Dec 14, 1972 - the liquid was decanted, the beads were washed on a filter funnel with several ml of ... gel electrophoresis by boiling 3 min in 60 Ml of sample buffer. 490 .... column (Biogel A-1.5 m), and eluted with the same buffer. (for 2-5 g of ... to the 0.4 M KCl fraction of the phosphocellulose column pre- cipitated core ...
Proc. Nat. Acad. Sci. USA Vol. 70, No. 2, pp. 490-494, February 1973
Isolation of a New RNA Polymerase-Binding Protein from Sporulating Bacillus subtilis (antibody precipitation)
ARNO L. GREENLEAF, THOMAS G. LINN, AND RICHARD LOSICK The Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138
Communicated by S. E. Luria, December 14, 1972 ABSTRACT RNA polymerase was precipitated from extracts of radioactively labeled vegetative and sporulating Bacillus subtilis with antiserum prepared against vegetative core polymerase. The precipitates were solubilized and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Antiserum added to an extract of veitive B. subtilis precipitated only the known subunits of core RNA polymerase, but antiserum added to an extract of sporulating cells precipitated a new polypeptide of 70,000 daltons in addition to the subunits of core enzyme. The 70,000-dalton polypeptide precipitated from an tiet of a mixture of vegetative and sporulating B. subb, eqarately labeled with two different radioisotopes, contained only the radioisotope characteristic of the sporulating cells. The 70,000-dalton protein has been freed of core RNA polymerase and extensively purified by chromatography on phosphocellulose. Precipitation of the purified 70,000dalton protein by the anti-polymerase serum requires the prior addition of vegetative or sporulation core RNA polymerase. The reaction is specific since the purified protein is not precipitated during antibody precipitation of either phage X repressor or bovine serum albumin. The RNA polymerase-binding protein appears during the third hour of sporulation and is apparently not synthesized by the sporulation-defective mutant rfr 10.
extensively purified. A function for this protein in sporulation transcription has not yet been demonstrated. METHODS
Buffers. Buffers used derive from those described by Burgess (7) and by Chamberlin (8). Buffer G: 0.05 M Tris HCl (pH 7.5)-0.01 M MgCl2-0.2 M KCl-0.1 mM EDTA-0.1 mM dithiothreitol-10%/0 (v/v) glycerol. Buffer A: 0.05 M Tris* HCl (pH 7.9)-0.1 mM EDTA-0.1 mM dithiothreitol-0.01 M MgCl2-stated glycerol and KCl concentrations. Buffer C: the same as buffer A but MgCl2 omitted. Storage buffer: 0.01 M Tris * HCl (pH 7.9)-0.01 M MgCh2-0.1 M KCl-0.1 mM dithiothreitol-0.1 mM EDTA-50% (v/v) glycerol.
Preparation of a Cell Extract for Antibody Precipitation. All steps were at 0-4o unless otherwise noted. Each gram of radioactively labeled cells was mixed with 5 ml of buffer G, 6 ml of 120-.um glass beads, 0.25 ml of phenylmethyl sulfonyl fluoride solution (6 mg/ml of 95% ethanol), and 1 drop tri-nbutylcitrate, and disrupted 3-4 min in a Braun shaker. After the liquid was decanted, the beads were washed on a filter funnel with several ml of buffer G; the filtrate and decanted liquid were combined and centrifuged at 110,000 X g for 1.5 hr. The resulting supernatant was adjusted to 60% saturation with solid ammonium sulfate, and after stirring for 0.5 hr, the solution was centrifuged at 100,000 X g for 0.5 hr. The precipitate was resuspended in buffer A (10% glycerol0.1 M KCl), and the ammonium sulfate precipitation was repeated. After centrifugation, the final precipitate was resuspended in a small volume (typically 1 ml for a preparation from 2 g of cells) of buffer A (5% glycerol-0.05 M KCl) and dialyzed for 2 hr with two changes of the same buffer. After dialysis, the solution was sometimes cloudy and was centrifuged 20 min at 12,000 X g. The supernatant was used for the antibody precipitations.
RNA polymerase of Bacillus subtilis undergoes at least two changes in subunit structure during the process of sporulation. During the first hour of sporulation, the loss of vegetative sigma factor activity causes a change in the template specificity of RNA polymerase (refs. 1, 2; Linn, T., Shorenstien, R., Greenleaf, A. & Losick, R., in preparation; and J. Brevet, personal communication). Later during sporulation, one of the 0 subunits of polymerase disappears and is apparently replaced by a smaller polypeptide of 110,000 daltons (refs. 2, 3; and Linn et al., in preparation). The loss of sigma factor activity early during sporulation has offered a possible explanation for the turn off of ribosomal RNA synthesis (4) and the failure of phage ye to grow in sporulating B. subtilis (5), since the sigma polypeptide is required for the transcription of ribosomal RNA genes (6) and 4e DNA in vitro (2). These findings prompted us to search in sporulating B. subtilis for similar factors that could direct the transcription of sporulation genes. Such polypeptides might be expected to bind to RNA polymerase and could be isolated by virtue of this binding. We report here the isolation of a 70,000-dalton polypeptide of sporulating B. subtilis that binds to RNA polymerase. This protein appears during the third hour of sporulation and is not present in a mutant blocked early in the sporulation process. The 70,000-dalton polypeptide binds tightly to phosphocellulose and has been
Antibody Precipitations. Preparation of partially purified rabbit antiserum against vegetative core RNA polymerase will be described (Linn et al., in preparation). Precipitation reactions contained phenylmethyl sulfonyl fluoride solution (6 mg/ml; 1 ul/20 IAl reaction mixture) and sufficient partially purified antiserum to precipitate about 30 jug of polymerase. The reaction was allowed to proceed at either 0° or 50 overnight. The precipitate was centrifuged 15 min at 12,000 X g and washed three times with 0.05 M phosphate buffer (pH 7.0). Typically, about 1% of the radioactivity of the extract prepared as described above was precipitated. The washed precipitate was solubilized for SDS (sodium dodecyl sulfate)gel electrophoresis by boiling 3 min in 60 Ml of sample buffer
Abbreviation: SDS, sodium dodecyl sulfate. 490
Proc. Nat. Acad. Sci. USA 70
(1973)
containing 3% SDS, 0.01 M P04 (pH 7.2), 0.3 M 2-mercaptoethanol, 10% glycerol, and 0.002% bromphenol blue.
SDS-Polyacrylamide Gel Electrophoresis. Gels were 5% acrylamide, 0.5 X 6 cm; solutions were as described by Weber and Osborn (9). Gels were cut into 1.1-mm slices. Each slice was placed in a vial containing 8 ml of a 3% solution of Protosol (New England Nuclear Corp.) in toluene-base scintillation fluid, and the vials were incubated at 370 overnight with shaking. Radioactivity recovered from a gel by this method is more than 90% of that applied to the gel. RESULTS A Sporulation Polypeptide Precipitated with RNA Polymerase. Goff and Weber (10) have used antibody precipitation of Escherichia coli RNA polymerase to show covalent attachment of a phosphate-containing compound to the a subunit of polymerase after phage T4 infection. We have taken advantage of this method to search for new subunits of RNA polymerase in sporulating B. subtilis 3610. Accordingly, antiserum prepared against vegetative core polymerase was first added to an extract (see Methods) of vegetative cells radioactively labeled with [3H]tryptophan. The precipitate was solubilized and analyzed by SDS-polyacrylamide gel electrophoresis. Fig. 1A shows that almost all the radioactivity in the precipitate was located in the (38' and a subunits of the vegetative core polymerase. Apparently, very little if any vegetative sigma factor (55,000 daltons) was precipitated. In a separate experiment an extract of [3HItryptophanlabeled sporulating cells harvested 4.5 hr after the end of logarithmic growth was treated with the antiserum. SDS-gel electrophoresis of the antibody precipitate revealed a protein of about 70,000 daltons in addition to the ( and a subunits of core enzyme (Fig. 1B). The coprecipitation of the polypeptide of molecular weight 70,000 with core polymerase suggested that in sporulating cells this polypeptide is associated with RNA polymerase. It should be noted that the two (3 subunits of vegetative core polymerase are not resolved into separate bands by the 5% acrylamide gels used here to analyze the antibody precipitates. A high resolution SDS-gel system using stacking gels allows separation of the ,B bands and has been used to demonstrate partial disappearance of one of the (3 subunits from the antibody precipitates during the fourth hour of sporulation (Linn et al., in preparation). Since the gels used in Fig. 1 do not separate the two ,B bands, the 155,000-dalton protein band in these gels has been labeled simply (3. Furthermore, we find that, as shown in Fig. 1B, the 110,000-dalton polypeptide previously described in cells harvested during the sixth hour of sporulation (2) frequently does not appear in precipitates derived from sporulating cells, possibly because it is lost during precipitation of polymerase by antibody. Is the 70,000-dalton protein in Fig. 1B a new component of sporulating cells, or could it be an artifact arising from in vitro lproteolysis of polymerase by a protease in the sporulating cell extract? To distinguish between these two possibilities an extract prepared from a mixture of 3H-labeled vegetative cells and 'IC-labeled sporulating cells was treated with the anti-polymerase serum and the precipitate was analyzed by SDS-gel electrophoresis (Fig. 1C). The radioactivity in the 70,000-dalton protein species is exclusively the 14C label of the sporulating cells. The reverse double-label mixing experi-
RNA Polymerase-Binding Protein
491
ment (not shown) with 14C-labeled vegetative cells and 3Ilabeled sporulating cells confirmed that the 70,000-dalton protein contains only the radioactivity of the sporulating cells. The 70,000-dalton protein, therefore, is present only in sporulating cells and is not a product of in vitro proteolysis of vegetative polymerase.
Purification of the 70,000-Dalton Protein. The coprecipitation of the 70,000-dalton protein with core enzyme suggests that it is precipitated by virtue of its binding to the core polymerase. An alternative explanation, however, is that the 70,000-dalton protein is directly precipitated by the antiserum. To distinguish between these possibilities it was necessary to separate the 70,000-dalton polypeptide from core enzyme. An extract of sporulating cells was prepared as described in Methods through the first ammonium sulfate precipitation. The resuspended precipitate was dialyzed for 1 hr against buffer A (10%? glycerol-0.3 M KCl), applied to an agarose
5000
AA Vegetative 70,000 Daltons
I
2500
3500.
70,000 Daltons
B
Sporulotion a
1750-
1200
~~~~~DyeI
a
Dye
C
Vegetative 70,000 Doltons 600-
11 t~prulotin
O
1
20
30
40
Dye
50
60
Slice No. FIG. 1. SDS-gel patterns of solubilized antibody precipitates of RNA polymerase. RNA polymerase was precipitated at 0° from an extract of either: (A) a 1500-ml culture of vegetative
B. subtilis 3610 grown in 121A medium (5) containing 2 mCi of [3H]tryptophan and harvested during late-logarithmic growth; (B) a 1500-ml culture of partially synchronized B. subtilis grown in 121B medium (5) containing 2 mCi of [3H]tryptophan and harvested 4.5 hr after the end of logarithmic growth; (C) a mixture of 2 g of [3H]vegetative cells grown as in (A) and 4 g of sporulating cells grown as in (B) but labeled with 250 ,uCi of [14C]tryptophan from 0.5 to 4.5 hr after the end of logarithmic growth. The antibody precipitates were solubilized and analyzed by SDS-gel electrophoresis as described in the Methods. ( *, 3H;. 14C)
Biochemistry: Greenleaf et al.
492
Proc. Nat. acad. Sci. USA 70
column (Biogel A-1.5 m), and eluted with the same buffer (for 2-5 g of cells, a 120-ml column was used). Peak fractions of polymerase activity were pooled, and RNase A was added (5X crystallized, Calbiochem; 0.2 mg/g of cells used). The RNase-treated sample was dialyzed against buffer C (20% glycerol-0.3 KCl) overnight. Then, MgCl2 (to give 0.01 M) and DNase (DPFF, Worthington; 0.1 mg/g of cells used) were added and the sample was dialyzed against buffer A (20% glycerol-0.3 M KCl) at 40 for 8 hr. After ammonium sulfate precipitation at 60% saturation, the resuspended sample was dialyzed against buffer C (20% glycerol-0.1 M KCl). SDS-gel analysis of an antibody precipitate of a small A
175070,000
Daltons
Dye
O.I~~~~~~~~~~~ 2200-
B
1100o
70000
-
t
0.4M
Daltons
KCI Step Stop
~~~~~plusLOM KC1
LOM KCI
Step D ye
5000
_70,0040
Daltons
(1,973)
portion of the dialyzed sample showed that both the 70,000dalton protein and core enzyme were still present (not shown). The dialyzed sample was applied to a 2-ml phosphocellulose column, and the column was then eluted with salt steps of 0.24 M, 0.4 M, and 1.0 M KCl in buffer C. Antiserum added to the 0.4 M KCl fraction of the phosphocellulose column precipitated core enzyme as expected (2), but none of the 70,000dalton protein was present in the precipitate (Fig. 2A). The 70,000-dalton polypeptide had apparently been removed from core polymerase in this phosphocellulose step. Antiserum added to the other salt-step fractions of the column also failed to precipitate the 70,000-dalton species. In particular, antiserum added to the 1.0 M KCl fraction precipitated neither the 70,000-dalton polypeptide nor core polymerase subunits (triangles in Fig. 2B). Coprecipitation of core polymerase and the 70,000-dalton protein did occur, however, when antiserum was added to a mixture of the 0.4 M and the 1.0 M KCl fractions (Fig. 2B). This result suggests that the 70,000-dalton polypeptide was present in the 1.0 M KCl fraction, but was only precipitated in the presence of added core enzyme. Fig. 2C shows that the 70,000-dalton protein was the major species in the 1.0 M KCl fraction, as determined by SDS-gel electrophoresis of a trichloroacetic acid precipitate of that fraction. Further purification of the 70,000-dalton protein can be obtained by washing the phosphocellulose column with 0.7 M KCl before the 1.0 M KCl step. Table 1 shows that the precipitation of the 70,000-dalton protein by antibody is also stimulated by vegetative core polymerase as well as by the late sporulation polymerase, lacking a , subunit (2), from cells harvested during the sixth hour of sporulation. The adsorption of the 70,000-dalton protein to the phosphocellulose column suggests that it might bind to nucleic TABLE 1. Specificity of antibody precipitation of the binding protein cpm of binding protein preAntigen 1. None 2. 15 ,g of vegetative
30
Slice No. FIG. 2.
Separation of the
70,000-dalton
protein from polymer-
ase by phosphocellulose chromatography. 2 g of sporulating cells radioactively labeled by the addition of 3 mCi of [3HJ tryptophan
core 3. 15 ,ug of sporulation core 4. 12 ,ug of X repressor 5. 20j(g of bovineserum albumin
Antiserum
cipitated
Anti-vegetative core Anti-vegetative core
98 2226
Anti-vegetative
1512
core
Anti- X repressor Anti-bovine-serum albumin
144 70
0.5 hr after the end of logarithmic growth and harvested during the fourth hour of sporulation were subjected to the purification
described in the text. The following were treated with antiserum at 50
and the precipitates
phoresis:
(A) an aliquot
were
analyzed by SDS-gel electro-
(0.2 of the total)
of
the
0.4
M
KCl
fraction of the phosphocellulose column in a total volume of 1.8 ml and a final KCI concentration of 0.38 M; (B) both an aliquot (0.2 of the total) of the 1.0 M KCl fraction alone
(zt-F--z)
and
a mixture of equal amounts (0.2 of the total) of the 0.4 M KCl
fraction and the 1.0
MI
KCl fraction
(@-.-*), adjusted to the 5°. (C) SDS-gel analysis
conditions of (A ) and incubated 2 hr at of protein precipitated in
10% trichloroacetic
acid at 00
from
0.2 of the total of the 1.0 M KCl fraction of the phosphocellulose
column.
Reaction mixtures contained 100 ,ul of 70,000-dalton protein purified through the phosphocellulose step as described in the text (7000 cpm; 510 cpm/,ug; in Storage Buffer), 0.62 ml of buffer A (10% glycerol-0.3 MV KCl), and the antigen to be tested. After incubation at 50 for 2 hr, 0.25 ml of the specified antiserum was added and the precipitin reaction proceeded overnight at 50. The precipitates were subjected to SDS-gel electrophoresis. The gels were stained and scanned in a densitometer to demonstrate that the antigen was precipitated in each case. The gels were then sliced to determine the amount of radioactive 70,000dalton protein precipitated. Sporulation core RNA polymerase (a gift of J. Pero) was purified from cells harvested during the sixth hour of sporulation.
Proc. Nat. Acad. Sci. USA 70
(1978)
RNA Polymerase-Binding Protein oo
5000
1.5k
300 200 1
/-7_
AZ~~~~~
~3000
100 I
1.0
J
Q: z
50~
Q1 k.
bq~~~~~~~~~~~~~~~~~~~4
I
/_
-2
-I
0
1
2
Time of Addition of Antiserum(hr)
FIG. 3. Effect of time of addition of antiserum on the amount of purified 70,000-dalton protein precipitated. To each of four tubes containing 15 ,ug of vegetative core polymerase in 0.615 ml of buffer A (10% glycerol-0.3 MI KCl) antiserum was added at a different time before or after the addition of purified 70,000dalton binding protein (12,000 cpm; 605 cpm/pug). The tubes were then incubated at 50 overnight (0.2 M KCI final concentration). Time of addition of the 70,000-dalton protein was called time 0. The precipitates were analyzed by SDS-gel electrophoresis as for Table 1.
acid. This raises the possibility that polymerase may not interact directly with the 70,000-dalton species but with small fragments of DNA or RNA that may be attached to the 70,000-dalton protein. To test this possibility, the purified 70,000-dalton binding protein was treated with approximately equimolar amounts of RNase and DNase for several hours at 50, then mixed with purified vegetative core polymerase and precipitated with antiserum at 0.4 M KC1. The nuclease treatment had no significant effect on the amount of the 70,000-dalton protein precipitated as compared to a parallel experiment using untreated 70,000-dalton protein (data not shown). In addition, it is known that the 0.4 M KCl concentration of the precipitation reaction greatly inhibits the binding of polymerase to nucleic acid (11). Direct Interaction of the 70,060-Dalton Protein with Core RNA Polymerase. As further evidence for a direct interaction of the 70,000-dalton polypeptide with polymerase, vegetative core enzyme was treated with antiserum at different times before and after the addition of radioactive 70,000-dalton protein. The experiment of Fig. 3 shows that addition of antiserum to core polymerase before the addition of the 70,000-dalton protein prevented precipitation of the 70,000dalton species. Addition of antiserum after addition of the 70,000-dalton protein, however, did not prevent the precipitation of this species. As a control, Fig. 3 shows that at each time point, the amount of vegetative core precipitated was nearly constant. Thus, precipitation of the 70,000-dalton protein by antibody requires interaction with polymerase before addition of antiserum, indicating that the 70,000dalton protein binds directly to free core enzyme and not to core enzyme complexed with antibody. (Hereafter the 70,000dalton protein will also be referred to as "binding protein.") Specificity of the Binding Protein. Is the binding of the 70,000-dalton species to core enzyme specific, or does it reflect a general affinity. of the binding protein for cellular protein? SDS-gel analysis shows that radioactivity in the binding protein represents less that 2% of the total radioactivity in an extract of sporulating cells (data not shown). Yet antibody precipitates of polymerase contain as much as 50% of their radioactivity in the 70,000-dalton protein. The
;s
.
O.SH
0
-2 -3
493
2
4
6
Hours into Sporulation FIG. 4. Time of appearance of the binding protein during sporulation. To a 1500-ml culture in 121B was added 2 mCi of ['H]tryptophan during vegetative growth, and 250-ml portions were harvested at various times. Antibody precipitates of extracts prepared from cells at each time point were analyzed by SDS-gel electrophoresis. The turbidity of the culture was measured in Klett units (green filter, 540 nm). Refractile prespores first appeared during the sixth hour after the end of logarithmic growth.
binding protein apparently has a much greater affinity for
polymerase than for cellular protein in general, for if it bound to all proteins equally, it should represent less than 2% of the radioactivity precipitated. As a more stringent test of its specificity, the purified binding protein was incubated 2 hr with purified phage X repressor (a gift of P. Chadwick) and then anti-repressor serum was added. Table 1 shows that whereas the X repressor was precipitated, there was no l)recipitation of the binding protein. Similarly, antiserum to bovine-serum albumin (Mann Research Laboratories) did not precipitate binding protein that had been previously mixed with the bovineserum albumin (Table 1). The 70,000-dalton protein thus appears to bind specifically to RNA l)olymerase. These experiments also indicate that l)recipitation of the 70,000dalton protein is not due merely to entrapment by antibody precipitates.
Stoichiometry and Radioactive Labeling of the Binding Protein. Densitometer tracings of stained SDS-gels of antibody precipitates indicate that the number of molecules of binding protein per core RNA polymerase molecule varies from 0.3 to 2.0, depending on cell preparation and l)recipitin reaction. 12-
8)
70,000 Daltons
Dye
Sj4-
10
20
30
40
Slice No. FIG. 5. Absence of the 70,000-dalton binding protein in stationary phase rfr 10. To a culture of rfr 10 (500 ml of 121B) synchronized as described for wild-type cells (5), was added 1 mCi of ['H] tryptophan during logarithmic growth (Klett = 100), and the culture was harvested 4 hr into stationary phase. RNA polymerase was precipitated from an extract of the cells and the precipitate was analyzed by SDkS-gel electrophoresis.
494
Biochemistry: Greenleaf et al.
On the other hand, estimates of stoichiometry based on radioactivity profiles of sliced gels appear to indicate a larger number of molecules of binding protein per polymerase (see Fig. 1B, for example). Measurement of both protein and radioactivity on the same gel shows, however, that the specific radioactivity of the binding protein is about five times higher than that of the i subunits, when the labeling was with radioactive tryptophan added either during vegetative growth or 1 hr after the end of logarithmic growth. The binding protein thus seems to be relatively rich in tryptophan. In contrast, we have been unable to label the binding protein with [05S]. sulfate or [3H]methionine under conditions resulting in incorporation of radioactivity into the , subunits of polymerase. Time of Appearance of the Binding Protein During Sporulation. A culture of cells was radioactively labeled during logarithmic growth, and samples were harvested at intervals during sporulation. From SDS-gel analysis of the antibody precipitates prepared from extracts of each of the samples, we determined the ratio of radioactivity in the binding protein to radioactivity in the j3 subunits. The experiment of Fig. 4 shows that the binding protein first appears during the third hQur of sporulation and persists until at least the sixth hour of sporulation. Appearance of the Binding Protein is Blocked in the Sporulation Defective Mutant rfr 10. If the polymerase-binding protein is associated specifically with sporulation, it should not appear during stationary phase in mutants blocked early in sporulation. Rfr 10 is an oligosporogenous rifampicin-resistant mutant of B. subtilis 3610 (12). During stationary phase, rfr 10 RNA polymerase does not undergo the change in the template specificity that occurs in wild-type cells during the first hours of sporulation (12). The experiment of Fig. 5 shows that no 70,000-dalton protein was coprecipitated with RNA polymerase by antibody treatment of an extract of radioactively labeled stationary-phase cells of rfr 10, One possibility is that the 70,000-dalton protein is present in stationary phase rfr 10 but fails to bind to core enzyme, since rfr 10 is a mutant in RNA polymerase. However, the 70,000-dalton binding protein was not precipitated from rfr 10 extracts even when wild-type core polymerase was added during the antibody reaction.
DISCUSSION A protein that binds to RNA polymerase has been isolated from sporulating B. subtilis. This protein is synthesized during sporulation since it incorporates a radioactive amino acid added after the end of logarithmic growth. Experiments in which the 70,000-dalton protein was isolated by antibody precipitation of polymerase from a mixture of vegetative and sporulating cells separately labeled with [3H]- and ['IC] tryptophan further demonstrate that the binding protein is present in precipitable form only in sporulating cells. It is not excluded, however, that the binding protein might also be synthesized during vegetative growth in a precursor form that does not bind to RNA polymerase. Appropriate pulsechase experiments to test this possibility are hampered by the turnover of proteins during sporulation (13).
Proc. Nat. Acad. Sci. USA 70
(1973)
The binding of the 70,000-dalton protein to RNA polymerase is apparently specific, since this protein does not bind to phage X repressor, to bovine-serum albumin, or to most of the proteins of sporulating cells. It is possible, however, that the binding to RNA polymerase is accidental and unrelated to the function of the 70,000-dalton protein during sporulation. The RNA polymerase-binding protein of sporulating B. subtilis differs in several respects from the vegetative sigma factor, a protein that also binds to polymerase (ref. 2, and Shorenstein and Losick, in preparation). First, while sigma factor is apparently lost from RNA polymerase early during sporulation (ref. 1, and Linn et al., in preparation), the binding protein first appears during the third hour of sporulation. Second, the 70,000-dalton binding protein is larger than sigma factor [about 55,000 daltons (2)] and is therefore not derived from proteolytic cleavage of vegetative sigma factor. Third, although both sigma (2) and binding protein are separated from core polymerase by chromatography on phosphocellulose, vegetative sigma factor does not bind to the resin and elutes in the flow-through, while the sporulation binding protein adheres tightly to phosphocellulose and elutes at 1 M KCl. A fourth difference between sigma and the 70,000-dalton protein is that while both proteins bind to vegetative polymerase, sigma does not bind to late sporulation polymerase, which lacks a , subunit (Shorenstein and Losick, in preparation). Possibly, sigma and the 70,000-dalton protein bind to different sites on RNA polymerase. The sigma subunit of vegetative polymerase directs asymmetric transcription of B. subtilis DNA in vitro (Pero in ref. 3) and the synthesis of ribosomal RNA in vitro (6). Attempts to demonstrate a role for the 70,000-dalton binding protein in the transcription of sporulation genes have not yet been successful. We thank J. Pero and A. L. Sonenshein for helpful discussions and a critical reading of the manuscript, and T. Landers for use of his computer program for correcting double-label radioactivity data. This work was supported by National Science Foundation Grant GB-27610. 1. Losick, R. & Sonenshein, A. L. (1969) Nature 224, 35-37. 2. Losick, R., Shorenstein, R. G. & Sonenshein, A. L. (1970) Nature 227, 910-913. 3. Losick, R. (1972) Annu. Rev. Biochem. 41, 409-446. 4. Hussey, C., Losick, R. & Sonenshein, A. L. (1971) J. Mol. Biol. 57, 59-70. 5. Sonenshein, A. L. & Roscoe, D. H. (1969) Virology 39, 265276. 6. Hussey, C., Pero, J., Shorenstein, R. G., & Losick, R. (1972) Proc. Nat. Acad. Sci, USA 69, 407-411. 7. Burgess, R. R. (1969) J. Biol. Chem. 244, 6160-6167. 8. Berg, D., Barrett, K. & Chamberlin, M. (1971) Methods Enzymol. 21, 506-519. 9. Weber, K. & Osborn, M. (1969) J. Biol. Chem. 244, 44064412. 10. Goff, C. G. & Weber, K. (1970) Cold Spring Harbor Symp. Quant. Biol. 35, 101-108. 11. Richardson, J. P. (1969) Progr. Nucleic Acid Res. Mol. Biol. 9, 75-116. 12. Sonenshein, A. L. & Losick, R. (1970) Nature 227, 906909. 13. Kornberg, A., Spudich, J. A., Nelson, D. L. & Deutscher, M. P. (1968) Annu. Rev. Biochem. 37, 51-78.
