sigma factor is not released during transcription in bacillus subtilis

Molec. gen. Genet. 174, 47 52 (1979) © by Springer-Verlag 1979

Sigma Factor is Not Released During Transcription in Bacillus subtilis

Valerie M. Williamson and Roy H. Doi Department of Biochemistry and Biophysics, University of California, Davis, California 95616, USA

Summary. The relationship between sigma (a) and delta (6) factors of Bacillus subtilis R N A polymerase has been analyzed during initiation of R N A synthesis. When core enzyme (E) containing delta factor (E6) binds to DNA, the delta factor is released with the formation of an E-DNA complex. The addition of sigma to the E-DNA complex results in the formation of a stable Eo--DNA complex which can synthesize R N A upon addition of nucleoside triphosphates. Sigma factor, significantly, is not released from the core during R N A synthesis. These results suggest that delta and sigma factors can act sequentially during initiation of R N A synthesis with delta acting as a D N A recognition factor and sigma acting as an initiation factor. The results do not preclude the possibility that Ea can initiate R N A synthesis correctly since Eo- alone can bind to D N A and initiate R N A synthesis.

in two size classes of approximately 90,000 and 55,000 daltons (Burgess, 1976). We report here that in Ba- cillus subtilis, which contains a 55,000 dalton a (Lo- sick et al., 1970; Avila et al., 1971), a factor is not released during the transcription process. In B. subtilis another subunit, 6 (Williamson and Doi, 1978; Pero et al., 1975; Tjian et al., 1977; Spiegelman et al., 1978), is involved in the initiation of transcription (Williamson and Doi, 1978). This subunit can displace o- from the R N A polymerase core (E) and can bind to E to form an E6 complex (Williamson and Doi, 1978), which binds specifically to DNA. However, after binding of E6 to DNA, 6 is released from the E and then a is able to bind to the E-DNA complex to initiate transcription. Thus in B. subtilis two factors act in a sequential manner to initiate transcription. The B. subtilis R N A polymerase forms will be desig- nated as follows: flfl'(X2(r will be called Ea instead of holoenzyme, and fl[3'ea6 will be called E6.

Introduction

The release of a factor from R N A polymerase core (E) during transcription, which was first demon- strated in Escherichia coli by Travers and Burgess (1969) and in Azotobacter vineIandii by Krakow (1969), has been generally accepted as occurring in all bacterial species. Early studies with non-denaturing gel electrophoresis methods indicated that a was released from Azotobacter (Krakow and von der Helm, 1970) and E. coli (Ruet et al., 1970) R N A polymerase after transcription had been initiated. Similar results were also obtained by density gradient centrifugation techniques for Pseudomonas putida (Gerard et al., 1972) and E. coli (Bordier and Rossetti, 1976).

Comparative studies of R N A polymerase from various prokaryotes indicate that the ~ factor occurs

For offprints contact: Dr. Roy H. Doi

Materials and Methods

Bacterial Strain. Bacillus subtilis 168 wild type was used in all the studies described here.

RNA Polymerase, a and 6 Purification. RNA polymerase species and cr and 6 factors were purified as previously described (William- son and Doi, 1978).

Non-Denaturing Polyacrylamide Gel Electrophoresis. The electrophoresis system described below is adapted from that of Hedrick and Smith (1968). The application buffer was similar to that used by Krakow (1971) in studies of a release from Azotobacter vinelandii RNA polymerase. Addition of glycerol was necessary to stabilize c~ binding to E. The discontinuous acrylamide concentrations (8% and 12%) were used to best display both the large (approximately 400,000 dalton) RNA polymerase and the small (21,000 daltons) a factor. The lower gel contained 60 mM Tris-Hcl, pH 7.9, 12% acrylamide, 0.32% bisacrylamide, 15% glycerol, 0.035% TEMED, and 0.023% ammonium persulfate. Upper gel composition was the same as the lower gel composition except that it contained 8% acrylamide and 0.21% bisacrylamide. The reservoir buffer con-

0026-8925/79/0174/0047/$01.20

48 V.M. Williamson and R.H. Doi: Sigma Factor is Not Released

tained 0.12 M asparagine titrated to pH 7.3 with Tris-base. The samples were incubated at 37°C for 3 rain in 60 mM Tris-HC1, pH 7.9; 40 mM mercaptoethanol; 5% sucrose; 15% glycerol; and 90 mM KC1, or as described in the figures. After incubation the mixtures were cooled on ice and electrophoresed. Electrophoresis was conducted at 100 V for 3-4 h.

Gels were stained overnight in 50% methanol, 7.5% acetic acid, and 0.5% Coomassie brilliant blue R-250 and destained in 45% methanol and 9% acetic acid, then further destained in 15% methanol and 22.5 acetic acid.

Glycerol Gradient Studies. Glycerol gradient centrifugation was used to study the subunit composition of RNA polymerase under various conditions. Approximately 0.3 mg of the RNA polymerase species to be tested was combined with appropriate amount of factor (either a or 6) as indicated and the volume was brought to 0.5 ml with Buffer G (50 mM Tris-HC1, pH 7.9, 5 mM MgC12, 0.4 mM dithiothreitol, 15% glycerol, and 0.12 M KC1). The mixture was dialyzed overnight versus 500 ml of Buffer G. Linear gradients of 20% to 40% glycerol (3.7 ml total volume) were pre- pared in 7/16 x 2 3/8 inch polyallomer tubes. The dialyzed sample was layered on the gradient and centifuged at 4 ° C for 26 h at 157,000 x g or as indicated in a Beckman type 56 or type 60 swing- ing bucket rotor. Gradients were collected by piercing a hole in the bottom of the tube. 0.5 ml fractions were collected for analysis.

Proteb~ Determinations. Protein concentration was determined by the method of Sedmak and Grossberg (1977).

Results

When 3 factor is added to Err in the absence of DNA, 6 displaces rr completely from Err (Williamson and Doi, 1978). However, a cannot displace 6 from E~. This was shown earlier in a series of glycerol gradient centrifugation experiments (Williamson and Doi, 1978). A non-denaturing polyacrylamide gel electrophoresis system was developed to screen a number of conditions for 6 and rr release, since comparatively little enzyme is required for each analysis by this method. When Ea is run by itself in this system, no free rr is observed (Fig. 1, lane 1). When an equimolar amount of 6 is added, rr is released and 6 is retained (Fig. 1, lane 2) confirming the results reported earlier with the glycerol gradient technique. When poly dAT is mixed with the E6, 6 is released from E and is seen as a discrete band in the gels (Fig. 1, lane 6). The 6 is released in the presence of template regardless of whether rr is present or not. However, since the amount of 6 could not be mea- sures accurately in these gels, it is not known whether 6 was released completely.

When nucleoside triphosphates were added to the reaction mixture containing poly dAT, Err, and 6, and the mixture was incubated for 3 min under conditions of RNA synthesis, ~5 was released, but no release of a from E was observed (Fig. 2, lane 6). Under the conditions of the reaction mixture, the specific activity of the enzyme was approximately 71 nmoles of UMP incorporated per mg of protein per 10min

Fig. 1. Factor release as determined by non-denaturing polyacrylamide gel elctrophoresis. The following mixtures were incubated at 37°C for 3 min, then cooled on ice and electrophoresed in a non-denaturing gel system of discontinuous acrylamide concentration (8% and 12%): 1) 24gg of Eo; 2) 24/ag of E a + l . 4 p g of 6; 3) 27pg of E6; 4) 24gg of E ~ + 5 p g of poly dAT; 5) 24 pg of Ecr + 1.4 gg of 6 + 5 gg of poly dAT; 6) 27 pg of E6 + 5 pg of poly dAT; 7) 27 gg of E6+4.6 gg of or+5 gg of poly dAT

and the presence of ~ factor did not inhibit the transcription rate. As a control E. coli RNA polymerase holoenzyme was tested under the same conditions and a release was found to occur (Fig. 2, lane 8). The enzyme activity for both bacterial species was approximately the same. These results explain prev- ious data (Halling and Doi, 1978) which showed that B. subtilis RNA polymerase eluted as two peaks containing E6 and Err, respectively, with no free E. E was always found associated with either c5 or or.

To further substantiate the results obtained by the non-denaturing gel system, the enzyme was sub- jected to glycerol gradient centrifugation after similar treatments. In the first experiment, 0.6 mg of RNA polymerase plus one fold excess rr and 0.4 mg of poly dAT were dialyzed for 24 h at 4°C versus Buffer G. After dialysis, 0.1 mmoles of ATP, CTP, GTP, and UTP were added to half of the RNA polymerase. The RNA polymerase mixtures were incubated for 3 min at 37 ° C, then chilled, loaded onto 2 0 4 0 % glycerol gradients, and centrifuged for 8 h at 257,000 x g. Protein concentration of rr/~2 ratio for each fraction is given in Table 1. The ~ content of the enzyme in the peak tubes after glycerol gradient centrifugation was 0.68 whether or not transcription

V.M. Will iamson and R.H. Doi: Sigma Factor is Not Released

Fig. 2. Factor release from B. subtilis and E. coli R N A polymerases during transcription as determined by non-denatur ing polyacrylamide gel electrophoresis. The following mixtures were incubated at 37°C for 3 m i n : I) 24~tg of Ecr; 2) 2 4 p g of Eer+l ,4 pg of 3; 3) 2 4 p g of E a + 1 0 p g o f p o l y dAT; 4) 2 4 g g o f E a + l . 4 ~ t g of 3 + 10 pg of poly dAT ; 5) 24 lag of E¢7 + 10 pg of poly dAT + 10 nmoles of ATP, GTP, CTP, and UTP; 6) 24 pg of Ecr+l .4 gg of 3 + 1 0 lag of poly d A T + 1 0 nmoles of ATP, GTP, CTP, and UTP; 7) E. coli holenzyme (Ea), 21 lag; 8) E. coil Ea, 21 p g + 1 0 lag poly d A T + 10 nmoles of ATP, GTP, CTP, and UTP. Other conditions are as in Fig. 1 except that the final glycerol concentration was reduced from 15% to 9% and KCI was present at 0.15 M instead of 0.09 M as in Fig. I. a 8 designates the position of B. subtilis ~7, and ~ designates E. coli

had occurred. The a to E ratio was less than one. It is not certain if this is because a was not bound to all the E or because Coomassie blue dye binds less well to ¢ than to e. The low o- content of the bottom fractions is curious since it occurs both with

49

Table 1. Analysis of R N A polymerase after glycerol gradient centrifugation in the presence of nucleic acid and excess a a

Fraction b Protein (pg) a/c~2 ~

( - R N A c) ( + R N A d) ( - R N A ) ( + R N A )

(Bottom) 4 I4 12 5 14 10 6 45 34 7 32 41 8 11 13 (Top)

0.42 0.41 0.68 0.67 1.11 O.68

a Enyzme used was Ecr plus an equimolar amoun t of cr factor. Centrifugation was for 8 h at 257,000 x g b Fractions 1, 2 and 3 contained no detectable protein and, therefore, were not analyzed c ( - R N A ) indicates that D N A but no ribonucleotides were incubated with the R N A polymerase before contrifugation

( + R N A ) indicates that D N A and ribonucleotides were incubated with the R N A polymerase before centrifugatlon, that is, R N A transcription was allowed to occur e SDS-Urea polyacrylamide gels (Halling et al., 1977) were run on glycerol gradient fractions. The Coomassie blue stained gels were scanned at 600 nm. The ratio of the a factor to E is assumed to be equal to the cr/~ 2 ratio

and without RNA synthesis and the DNA (as determined by Az60,m) was present mainly in the enzyme peak tubes.

In a similar experiment, the centifugation patterns of l) Ea; 2) Ea plus 6 and poly dAT; and 3) Eo- plus b, poly dAT, and nucleoside triphosphates (that is, a complete transcription mixture) are compared in Table 2 and Fig. 3. The RNA polymerase bound to DNA centrifuged somewhat further than free enzyme, but there was considerable overlap. The RNA transcribed in 3) was mostly released from the enzyme-DNA complex. This could have occurred before or during centrifugation. The subunit composition

Table 2. Analysis of R N A polymerase after glycerol gradient centrifugation in the presence of nucleic acid and 3 o

Fraction b Protein (pg) o/e 2 6/~:2

( - R N A ~) ( + R N A d) ( - RNA) ( + RNA) ( - RNA) ( + RNA)

I(bottom) 25 20 0.27 0.23 0.27 0.20 2 44 28 0.44 0.34 0.46 0.29 3 42 37 0.54 0.54 0.52 0.44 4 28 24 0.64 0.47 0.57 0.54 E-before

centrifugation 0.76 0.16

° Enzyme used was E~ plus an equimolar amount of c~ factor. Centrifugation was for 26 h at 157,000 × g b Fractions 5 8 contained very httle protein, therefore it was not possible to determine 6/~2 or 3/~2 on these fractions c ( - RNA) indicates that D N A but no ribonucleotides were incubated with the R N A polymerase before centrifnga- tion e ( + RNA) indicates that D N A and ribonucleotides were incubated with the R N A polymerse before centrifugation, that is, R N A transcription was allowed to occur


611

50

4O

Z

o

O.

~. ao

Io

0 I 2 3 4 5 6 7 8 9 I0

BOTTOM FRACTIONS TOP

12

I0

8

io

e b x

4 o

2

0

Fig. 3. Glycerol gradient analysis of transcription mixtures. The following mixtures (0.4 ml final volume) were dialyzed overnight against dialysis Buffer G: 1) 0.3 mg of Ecr ( . ) ; 0.3 mg Ecr+14 gg 3+0.2rag poly dAT (o); 3) 0.3mg of E a + 1 4 g g of ~+0 .2mg of poly dAT (zx). After dialysis 0.1 mmoles of ATP, CTP, GTP, and UTP (aH-UTP), specific activity=2.8 × 10' cpm per nmole) were added to mixture 3) and all mixtures were incubated at 37 ° C for 3 min. The reaction mixtures were then chilled, loaded onto 20-40% glycerol gradients and centriguged 24 h at 157,000xg. Protein concentration of each fraction was determined by the method of Sedmak and Grossberg (1977). The RNA synthesized (A) in reaction mixture 3) was determined by trichloroacetic acid precipitation of a 50 gl aliquot of each fraction from the glycerol gradient

of the RNA polymerase in the glycerol gradient fractions was analyzed by SDS-urea polyacrylamide gel electrophoresis (Fig. 4). Again the DNA-bound RNA polymerase subunit composition appeared to be the same whether or not nucleoside triphosphates were present (i.e., whether or not RNA synthesis had occurred) (see Table 2). Fractions 2 and 3 (Table 2) contained approximately the same proportion of G with or without RNA synthesis. The amount of b associated with RNA polymerase was significant, but somewhat lower in the case where transcription was occurring (Table 2). Some free 8, but no free ~ is seen near the top of the gradient. It may be that some b was bound because o- was not saturating the enzyme applied or because fi had cycled back during the course of the experiment. Another possibility is that an intermediate state exists in which o- and can bind at the same time. These results are quite different from those seen when no DNA was present (Williamson and Doi, 1978) and show that the presence of DNA (poly dAT) prevents b from displacing o- from E and reduces the affinity of ~ for E.

Discussion

The evidence presented here indicates that o- is retained with the core during transcription. The non- denaturing gel experiments showed that while o- is released from E. coli enzyme, under the same conditions the smaller a is not released from B. subtilis

Fig. 4A and B. Analysis of glycerol gradient fractions by 10% polyacrylamide SDS-urea gels. Glycerol gradient fractions from Fig. 3 were analyzed by gel electrophoresis under denaturing conditions (Halling et al., 1977). A Gradient 2 of Fig. 3, which contains RNA polymerase and DNA; B Gradient 3 of Fig. 3 which contains nucleoside triphosphates and newly synthesized RNA in addition to RNA polymerase and DNA. In both runs, #1 is the bottom fraction, #9 is the top fraction, and//10 is a control containing Ecr

V.M. Williamson and R.H. Doi: Sigma Factor is Not Released 51

DNA + RNA

y+

Fig. 5. Model for the role of b and a during transcription in B. subtilis. 1) 3 displaces cr from Ecr; 2) E6 bmds to promoter; 3) b falls off of E-DNA; 4) a brads to E-DNA to form an E~-DNA initiation complex; 5) RNA synthesis is initiated; 6) after elonga- tion and termination Ecr is released. Thus, ~ and a act sequentially and each has its own cycle

enzyme. As discussed in the In t roduc t ion , o- release has been descr ibed for several bac ter ia l species by a var ie ty of techniques ; all o f the species s tud ied conta in the large class a subunit . C h a m b e r l i n (1974) suggests tha t E. coli a release occurs when the enzyme is conver t ed f rom the in i t ia t ion to the e longa t ion con- fo rmat ion . Pe rhaps release o f the smal ler a is not requi red for a s sump t ion of the e longa t ion con fo rma- t ion in B. subtilis.

The m o d e l o f o- and 6 fac tor in te rac t ion with R N A po lymerase and t r ansc r ip t ion presen ted in an ear l ier pape r (Wi l l i amson and Doi , 1978) hypo- thesized the release of 6 f rom E after the Eb complex had b o u n d to D N A . Evidence p resen ted here indicates that , in fact, the b ind ing o f 6 to E is great ly reduced when E6 binds to D N A . It appea r s tha t 6, the recogni t ion factor , is re leased f rom the core p r io r to in i t ia t ion, and tha t a, the in i t ia t ion factor , does not have to be re leased dur ing the e longa t ion phase o f synthesis. The work ing m o d e l has been mod i f i ed to show the re ten t ion o f o- dur ing t r ansc r ip t ion and is i l lus t ra ted in Fig. 5. The da t a do no t prec lude the poss ib i l i ty tha t E a a lone can p r o p e r l y recognize s t rong p r o m o t e r s such as those present for " e a r l y " phage genes (Sp iege lman et al., 1978). This type of m o d e l suggests the poss ib i l i ty of o ther " 6 " fac tors ( F u k u d a and Doi , 1977) t ak ing pa r t in dif ferent ia l t r ansc r ip t ion dur ing spo ru l a t i on o f B. subtilis (Doi , 1977). F u r t h e r m o r e these da ta c o m p l e m e n t the prev- ious obse rva t ions ( F u k u d a , et al., 1977; Ha l l ing et al., 1977; Hal l ing et al., 1978) which indicate tha t the

B. subtilis t r ansc r ip t ion mach ine ry differs s ignif icant ly f rom tha t for E. coli in subuni t s t ructure and funct ion.

W e have no t iced tha t in the cases s tudied thus far, the larger a occurs in g ram negat ive o rgan isms and the smal ler o- occurs in g ram posi t ive bacter ia . F r o m our observa t ions we wou ld predic t a 6-like pep- t ide in o ther g r am posi t ive bacter ia . As yet the smal ler pep t ides assoc ia ted with R N A po lymerase in other g r am posi t ive bac te r ia have not been well s tudied. How- ever, in Lactobaci l lus casei there is an R N A po lymer - ase f rac t ion which elutes ear ly f rom the D N A -ce l l u - lose column, has low enzyme activity, and conta ins a 28,000 da l t on pep t ide but no o- (Stetter , 1977). This enzyme m a y be ana logous to the E 6 fo rm found in B. subtilis. Curren t s tudies are being d i rec ted to de te rmine whether 6 is present and has a s imilar func- t ion in o ther g r am posi t ive organisms.

Acknowledgements. Research supported by NIH grant USPHS GM- 19673-07 and NSF grant PCM 76-80788-02.

References

Avila, J., Hermoso, J., Vinuela, E., Salas, M.: Purification and properties of DNA-dependent RNA polymerase from Bacillus subtilis vegetative cells. Eur. J. Biochem. 21, 526-537 (1971)

Bordier, C., Rossetti, G.P.: Subunit composition of Escherichia coli RNA polymerase during transcription in vitro. Eur. J. Bio- chem. 65, 147-153 (1976)

Burgess, R.R. : Purification and physical properties of E. coil RNA polymerase. In: RNA polymerase (R. Losick and M. Chamber- lin, eds.), pp. 69 100. New York: Cold Spring Harbor Labo- ratory 1976

Chamberlin, M.J.: The selectivity of transcription. Annu. Rev. Biochem. 43, 721-775 (1974)

Doi, R.H.: Genetic control of sporulation. Annu. Rev. Genet. 11, 29 48 (1977)

gnkuda, R., Doi, R.H. : Two polypeptides associated with ribonu- cleic acid polymerase core of Bacillus subtilis during sporulation. J. Bacteriol. 129, 422 432 (1977)

Fukuda, R., Ishihama, A., Saitoh, T., Taketo, M.: Comparative studies of RNA polymerase subunits from various bacteria. Mol. Gen. Genet. 154, 135 144 (1977)

Gerard, G.F., Johnson, J.C., Boezi, J.A. : Release of the ~ subunit of Pseudomonasputida deoxyribonucleic acid dependent ribonu- cleic acid polymerase. Biochemistry 11,989 997 (1972)

Halling, S.M., Burtis, K.C., Doi, R.H.: Reconstitution studies show that rifampicin resistance is determined by the largest polypeptide of Bacillus subtilis RNA [~olymerase. J. Biol. Chem. 252, 9024-9031 (1977)

Halling, S.M., Burtis, K.C,, Doi, R.H.: fl" subunit of bacterial RNA polymerase is responsible for streptolydigin resistance in Bacilh~s subtilis. Nature 272, 837 839 (1978)

Halling, S.M., Doi, R.H. : A template independent, rifampicin sensitive poly (A). poly (U) synthesizing activity present in Bacillus subtilis. Biochem. Biophys. Biophys. Res. Commun. 80, 349-354 (1978)

Hedrick, J.L., Smith, A.J.: Size and charge isomer separation and estimation of molecular weights of proteins by disc gel electrophoresis. Arch. Biochem. Biophys. 126, 155-164 (1968)

Krakow, J.S.: A. vinelandii RNA polymerase: template dependent release of the 7 protein. Fed. Proc. 28, 659 (1969)


Krakow, J.S.: Acrylamide gel electrophoresis as a tool for the study of RNA polymerase and the sigma initiation factor. In: Methods in enzymology (L. Grossman and K. Moldave, eds.), Vol. XXI, pp. 520 528. New York: Academic Press Inc. 1971

Krakow, J.S., v o n d e r Helm, K.: Azotobacter RNA polymerase transcripts and the release of sigma. Cold Spring Harbor Symp. Quant. Biol. 35, 73-83 (1970)

Losick, R., Shorenstein, R.G., Sonenshem, A.L. : Structural alter- ation of RNA polymerase during sporulation. Nature 227, 910-913 (1970)

Pero, J., Nelson, J., Fox, T.D.: Highly asymmetric transcription by RNA polymerase containing phage-SP01-induced polypeptides and a new host protein. Proc. Natl. Acad. Sci. U.S.A. 72, 1589-1593 (1975)

Ruet, A., Sentenac, A., Fromageot, P.: On the liberation of cr and the molecular weight of E. coli RNA polymerase. FEBS Lett. l l , 169-17i (1970)

Sedmak, J.J., Grossberg, S.E.: A rapid, sensitive, and versatile assay for protein using Coomassie brilliant blue G250. Anal. Biochem. 79, 544-552 (1977)

Spiegelman, G.B., Hiatt, W.R., Whiteley, H.R.: The role of the 21,000 molecular weight polypeptlde of Bacillus subtilis RNA

polymerase in RNA synthesis. J. Biol. Chem. 253, 1756-1765 (1978)

Stetter, K.O. : Transcription in Lactobacillaceae - DNA - dependent RNA polymerase from Lactobacillus casei isolation of transcription factor y. Hoppe-Seylers Z. Physiol. Chem. 358, 1093-1104 (1977)

Tjian, R., Losick, R., Pero, J., Hinnebush, A.: Purification and comparative properties of the delta and sigma subunits of RNA polymerase from Bacillus subtilis. Europ. J. Biochem. 74, 149-154 (1977)

Travers, A.A., Burgess, R.R. : Cyclic re-use of the RNA polymerase sigma factor. Nature 222, 537 540 (1969)

Williamson, V., Doi, R.H. : Delta factor can displace sigma factor from Bacillus subtilis RNA polymerase holoenzyme and reg- ulate its initiation activity. Mol. Gen. Gent. 161, 135 141 (1978)

Communicated by E. Bautz

Received February 19, 1979

sigma factor is not released during transcription in bacillus subtilis

Documents