Cooperative DNA binding by CI repressor is dispensable in a phage λ variant

Babić and Little. 10.1073/pnas.0602223104.

Supporting Information

Files in this Data Supplement:

SI Figure 4
SI Table 1
SI Table 2
SI Table 3
SI Text
SI References




SI Figure 4

Fig. 4. Construction of templates for footprinting. (A) Map of the O_R region and flanking regions of the l genome showing the locations of restriction sites used in cloning and the prm252 andO_R2up mutations. Only the initial portions of cI and cro are shown. (B) Multiple cloning site of pDWW12. H, HindIII; Sp, SphI; P, PstI; X, XbaI, B, BamHI; N, NsiI; (St), StuI site blocked by Dcm methylation; Bg, BglII; K, KpnI. (C) Construction of plasmids for DNase I footprinting. The location of primers is shown. The resulting fragments were cloned into pDWW12, producing the inserts shown in D. (D) Structure of templates for DNase I footprinting. These plasmids were used as templates for PCR; see SI Text.

Table 1. Mutations suppressing the lJL506 lysogenization defect

These mutations were identified in lysogenizing variants of lJL506, isolated as described in the main text. In the O_R3 mutant, the bolded underlined base is G WT.

Table 2. Plasmids used

 

 

Plasmid	Relevant genotype	Source or SI ref.
pDWW12	pBS(-), trp attenuator between KpnI and EcoR1	David Wert, this laboratory
pET11a	T7 expression vector	Novagen, 9
pFG600 cI Y210H		Gary Ackers, 8
pFG600	Lac promoter::wild type cI	6
pJWL236	Fusion of residues 5-93 of l cI :CTD of HK022 cI	J.W.L., unpublished
pJWL244	OR v1 v3	3
pJWL 500	NsiI-BglII fragment from wild-type, resected SalI site in polylinker	This work
pJWL501	OR2up	This work
pJWL505	cI Y210H	This work
pJWL504	Ptac:: cI Y210N	This work
pJWL506	OR2up, prm252	This work
pJWL509	T7p:: cI Y210N	This work
pJWL510	T7p:: cI Y210H	This work
pJWL511	OR1+ OR2+	This work
pJWL512	OR1ºOR2+	This work
pJWL514	OR1ºOR2up	This work
pJWL515	T7p:: cI Y210F	This work
pJWL516	OR1+ OR2º	This work
pLR1 cI Y210N	cI Y210N	Ann Hochschild, 10
pShep84	T7p::cI+	Donald Shepley, this laboratory

Table 3. Primers employed

 

SI Text

Isolation of suppressors of cI Y210N. Suppressors of cI Y210N were isolated as described in the main text. After 30 min, a portion of the infected culture was streaked onto kanamycin plates; several isolates were transformed with pFG600, and their lysogen number was tested. Each was found to be a multiple lysogen. This might have arisen either because polylysogeny can occur after single infection (because some replication of the phage DNA occurs after infection) or because lJL506 and a variant phage colysogenized.

Mutations found in suppressor screen. DNA sequence analysis of the cI-cro interval identified mutations in cI, cro, O_R, and P_RM (SI Table 1). None of these variants fulfilled our criteria for near-normal regulatory behavior. Most of the cro mutants formed tiny plaques, indicating a major defect in lytic growth (as previously found for cro mutants; see refs. 1 and 2). For many isolates, single lysogens could not be obtained. Isolates able to form stable lysogens and give good lytic growth after infection had a major growth defect after UV induction (see main text). We surmise, but have not tested directly, that most or all of these isolates passed our selection procedure by virtue of having been present in the pool of lysogens as double lysogens with the starting lJL506 parent, and that this combination somehow allowed a substantial burst to occur after UV induction.

Isolation and characterization of the cI Y210F phage lJL504. In the crystal structure of the C-terminal domain (CTD) of CI, subunits interact both by the dimerization interface and by a second interface (17). The latter was suggested to mediate cooperative interactions, because several cooperativity mutations affect residues lying in that interface. Among these is Tyr-210, which was suggested to interact with Gln-204 by a hydrogen bond with the Tyr-OH as donor via a bridging water molecule.

To examine this prediction, we characterized another mutant with a change in Y210, in this case to Phe, which should be unable to form a hydrogen bond. A phage carrying cI Y210F, lJL514, was isolated in four steps. First, lJL282 = lJL163 prmup1 was made by methods like those used to make lO_R323 (3). Second, lJL413 cI Y210C prmup-1 was isolated as a suppressor of poor lytic growth by lJL282 on JL6039 (J.W.L., unpublished data); the cI Y210C allele was identified by sequencing of the cI-cro interval. Third, lJL413 was crossed with lJL163. JL2497 was grown in LBMM (LB supplemented with 0.2% maltose and 1 mM MgSO₄), concentrated in TMG, infected at a multiplicity of infection of 5 of each parent, and grown for 90 min in LBMM. Desired recombinants were identified by formation of clear plaques on JL2497, under the assumption that l cI Y210C would form clear plaques, and the cro-O_R-cI region was sequenced, yielding lJL500. Fourth, we selected lysogenizing variants of this phage, by using methods similar to those described in the main text for Y210N, except that lysogens arising in the initial infection were characterized individually rather than grown and UV-induced as a pool. Among these was a pseudorevertant carrying cI Y210F.

This phage, lJL504, had wild-type behavior in vivo by several criteria. First, it formed turbid plaques indistinguishable from those of wild type. Second, it formed stable single lysogens with CI levels similar to the wild-type lysogen (data not shown). Third, it had a burst size after infection and lysogenization frequency similar to a wild-type lysogen (data not shown). Fourth, the threshold behavior, set point, and burst sizes after UV induction were similar to wild type (data not shown), indicating that the physiology and regulatory circuitry were similar between wild type and lJL504. In addition, preliminary in vitro footprinting data suggested that cooperativity of purified Y210F protein is similar to that of wild type.

We conclude that a change of Tyr-210 to Phe has little or no effect on cooperativity. These findings suggest that the interactions made by the aromatic ring of Tyr-210 are important for cooperativity, but that the hydrogen bond observed in the crystal structure is not. Perhaps this bond contributes very little free energy to the interaction. Alternatively, perhaps the hydrogen bond seen in the crystal is important for Tyr-210, but alternative contacts are made in the Y210F mutant.

Plasmid constructions. In the constructions described below, PCR of the O_R region was done by using primers croleft and cIright except as specified. Location of restriction sites in and around the O_R region are shown in SI Fig. 4A, as are the location of P_RM and the prm252 and O_R2up mutations. Plasmids are listed in SI Table 2.

pJWL236: This plasmid was made in multiple steps, not described. The vector was pBR322. The insert contains residues 5-93 of l cI, the C-terminal portion of HK022 cI (4), and 330 bp of l sequence lying downstream of l cI, ending beyond a ClaI site. It was used as a vector for construction of pJWL505, during which the HK022 cI segment was removed.

pJWL500: This plasmid was made from pJWL344 (supplemental material in ref. 5), which contains a NsiI-BglII fragment spanning O_R cloned into pJWL334 (almost identical to pDWW12) (SI Fig. 4B). pJWL344 has a HincII site in O_R as well as the site also cut by SalI. pJWL344 was cut with SalI, filled in with DNA polymerase, and religated. The resulting plasmid had a unique HincII site within O_R. It also bore XhoI and SpeI sites (5), not present in l, which were removed during construction of pJWL501.

pJWL501 = O_R2up: A three-piece ligation was done. pJWL500 was cut with NsiI and BglII. A PCR product of the lJL163 O_R region was cut with HincII and BglII. The third fragment, containing the O_R2up mutation, was isolated from a PCR of the O_R region of lJL300 cut with NsiI and HincII.

pJWL505 = cI Y210H: pFG600 carries a lacP::cI fusion and 264 bp of l DNA downstream of cI, ending at a ClaI site (6). A derivative of pFG600 (7, 8) carrying cI Y210H was obtained from Prof. Gary Ackers. An NsiI-ClaI fragment from this plasmid included the mutant allele, and was ligated into pJWL236 cut with NsiI and ClaI.

pJWL506 = O_R2up, prm252: A PCR product of the O_R region of lJL511 (which contains prm252) was cut with NsiI and BsaBI and was cloned into pJWL501 cut with the same enzymes.

pShep84 = T7p::wild-type cI (made by D. P. Shepley, this laboratory): The backbone for this three-piece ligation was pET11a (9), obtained from Novagen, cut with BamH1 and NdeI. The first insert was an NsiI-to-BamH1 fragment from pFG600; the BamHI site lies distal to cI within the tet gene of pBR322. The second insert was a PCR product of wild-type l cI done with a primer that introduced an NdeI site at the start codon of cI, followed by cutting with NdeI and NsiI.

pJWL509 = T7p:: cI Y210N: A derivative of plasmid pLR1 (10) carrying cI Y210N was obtained from Prof. Ann Hochschild (Harvard University, Cambridge, MA); it was cut with NsiI and BamHI, and the fragment bearing the mutation was cloned into pShep84 cut with the same enzymes.

pJWL510 T7p::cI Y210H: pShep84 was cut with NsiI and BamHI. An insert containing the Y210H mutation was obtained from pJWL505 cut with NsiI and ClaI. The second insert was a BamHI to ClaI fragment from pShep84.

pJWL511, pJWL512, pJWL514, pJWL516: The backbone pDWW12 (SI Fig. 4B) was made by David Wert (this laboratory). It is nearly identical to pJWL334 (5); both contain a polylinker modified from that of pBS(-), and differ only in the sequence between the NsiI and BglII sites. pDWW12 includes sites for KpnI and BamHI and a strong terminator (trp a) beyond the KpnI site, used here to terminate transcription from P_R. pDWW12 was cut with KpnI and BamHI. Inserts were made by using two primers, one containing a KpnI restriction site and the other containing a BamHI restriction site (SI Fig. 4C and SI Table 3). For versions with mutant operators, three mutations were introduced into O_R1 or O_R2 to abolish CI binding by using versions of the primers containing the mutations. Mutant operators are symbolized as O_R1^≡ or O_R2^≡, respectively. These PCR products were cut with KpnI and BamHI and cloned into the pDWW12 backbone (SI Fig. 4D).

For pJWL511, wild-type l was used as the template for a PCR by using the primers KpnIwt and BamHIO_R3-1. For pJWL512, lJL163 was used as the template for a PCR using the primers KpnIO_R1-3 and BamHIO_R3-1. For pJWL514, pJWL501 was used as the template for primers KpnIO_R1-3 and BamHIO_R3-1. For pJWL516, pJWL511 was used as a PCR template by using the T7 universal primer and the O_R2-3 primer.

pJWL515 = T7p::cI Y210F: pShep84 was cut with NsiI and BamHI. An insert containing the Y210F mutation was obtained from the PCR product of the primers rexmid and Nsileft by using lJL504 as template, which was then cut with NsiI and ClaI. The second insert was a BamHI to ClaI fragment from pShep84.

Isolation of CI proteins. Buffers used included the following. Lysis buffer was 100 mM Tris·HCl (pH 8) + 200 mM KCl + 1 mM EDTA + 2 mM CaCl₂ + 10 mM MgCl₂ + 0.1 mM DTT + 5% glycerol (vol/vol). Standard buffer (SB) was 10 mM Tris·HCl (pH 8) + 2 mM CaCl₂ + 0.1 mM EDTA + 0.1 mM DTT + 5% glycerol (vol/vol). Wash buffer was SB to which was added solid (NH₄)₂SO₄ to 3.03 M. Buffer B (BB) was 20 mM potassium phosphate + 0.1 mM EDTA + 10% glycerol + 1 mM DTT. Hydroxyapatite (HA) column buffer was 0.1 M potassium phosphate (KP) + 1.5 mM b-mercaptoethanol. The KP buffer was .7 mol dibasic (K₂HPO₄) + .3 mol monobasic (KH₂PO₄) per liter; its pH was 7.0.

Wild-type and mutant CIs were purified by extensive modification of a published method (11). Wild-type and various cI alleles were cloned in the expression vector pET11a (Novagen) as described above. Plasmids were transformed into BL21(lDE3)/pLysS. One liter of cultures were grown in LB to OD₅₉₀ = 0.2, induced with 1 mM isopropyl b-D-thiogalactoside (IPTG) (Sigma), and grown another 3 h. Induced cells were concentrated by centrifugation, resuspended at 1 g cells per 10 ml lysis buffer, and flash frozen. They were then thawed and sonicated until no longer viscous, followed by centrifugation at 13,000 ´ g for 15 min, and recovery of the supernatant. All subsequent steps were carried out at 4ºC.

The supernatant fluid was made 600 mM KCl by addition of solid KCl. This was stirred for 6 min, made 0.6% in polyethyleneimine-HCl by slow addition of 10% polyethyleneimine-HCl (pH 7.9) with stirring, and stirred another 10 min, followed by centrifugation at 6,500 ´ g for 15 min. The supernatant fluid, containing CI, was brought to 3.03 M ammonium sulfate (BRL) by addition of solid (NH₄)₂SO₄, stirred for 30 min, and centrifuged at 13,000 ´ g for 45 min. The pellet was resuspended in wash buffer in a volume equal to that of previous step, gently agitated for 6 min, and centrifuged at 14,000 ´ g for 10 min. The pellet was redissolved in 1 ml of 1x BB for every 2 g of starting cells and dialyzed overnight against 100x volume BB + 100 mM NaCl (100BB) with one buffer change. The protein was centrifuged at 2,600 ´ g for 5 min to remove a small amount of precipitate. The CI remained in the supernatant.

CI was purified by using three sequential columns: phosphocellulose (Whatman), hydroxyapatite (Bio-Rad), and Affigel Blue (Bio-Rad). Y210N mutant protein eluted earlier than wild type from all three columns; a direct comparison with Y210H protein was not made, but its properties were similar to those of Y210N. A 7- to 8-ml phosphocellulose column was equilibrated to pH 7. The sample was loaded onto the column and washed with 100BB. CI was eluted with a 100-ml gradient in BB from 100-900 mM NaCl.

Peak fractions were loaded (without dialysis) onto a hydroxyapatite column in 0.1M KP + 1.5 mM bmercaptoethanol (bME). The column was washed with 0.1M KP + 1.5 mM bME. CI was eluted with a 100-ml gradient from 0.1M KP + 1.5 mM bME to 1M KP + 1.5 mM bME. The peak fractions were dialyzed against SB + 100 mM KCl, loaded onto the affigel blue column, and washed with 100SB for ≈20 min. CI was eluted with a 40-ml gradient from 100 mM to 1 M KCl in SB. The peak fractions were dialyzed against SB + 200 mM KCl, aliquoted, and flash frozen. The concentration of the protein was determined in 6 M guanidine-HCl as described (12).

Purification of rabbit anti-CI antibody

Rabbit antibody to CI (6) was purified by affinity chromatography. To immobilize CI, wild-type CI was dialyzed into 0.1 M NaHCO₃ - 0.5 M NaCl. Cyanogen-bromide activated Sepharose 4B (Pharmacia) (0.3 g) was activated according to the manufacturer's instructions, mixed with dialyzed CI (0.7 mg), mixed gently for 2.5 h at room temperature, blocked with Tris·HCl (pH 8) for 2 h, and washed, yielding immobilized CI beads. Antiserum was made 0.3 M NaCl - 20 mM Tris·HCl (pH 8.0), 0.1% Tween 20 (TTS), mixed with the immobilized CI beads, and mixed gently overnight at 4°C. The slurry was applied to a small column and washed extensively with TTS. Antibody was eluted with 0.2 M glycine-HCl (pH 2.8), promptly neutralized with 0.05 vol of 1 M Tris base, made 0.02% NaN₃, and stored at -70°C.

DNase I footprinting. Our approach to measuring cooperativity was designed to be sensitive to low residual levels of cooperativity. We used three templates with only two operators, O_R1 and O_R2. In the first, both operators were intact, allowing cooperative binding; in the other two, one or the other operator was inactivated by three mutations, allowing measurement of intrinsic affinity. Because O_R2 binds CI much more weakly than O_R1, a mutant CI with a low degree of cooperativity should influence primarily the affinity for O_R2, and the change in affinity when O_R1 is present should be close to the cooperativity parameter.

Template DNAs were made by PCR with a 5'-³²P-end-labeled primer (pDW12t7) and a second primer (pDW12t3), essentially as described in ref. 13, with the following modifications. The templates for PCR were pJWL511, pJWL512, and pJWL516. PCR was done with Pfu polymerase (Stratagene) with minor modifications to the conditions for the PCRs. Labeled fragments were gel purified. To estimate the molar amounts of recovered fragments, we assumed that the molar yield from the PCR equaled the molar amount of primers used; the ratio of counts recovered from the gel to counts applied was used to calculate the yield of labeled template. This method probably overestimates the concentration, because not all of the primer was incorporated into product. In all cases, DNA concentrations were well below the measured dissociation constant.

The concentration of wild-type CI dimers was determined from the input concentration of CI and a K_dimer of 5.6 nM (14, 15). The dimer dissociation constant for CI Y210H was measured as 7.6 nM (15), and we assumed this value for CI Y210N as well.

Reaction conditions for footprinting reactions were as described in ref. 8. Reactions were carried out as described in ref. 4, and binding curves were measured as described in ref. 4, except that gels were scanned on the Typhoon 9410 scanner (Molecular Dynamics) and analyzed by using ImageQuant (Molecular Dynamics).

SI References

1. Eisen H, Brachet P, Pereira da Silva L, Jacob F (1970) Proc Natl Acad Sci USA 66:855-862.

2. Folkmanis A, Maltzman W, Mellon P, Skalka A, Echols H (1977) Virology 81:352-362.

3. Little JW, Shepley DP, Wert DW (1999) EMBO J 18:4299-4307.

4. Carlson NG, Little JW (1993) J Mol Biol 230:1108-1130.

5. Michalowski CB, Short MD, Little JW (2004) J Bacteriol 186:7988-7999.

6. Mustard JA, Little JW (2000) J Bacteriol 182:1659-1670.

7. Beckett D, Burz DS, Ackers GK, Sauer RT (1993) Biochemistry 32:9073-9079.

8. Burz DS, Ackers GK (1994) Biochemistry 33:8406-8416.

9. Studier FW, Rosenberg AH, Dunn JJ, Dubendorff JW (1990) Methods Enzymol 185:60-89.

10. Whipple FW, Kuldell NH, Cheatham LA, Hochschild A (1994) Genes Dev 8:1212-1223.

11. Johnson AD, Pabo CO, Sauer RT (1980) Methods Enzymol 65:839-856.

12. Gill SC, Von Hippel PH (1989) Anal Biochem 182:319-326.

13. Liu Z, Little JW (1998) J Mol Biol 278:331-338.

14. Koblan KS, Ackers GK (1991) Biochemistry 30:7817-7821.

15. Burz DS, Beckett D, Benson N, Ackers GK (1994) Biochemistry 33:8399-8405.

16. Powell BS, Court DL, Nakamura Y, Rivas MP, Turnbough CL, Jr. (1994) Nucleic Acids Res 22:5765-5766.

17. Bell CE, Frescura P, Hochschild A, Lewis M (2000) Cell 100:810-811.

• Early Edition
• Archives
• Online Submission

• Feature Articles

  (Expanded research articles of exceptional breadth)

• Commentaries

  (Expert commentary on leading research)

• Letters

  (Online-only letters to the editor)

• Inaugural Articles

  (Articles by recently elected NAS members)

• PNAS Profiles

  (Biographical profiles of Academy members)

• Sustainability Science

  (Interactions of human and environmental systems)

• Collected Papers

  (Editorials, Perspectives, Reviews, Colloquia, etc.)

• Special Features

  (Solicited articles on exceptional research topics)

• Sackler Colloquia

  (Scientific reports held under NAS auspices)

• Podcasts

  Brief podcasts with leading scientists