes: dimerization or association with the membrane. When no membrane sites are available, PutA protein would dimerize in the cytoplasm, allowing intermolecular dephosphorylation, and thus resulting in an increased DNA affinity and repression of the put operon. When membrane sites are available, PutA protein would associate with the membrane allowing the functional interaction with the electron transport chain required for enzyme activity and preventing dephosphorylation, thus resulting in a lower DNA affinity and derepression of the put operon. Hence, phosphorylation would allow PutA protein to rapidly respond to the presence or absence of both membrane sites and proline.

This model can account for the modulation of PutA expression under a spectrum of physiological conditions. When proline and membrane sites are available, PutA protein would remain phosphorylated so PutA-DNA binding would be weak, resulting in the maximal expression of the put genes. In the absence of one of these components, PutA protein would have intermediate DNA binding affinity resulting in intermediate levels of put gene expression. Finally, when neither proline or membrane sites are available, PutA protein would be fully dephosphorylated so PutA would have the highest DNA binding affinity, resulting in minimal levels of put gene expression. These changes in PutA protein, caused by proline, phosphorylation, and the interaction of PutA protein with the membrane, would result in a gradient of DNA binding affinities that would optimize the level of put gene expression depending on the physiological conditions. Because the availability of membrane sites may depend on the intracellular concentration of other dehy- drogenases or other aspects of membrane physiology, the phosphorylation state of PutA protein could provide a means for the cell to monitor the availability of functional membrane sites and to modulate put gene expression in response to the availability of these sites.

This model accounts for most of the available data, but many critical questions about the phosphorylation of PutA protein still need to be answered. We do not yet know which particular serine, threonine, and tyrosine

~ CM

Prollne M e m b r a n e s i t e s

+ Prollne ~ - M e m b r a n e s i t e s

DNA

Figure 6. A model for how phosphorylation may modulate the repression of the put operon by PutA protein in response to the availability of membrane sites. The availability of proline, ATP, and membrane sites shifts the equilibrium between the phosphorylated and unphosphorylated forms of PutA protein resulting in the different DNA binding affinities {Fig. 4C). In the presence of proline, PutA protein becomes more hydrophobic and it either associates with the membrane or, if membrane sites are not available, it dimerizes. When the PutA monomers are associated with the membrane the protein functions as an enzyme. In addition, the membrane association prevents intermolecular dephosphorylation so PutA protein remains phosphorylated, resulting in a lower DNA-binding affinity. In contrast, when membrane sites are not available, PutA protein dimerizes, allowing intermolecular dephosphorylation, resulting in a higher DNA-binding affinity and thus stronger repression of the put genes. The shaded ovals represent PutA protein molecules {CM) cytoplasmic membrane; {P) phosphate incorporated on amino acid residues; {Pi) inorganic phosphate.

residues in PutA protein are phosphorylated. The analo- gous eukaryotic proteins are often phosphorylated on multiple serine, threonine, and tyrosine residues. Sys- tematic site-directed mutagenesis to change each of these amino acids in the phosphorylated protein is a common approach to elucidate which residues are phosphorylated and the physiological relevance of their phosphorylation. However, this approach is not feasible for PutA protein because it is quite large {144.2 kD} and contains 61 serine residues, 67 threonine residues, and 35 tyrosine residues.

In summary, phosphorylation seems to modulate the





Ostrovsky and Maloy

s w i t c h tha t d e t e r m i n e s w h e t h e r P u t A pro te in f rom S. typhimuriurn f u n c t i o n s as a m e m b r a n e - a s s o c i a t e d en- z y m e or as a D N A - b i n d i n g pro te in . P h o s p h o r y l a t i o n of P u t A p ro t e in on serine, t h reon ine , and ty ros ine res idues changes the D N A b ind ing aff ini ty of P u t A prote in , and c o n s e q u e n t l y f ine t unes the express ion of the put genes in r e sponse to changes in i nduce r c o n c e n t r a t i o n and ava i lab i l i ty of m e m b r a n e si tes in the cell. T h e pa t t e rn of p h o s p h o r y l a t i o n of this bac ter ia l p ro te in is s imi la r to cer ta in e u k a r y o t i c p ro t e in k inases . T h e s e resu l t s h in t t ha t p ro te in ty ros ine k inases m a y play i m p o r t a n t regu- l a to ry roles in p r o k a r y o t e s as they do in euka ryo te s .

Materia ls and m e t h o d s

Strains and growth conditions

The following strains were used: EM41, E. coli K-12 trp thi J(putPA)lO1; MST2830, S. typhimurium LT2 leu-414(Am) putA826::TnlO/pPC34 (pCKR101 ]acI q Ptac-putA ~ ), MST2489 S. typhimurium LT2 leu-ala(Am) putA826::TnlO/pPC34 (pCKR101 laclq), MST58 S. typhimurium LT2 J(putPA)523.

Difco nutrient broth (0.8%) containing NaC1 at a final concentration of 0.5% (wt/vol) was used for rich medium. The minimal medium used was E medium (Vogel and Bonner 1956! supplemented with 0.6% succinate. When specified, 0.2% proline was added to E medium. Sodium ampicillin (100 ~g/ml} was added to rich medium for strains carrying plasmids.

Purification of PutA protein

PutA protein was purified from strain MST2830 (pCKR 101 Ptac-- putA) following induction with 0.1 mM isopropyl-~3-D-thiogalac- toside (IPTG)as described previously (Menzel and Roth 1981b; Ostrovsky de Spicer and Maloy 1991). The final purified samples contained between 1.6 and 4 mg of PutA protein per mil- liliter. Purified samples were stored frozen at -70°C until use. Electrophoresis in SDS-polyacrylamide gels was carried out as described with discontinuous gels composed of a 4% stacking gel and an 8% separating gel. Proteins in SDS or native gels were stained with FastStain (ZoionResearch, Allston, MA/.

Phosphorylation with [3gP]-ATP

In vitro phosphorylation of PutA protein was carried out as follows: 2.5 ~g of purified PutA protein was incubated with 2.5 mM {y-32p]ATP in 1 × buffer A(12 mM Tris, 1 mM MgCI,, 1 mM CaCI> 5 mM NaC1, 0.1 mM dithiothreitol, 100 ~g/ml bovine serum albumin, and 5% glycerol at pH 8.0 at 21°C)for 5 min at 37°C. For the experiments shown in Figure 1, PutA protein was phosphorylated as described above, incubated at room temper- ature for 5 min under the indicated conditions, then the samples were analyzed by polyacrylamide gel electrophoresis and autoradiography.

Membrane preparations from S. typhimurium have a high background of nuclease activities, which are absent from membrane preparations from E. coil (A.M. Muro-Pastor and S. Maloy, unpubl.). Therefore, the membranes used were purified from strain EM41 (E. coli Aput) grown in minimal medium with succinate and proline (Muro-Pastor and Maloy 1995).

Phosphoamino acid analysis

Purified, labeled PutA protein (100 ~g ) tha t was untreated, treated with 5 units of calf thymus alkaline phosphatase, or treated with 100 mM proline, precipitated in 20% TCA with 10

mg of bovine serum albumin by incubation on ice for 1 hr followed by centrifugation for 20 min, and air-dried. The dried pellet was resuspended in boiling 6 N HC1 then incubated at 110°C for 2 hr. The hydrolysate was resuspended in pH 1.9 buffer containing 1 mg/ml free phosphothreonine, phosphoserine, and phosphotyrosine (Sigma)then analyzed on a thin- layer cellulose plate by electrophoresis at pH 1.9 in the first dimension and pH 3.5 in the second dimension (Boyle et al. 1991 }. The position of the phosphoamino acids was determined by ninhydrin staining followed by autoradiography. The amount of label in each phosphoamino acid was quantitated with a Phosphorlmager.

.31P-NMR analysis

The sample prepared for 3~P-NMR analysis contained 0.2 mM PutA protein, 2 mM ATP, 30% D20 in 1 x buffer A (see above). The acquisition parameters were set as follows: 39 ° pulse (15 msec), spectral width _ 3048.78 Hz, data size 32,768, DW 164 msec, pulse width 15 msec, delay time 500 msec, 10 Hz line broadening, 5000 scans for Figure 3A and 3388 scans for Figure 3B.

Enzyme-linked immunoassays

Enzyme linked immunoassays were done essentially as described by Harlow and Lane (1988). Crude extracts of MST2830 (JputA/pCKRlO1 Pt~c-putA), induced with 0.1 mM IPTG and grown in minimal medium with or without proline, and MST58 (JputA), or samples of purified PutA protein were used for the immunodetection assays shown in Figure 4. Positive controls contained free phosphoserine, phosphothreonine, or phosphotyrosine, and negative controls contained bovine serum albumin. Samples with crude cell extracts contained 500 ~g of total protein per assay. Samples with purified PutA contained 10 ~,g of protein per assay. Samples with free phosphoserine, phosphothreonine, or phosphotyrosine (Sigma) contained 1 ~g of each phosphoamino acid per assay. Control samples with bovine serum albumin (Sigma} contained 20 ~zg of protein per assay. Each assay was done with 5 ~g/ml anti-phosphotyrosine polyclonal antibody (UBI, Lake Placid, NY).

We.stern blot immunodetection

Crude extracts of strain MST2830 (pCKR101 Ptac-putA)induced with 0.1 mM IPTG or uninduced, strain MST2489 (pCKR 101 ) treated in a similar manner, or purified PutA protein were separated by SDS-polyacrylamide gel electrophoresis. The proteins were then electrophoretically transferred to nitrocellu- lose membranes. The transfer was followed by two washes with phosphate-buffered saline (PBS, 137 mM NaC1, 2.7 mM KC1, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, at pH 7.3)plus 0.05% Tween 20. The washed membranes were blocked by soaking in PBS plus 5% nonfat dry milk for 1 hr, washed 3 times with PBS plus 0.05% Tween 20, then incubated for 3 hr in PBS plus 5% nonfat dry milk with anti-phosphoserine (clone PSR-45), anti-phosphotyrosine {clone PT-66/, or anti-phosphothreonine (clone PTR-8) monoclonal antibodies (Sigma)at the dilution suggested by the manufacturer. After incubation with the primary antibody, the membranes were washed 4 times with PBS plus 0.05% Tween 20, and incubated in in PBS plus 5% nonfat dry milk with 0.2 ~g/ml of horseradish peroxidase (HRP)-conjugated goat anti- mouse IgG (GIBCO; BRL} for 1 hr. After incubation with the secondary antibody, the membranes were washed 4 times with PBS plus 0.05% Tween 20. The membranes were then incubated in the chemiluminescent ECL reagent mix (Amersham) for 1 min, covered with plastic, and exposed to X-ray film for 1 min.





Tht, Set, and Tyt phosphorylation in bacteria

Gel mobility retardation assays

PutA-DNA binding was quantitated by use of a gel mobility retardation assay with 32p-labeled DNA from the put control region. The DNA binding reactions were carried out in 1 x buffer A {see above} as previously described {Ostrovsky de Spicer et al. 1991). The free DNA was quantitated using a Phospho- rImager.

A c k n o w l e d g m e n t s

We thank Benita Katzenellenbogen for use of the Hunter thin layer plate electrophoresis system and Dan Celander for use of the PhosphorImager. The membrane preparations were gener- ously provided by Alicia Muro-Pastor. The 3tp-NMR data were collected at the Molecular Spectroscopy Laboratory, School of Chemical Sciences, University of Illinois at Urbana-Champaign by Dr. Feng Lin. We thank David Nunn for his valuable help with the Western blots. We benefited from critical comments on the manuscript by John Cronan, Jr., Alan Horwitz, Kelly Hughes, Tony Hunter, Sandy Parkinson, John Roth, Tom Sil- havy, Jeremy Thorner, and Barry Wanner. This work was sup- ported by the National Institute of General Medical Sciences Grant GM34715 to S.M.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

References

Abrahamson, J.L.A., L.G. Baker, J.T. Stephenson, and J.M. Wood. 1983. Proline dehydrogenase from Escherichia coli K12: Properties of the membrane-associated enzyme. Eur. ]. Biochem. 134: 77-82.

Boyle, W. J., P. van der Geer, and T. Hunter. 1991. Phosphopep- tide mapping and phosphoamino acid analysis by two-dimensional separation on thin-layer cellulose plates. Meth- ods Enzymol. 201: 110--149.

Brandriss, M.C. and B. Magasanik. 1979. Genetics and physiology of proline utilization in Sacharomyces cerevisiae: En- zyme induction by proline. I. Bacteriol. 140: 498-503.

Burt, C.T. 1987 Phosphorus NMR in biology. CRC Press, Boca Raton, FL.

Buss, J.E., and J.T. Stull. 1983. Measurement of chemical phosphate in proteins. Methods Enzymol. 99: 7-14.

Carey, J. 1991. Gel retardation. Methods Enzymol. 208: 103- 117.

Chen, L.M. and S.R. Maloy. 1991. Regulation of proline utilization in enteric bacteria: Cloning and characterization of the Klebsiella put control region. ]. Bacteriol. 173: 783--790.

Cortay, J.C., B. Duclos, and A.J. Cozzone. 1986. Phosphoryla- tion of an Escherichia coli protein at tyrosine. I. Mol. Biol. 187: 305-308.

Costilow, R.N. and D. Cooper. 1978. Identity of proline dehydrogenase and pyrroline-5-carboxylic acid reductase in Clostridium sporogenes. J. Bacteriol. 134: 139-146.

Cozzone, A.J. 1993. ATP-dependent protein kinases in bacteria. ]. Cell. Biochem. 51: 7-13.

Downing, S.J., J.M. Phang, E.M. Kowaloff, D. Valle and R.J. Smith. 1976. Proline oxidase in cultured mammalian cells. ]. Cell Physiol. 91: 369-376.

Edelman, A.M., D.K. Blumenthal, and E.G. Krebs. 1987. Protein serine/threonine kinases. Annu. Rev. Biochem. 56: 567-613.

Foster, R., J. Thorner, and G.S. Martin. 1989. Nucleotidylation, not phosphorylation, is the major source of phosphotyrosine detected in bacteria. J. Bacteriol. 171: 272-279.

Fried, M. and D. Crothers. 1981. Equilibria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9: 6505-6525.

Hanks, S.K., A.M. Quinn, and T. Hunter. 1988. The protein kinase family: Conserved features and deduced phylogeny of the catalytic domains. Science 241: 42-52.

Harlow, E. and D. Lane. 1988. Antibodies: A laboratory man- ual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.

Hunter, T. and I.A. Cooper. 1985. Protein-tyrosine kinases. Annu. Rev. Biochem. 54: 897-930.

Maloy, S. 1987. The proline utilization operon. In Escherichia coli and Salmonella typhimurium: Cellular and molecular biology {ed. Neidhardt et al.). American Society for Micro- biology, Washington, D.C.

Maloy, S. and J. Roth. 1983. Regulation of proline utilization in Salmonella typhimurium: characterization of put::Mud(A- p,lac) operon fusions. I. Bacteriol. 154: 561-568.

Martin-Zanca, D., S.H. Hughes, and M. Barbacid. 1986. A human oncogene formed by the fusion of truncated tropomyo- sin and protein tyrosine kinase sequences. Nature 319: 743- 748.

Menzel, R. and J. Roth. 1981a. Enzymatic properties of the purified PutA protein from Salmonella typhimurium. ]. Biol. Chem. 256" 9762-9766.

1981b. Purification of the putA gene product. I. Biol. Chem. 256: 9755-9761.

Muro-Pastor, A.M. and S.R. Maloy. 1995. Proline dehydrogenase activity of the transcriptional repressor PutA is required for induction of the put operon by proline. ]. Biol. Chem. 270: 9819-9827.

Ninfa, E.G., M.R. Atkinson, E.S. Kamberov, and A.J. Ninfa. 1993. Mechanism of autophosphorylation of Escherichia coli nitrogen regulator II (NRn or NtrB): trans-phosphorylation between subunits. J. Bacteriol. 175: 7024-7032.

Nishida, E. and Y. Gotoh. 1993. The MAP kinase cascade is essential for diverse signal transduction pathways. Trends Biochem. Sci. 18: 128-131.

Ostrovsky de Spicer, P. and S. Maloy. 1993. PutA protein, a membrane-associated flavin dehydrogenase, acts as a redox dependent transcriptional regulator. Proc. Natl. Acad. Sci. 90: 4295--4298.

Ostrovsky de Spicer, P., K. O'Brien, and S. Maloy. 1991. Regu- lation of proline utilization in Salmonella typhimurium: A membrane-associated dehydrogenase binds DNA in vitro. J. Bacteriol. 173:211-219.

Sacktor, B. and C.C. Childress. 1967. Metabolism of proline in insect flight muscle and its significance in stimulating the oxidation of pyruvate. Arch. Biochem. Biophys. 120: 583- 588.

Stewart, C.R. and E.Y. Lai. 1974. Pyrroline-5-carboxylic acid dehydrogenase in mitochondrial preparations from plant seedlings. Plant Sci. Lett. 3" 173-181.

Sylvester, D., S.M. Krassner, and K.B. Munson. 1974. Proline metabolism in Trypanosoma scelopori epimastigotes. Comp. Biochem. Physiol. 48B: 401--409.

Vogel, H.J. and D.M. Bonnet. 1956 Acetyl ornithinase of Esch- erichia coli: Partial purification and some properties. ]. Biol. Chem. 218: 97-106.

Wood, J. 1987. Membrane association of proline dehydrogenase in Escherichia coli is redox dependent. Proc. Natl. Acad. Sci. 84: 373-377.

Wood, J.M., K.A.C.C. Taylor, D.J. McClellan, G.G. Lawrie, R.L. Krogsrud, and T.J. Beveridge. 1987. Isolation and characterization of monoclonal antibodies to proline dehydrogenase from Escherichia coli K-12. Biochem. Cell Biol. 65: 507-513.





10.1101/gad.9.16.2034Access the most recent version at doi: 9:1995, Genes Dev.

P C Ostrovsky and S Maloy regulation in Salmonella typhimurium.modulates membrane-protein interactions and transcriptional Protein phosphorylation on serine, threonine, and tyrosine residues

References

http://genesdev.cshlp.org/content/9/16/2034.full.html#ref-list-1

This article cites 30 articles, 14 of which can be accessed free at:

License

ServiceEmail Alerting

click here.right corner of the article or

Receive free email alerts when new articles cite this article - sign up in the box at the top

Copyright © Cold Spring Harbor Laboratory Press


http://genesdev.cshlp.org/lookup/doi/10.1101/gad.9.16.2034

http://genesdev.cshlp.org/content/9/16/2034.full.html#ref-list-1

http://genesdev.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=protocols;10.1101/gad.9.16.2034&return_type=article&return_url=http://genesdev.cshlp.org/content/10.1101/gad.9.16.2034.full.pdf



protein phosphorylation on serine, threonine, and tyrosine...

Documents