- Email: [email protected]
Gene pvuIIW: A possible modulator of PvuII endonuclease subunit association
Gene, 157 (1995) 193-199 © 1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50
193
GENE 08493
Gene pvuHW: a possible modulator of PvulI endonuclease subunit association* (Protein-protein interaction; restriction-modification system; regulation)
Gail M. Adams and Robert M. Blumenthal Department of Microbiology, Medical College of Ohio, Toledo, OH 43699-0008, USA Received by A.S. Bhagwat: 5 August 1994; Accepted: 30 September 1994; Received at publishers: 6 October 1994
SUMMARY
The PvulI restriction-modification system has been found to contain three genes which code for a DNA methyltransferase (MTase), a restriction endonuclease (ENase) and a small protein required for expression of the ENase-encoding gene. In addition, there is a small open reading frame (ORF) within and opposite to the MTase-encoding gene. The region containing this ORF is transcribed, and the ORF has an excellent Shine-Dalgarno sequence with an ATA start codon. A closely related ORF is present in the Sinai system. The 28-amino-acid (aa) predicted peptide from the PvulI ORF resembles a region of the PvulI ENase at the dimer interface. We have cloned this ORF, giving it an ATG start codon and putting it under the control of an inducible promoter: induction leads to a slight but significant decrease in restriction of bacteriophage ~. We also have obtained the 28-aa synthetic peptide, and are exploring the possibility that it modulates ENase subunit association. While this peptide has no detectible effect on dimeric PvuI1 ENase, it inhibits renaturation of urea-denatured ENase in a concentration-dependent manner. The ORF may represent an additional safeguard during establishment of the PvulI restriction-modification system in a new host cell, helping to delay the appearance of active ENase dimers, while the MTase accumulates and protects the host chromosome.
INTRODUCTION
Type-II R-M systems (RMS2s) are widespread among bacteria; and many RMS2 genes appear to be horizontally exchanged on plasmids. The expression of these Correspondence to: Dr. R.M. Blumenthal, Department of Microbiology, Medical College of Ohio, Toledo, OH 43699-0008, USA. Tel. (1-419) 381-5422; Fax ( 1-419) 381-3002; e-mail: [email protected] * Presented at the Third New England BioLabs Workshop on Biological DNA Modification, Vilnius, Lithuania, 22-28 May 1994. Abbreviations: aa, amino acid(s); BSA, bovine serum albumin; DMSO, dimethylsulfoxide; MTase (M.), DNA methyltransferase; ENase (R.), restriction endonuclease; HIV, human immunodeficiency virus; nt, nucleotide(s); ORF, open reading frame; pvullC and C.PvuII, gene and protein for the activator of pvullR expression; pvullM, gene encoding M.PvuII; pvulIR, gene encoding R.PvuII; pvuHW and W.PvuII, gene and peptide for the ORF opposite pvulIM; WS.PvuII, the synthetic W.PvuII peptide; RMS2, type-II R-M; R-M, restriction-modification system; TAE, 0.04 M Tris-acetate/0.002 M EDTA pH 7,5-7.8
SSDI 0378-1119(94)00704-7
RMS2 genes must be tightly controlled: methylation of the host chromosome must precede appearance of active ENase dimers. This control is especially important when an RMS2 is first introduced into a new host, but may also play an important role in recoveries from starvation or similar stresses. The PvuII RMS2 was isolated from the Gram- bacterium Proteus vulgaris (Gingeras et al., 1981), and its genes cloned from their native location on a plasmid (Blumenthal et al., 1985; Blumenthal, 1987; Calvin\Koons and Blumenthal, 1995). Three genes have been identified: pvulIM codes for the N4-methylcytosine DNA MTase (Tao et al., 1989), pvulIR codes for the homodimeric ENase (Athanasiadis et al., 1990; Tao and Blumenthal, 1991; 1992) and pvulIC specifies a regulatory protein required for expression ofpvulIR (Tao et al., 1991; Tao and Blumenthal, 1992). We currently view the role of pvuHC, and of other
194 similar genes (Brooks et al., 1991; Tao et al., 1991), as providing a delay in expression of the R gene. This does not preclude the possibility that additional genes play a role in delaying the appearance of ENase activity. We are exploring the role of a fourth O R F in the PvulI system, and present evidence that it may provide an additional regulatory mechanism acting to prevent premature appearance of ENase. The product of this O R F may be an inhibitor of PvuII ENase subunit association. If so, this would be the first reported example of a naturallyoccurring enzyme inhibitory peptide that inhibits subunit association, thus acting as an 'anti-chaperonin.'
and is probably expressed, we began to investigate its possible functions. First, is pvuHW unique? Several other RMS2s are related to PvuII in genetic organization and in their use of C-proteins such as the product of pvu11C (Tao et al., 1991). If pvulIW is playing a significant role then one might expect to find homologs. In fact, we found a very similar O R F opposite the M gene in the SmaI RMS2 (Fig. 1B). These two ORFs have 11 identities over 35 positions (31% identity), with three conservative substitutions and three gaps. As the PvuII and Sinai MTases have some sequence similarity to one another, it is important to note that the two W ORFs do not lie opposite analogous regions of the two M genes.
RESULTSAND DISCUSSION
(a) A fourth ORF is present in the PvuII RMS2 We have named the new O R F pvuHW, because it lies within pvulIM but has the opposite polarity (W vs. M). W'PvuII has a predicted Mr of 3035, an estimated pI of 3.77, and its gene is preceded by a strong Shine-Dalgarno sequence and an ATA start codon (Fig. 1A). Furthermore, transcript analyses, both in E. coli clones and in the native host P. vulgaris, indicate that this O R F begins about 40 nt from the 5' end of a long mRNA molecule that extends through pvulIC and pvulIR (Vijeseurier et al., 1993). Because this O R F is transcribed, A M >
MetMetThrLeuAsnLouGlnThzMetSezSerAsnASl~4etLeuAsn> ATGATGACTTTGAATCTACAGACTATGAGTAGCAATGATATGTTGAAT TACTACTGAAACTTAGATGTCTGATACTCATCGTTACTATACAACTTA
M >
PheGlyLysLysProAlsTyzThzThrSerAsnGlySorMotTyr~lo> 32
< W
TTTGGAAAAAAACCTGCCTACACAACTAGTAATGGGTCTATGTATATA AAACCTTTTTTTGGACGGATGTGTTGATCATTACCCAGATACATATAT <***ValValLeuLeuProAspIleTyrIle
M >
16 48
96
20
GiyAspSerLeuGluLeuLeuGluSerPhoProGluGluSerIleSer> GGTGACTCATTGGAGCTATTAGAATCATTCCCAGAAGAAAGTATTAGT CCACTGAGTAACCTCGATAATCTTAGTAAGGGTCTTCTTTCATAATCA < W ProSeEGluAsnSerSerAsnSezAsl~AsnGlySozSerLeuXleLeu
48 144
M >
LeuValMotThrSerProPzoPheAlaLeuGImAzgL¥sLysGluTyr> CTGGTTATGACTAGTCCTCCCTTCGCATTACAACGAAAAAAAGAATAC GACCA~Yu~CTGATCAGGAGGGAAGCGTAATGTTGCTTTTTTTCTTATG < W ArgThr(Met)
64 192
M >
80... 240
GlyAsnLeuGluGlnHisGluTyzV&iAspTrpPheLouSerPheAla> GGAAACTTAGAACAGCATGAGTATGTGGATTGGTTTTTGTCATTTGCT CCTTTGAATCTTGTCGTACTCATACACCTAACCAAAAACAGTAAACGA
4
B W-PvuII>
MTRL-ILSSG~DSNSSNES-PIYID
II
III
•
:.
II
I
I
..... PLLW*
II
W-SmaI>MEFGILFYRRLNILflLAREAYflSLflflZPTHSLflSflWQTTPVVPDP W.SmaI>KNGSRTFPPGREPANTHNLAMARGKTAK*
Fig. 1. The W.PvulI ORF. (A) Location and sequence of the W.PvulI ORF opposite the MTase gene,pvulIM (shown in part). Numbers refer to the aa positions. Potential Shine-Dalgarnoand promoter sequences are underlined.Asterisks (***)mark a stop codon; arrowheads (< and >) indicate gene orientations. (B) Comparison of the aa sequences of pvuHW and a similar ORF from the Sinai RMS2. Short vertical lines (J)indicateidentities,colons(:) indicatehighly-conservativesubstitutions and dots (.) indicate less-conservativesubstitutions Asterisks (*) mark the stop codons.
(b) W'PvulI resembles a portion of R'PvulI In searching for such W homologs, we found that pvullW resembles a segment of the PvuII ENase (Fig. 2A). The matched sequences have 10 identities over 28 positions (36% identity), with five conservative substitutions and two gaps. The structure of R.PvuII has been determined (Athanasiadis et al., 1994; Cheng et al., 1994). The region of the structure apparently mimicked by W'PvuII is a portion of the [~-sheet cluster that constitutes the catalytic core, and a contiguous loop that comes to within 3.7 ,& of the second subunit at the interface (Fig. 2A). If this sequence similarity is relevant, then one would expect the peptide could adopt a conformation similar to the corresponding portion of the ENase. To assess this, W.PvuII side chains were substituted into one subunit of the R'PvuII dimer structure where they differed, and the remainder of that subunit was deleted. This R'PvuIIW.PvuII complex was subjected to energy minimization, with the Ca atoms immobilized, using the program X-PLOR (BrOnger, 1992). The R.PvuII-proximal portion of the resulting W.PvuII structure is shown in spacefilling form (Fig. 2B). The optimized peptide structure can fit in the original position of the second ENase subunit with no van der Waals clashes, but some docking movement would allow the peptide tyrosine to fit more closely into a hydrophobic pocket (L 24, 131, F 32) and form a hydrogen bond to the carbonyl of N 29. This is consistent with the possibility that the peptide-ENase sequence similarity is structurally relevant. Accordingly, we proceeded to test the hypothesis that W.PvuII is an inhibitor of R'PvulI.
(c) Does W.PvuII inhibit the activity of R.PvulI? Peptides that inhibit ENase activity are known (e.g., Mark and Studier, 1981), and we tested W.PvuII for this ability in two ways. First, pvuHW was PCR amplified, given an ATG start codon and put downstream from the
195
PvuIIR
Helix A Loop AB Helix B (M)SHPD~NKLL ELWPHIQEYQ DLALK~IND IFQDN~GKLL QVLLI~GL~
PvuIIW
/ /,~ 2"
----Antiparallel ~ sheet-Apparent Mg++ contact PvuIIR PvuIIW
GNDAV I Ill LILSSGND--
/.,~/ I~I
active site
J_
DNAGQEYELK
I...I SNSSNES---
H - b o n d s same His on o t h e r subunit
. L,
I.
SIN~DL~KGF STHHHMN~VI IAKYRQVPWI... .I
il
PIYIDPLLVV Y/////~////~
Region shown as spacefilling in structure B.
Fig. 2. Sequence similarity between W.PvuII and R.PvulI and possible structural implications. (A) Sequence comparison. The symbols used are described in the legend to Fig. 1. The indicated close approach (3.7 A) is between the two subunits of the native ENase dimer. Helix A-Loop AB-Helix B constitutes the primary subunit interface in the native structure (Cheng et al., 1994). This is followed in the primary sequence by part of the l-sheet cluster that includes the sidechains critical for catalysis, and then by a loop (34) that includes the close contact described above. (B) The 6-aa region indicated in (A) was examined as described in the text. Briefly, the sidechains for one ENase subunit in the dimer crystal structure (Cheng et al., 1994) were substituted with the corresponding sidechains from W.PvuII. All other aa in the substituted subunit were eliminated. This was followed by energy minimization, with the positions of the s-carbons fixed, using the program X-PLOR (Brtinger, 1992). The result is shown with the intact ENase subunit as a ribbon diagram, and the 6-aa portion of W.PvuI1 in space-filling form, As shown, the N-terminal dimerization region of the ENase is at the right, the catalytic core is in the center, and the recognition region is to the left. In an active dimer, the second subunit would occupy the bottom half of the picture with its N terminus at the upper right. The arrow points to the loop similar in sequence to that portion of W.PvuII shown in space-filling form. i n d u c i b l e lac p r o m o t e r o n a h i g h - c o p y - n u m b e r v e c t o r ( p B l u e s c r i p t ; S t r a t a g e n e , L a Jolla, CA, U S A ) . We c o n f i r m e d t h a t the s e q u e n c e of the c l o n e d i n s e r t was u n c h a n g e d f r o m the wild-type. T h i s c l o n e was p u t i n t o a n E, coli s t r a i n ( J M 1 0 7 M A 2 ) w h i c h a l r e a d y c a r r i e d a pACYC184-derived p l a s m i d b e a r i n g the p v u H M C R genes. T h e p B l u e s c r i p t p l a s m i d s h a v e a c o p y n u m b e r o v e r tenfold h i g h e r t h a n pACYC184 ( M i n t o n , 1984), so b o t h
p r o m o t e r s t r e n g t h a n d gene d o s a g e s h o u l d h a v e i n c r e a s e d the a m o u n t of W.PvulI relative to R.PvulI t h o u g h a c t u a l a m o u n t s in the cell were n o t d e t e r m i n e d . I P T G i n d u c t i o n of p v u H W d i d n o t d e t e c t i b l y r e d u c e in vitro activity w h e n R . P v u I I was a s s a y e d in cell extracts (not shown). W h e n the i n d u c e d a n d u n i n d u c e d cells d e s c r i b e d a b o v e were c o m p a r e d for the a b i l i t y to restrict b a c t e r i o p h a g e ~, a small b u t statistically-significant effect
196 was seen (Table I). I n d u c t i o n of p v u H W led to a fivefold decrease in apparent restriction, and this effect was not seen in either induction of the vector control strain or with the use of M.PvuII-modified ~. O u r second a p p r o a c h was to prepare W'PvuII synthetically (Chiron Mimotopes). Circular dichroism spectra provided no evidence that the free peptide maintains stable secondary structure, either in peptide solvent (50% ethanol/0.1% acetic acid) or in neutral 50 m M phosphate buffer with or without 20% ethylene glycol (not shown). The synthetic 28-aa peptide, which we designate WS.PvuII, was preincubated with R.PvuII and then the ENase activity was assayed by digestion of bacteriophage ~, or plasmid D N A . An equivalent a m o u n t of the peptide solvent was added to control incubations. Adding up to a 121-fold m o l a r excess (66 ~tM) of WS'PvuII did not affect in vitro activity of R.PvuII, even when the activity assay was preceded by a 24 h coincubation with W ~.PvuII at 4°C (not shown). In other experiments, W ~.PvuII was added to the reaction as well as the co-incubation, to exclude the possibility that dilution into the reaction buffer led to rapid dissociation of inhibited complexes (not shown). These experiments showed no detectible inhibition of ENase activity relative to the control incubations lacking peptide. These results have several possible interpretations, of which two seem m o s t likely. First, W.Pvull m a y normally play no role in controlling R.PvuII activity, notwithstanding the small in vivo effect shown in Table I, t h o u g h it could be the r e m n a n t of a previously-used control system. Second, W.PvuII m a y only bind to m o n o m e r s of R.PvuII and prevent their dimerization. While gel-filtration c h r o m a t o g r a p h y of R'PvuII in 0.5
M NaC1 yielded h o m o d i m e r s as the only detectible active form (Tao and Blumenthal, 1992), the subunit-subunit Ka for R.PvulI has not yet been determined. This Kd appears to be quite low for the few ENases that have been characterized for multimerization constants ( M o d r i c h and Zabel, 1976; D ' A r c y et al., 1985; Luke et al., 1987; N a r d o n e and Chirikjian, 1987). If the R.PvulI dimer dissociation constant is comparable in magnitude, then the a m o u n t and appearance rate of free m o n o m e r would be minute under the conditions used here (approx. 0.5 pM ENase).
(d) Does WS.PvulI affect subunit association of R.PvulI? Peptides that p r o m o t e subunit association are k n o w n (Wagner and Green, 1993), and peptides that inhibit enzyme subunit association have been synthesized t h o u g h not yet found to occur naturally (Bab6 et al., 1992; Perry et al., 1992; K r o g s r u d et al., 1993). Some inhibitory synthetic peptides that mimicked a large proportion of the H I V protease subunit interface actually p r o m o t e d dimer dissociation, but most of the inhibitory peptides only prevented association of m o n o m e r s (Bab~ et al., 1992). If W'PvulI is m o d u l a t i n g subunit association at all, the results described in section c suggest it is acting in this latter manner. Testing this would require a population of R.PvulI monomers. O u r a p p r o a c h was to find denaturation/renaturation conditions for R.PvulI, and we began with the urea denaturation/renaturation conditions reported for subunit association studies with the H I V protease (Bab6 et al., 1992). In contrast to the protease, R.PvulI is quite stable and appears to refold and reassociate very quickly with no special treatment when the urea is diluted out.
TABLE I In vivo effect of pvuIIW induction on restriction of bacteriophage ?~ Plasmida
)wir (unmod)b
Xvir (mod)b
1
2
-- IPTG
+ IPTGc
- IPTG
+ IPTGc
(none)
(none)
ND
pBluescript-KS
pPvuRM 184
pPvuWKS
(2.2+0.1) x 101° (1.00) (4.5 + 0.4) x 10 9 (2.1×10 -1 ) ( 1.8 + 0.0) X 101° (8.2X10 1)
ND
pPvuRM184
(8.5 +_0.2)z 109 (1.00) (3.9 + 0.2) x 105 (4.6x10 -s) (4.8 ± 0.2) × l0 s (5.6X10 5)
( 1.3 + 0.1 ) x 10s (1.4x10 5) (2.4 +_0.2) x 10 6 (2.8X10-4)
± 0.3) × 109 (1.7z10 -1 ) ( 1.3 ± 0.0) × 10l° (5.8X10 1) (3.7
a The host strain is E. coli JM107MA2 (Blumenthal et al., 1985). Plasmid pPvuRM184 is a pACYC184-based clone of the intact PvulI RMS2. The two plasmid-bearing strains differ in that one also contains pBluescript (Stratagene), while the other contains a pBluescript clone of the PCR-amplified pvullW gene under the control of Ptac-(pPvuWKS). b The numbers indicate apparent phage titers + standard error, with normalized values (relative efficiency of plating) shown in parentheses. Each determination was made in triplicate. Each strain was grown in LB medium containing carbenicillin and tetracycline, dilutions of the phage were adsorbed for 20 min, and the mix was then plated onto LB agar in )~ top agar (media are as described in Sambrook et al. 1989). Unmodified ~ vir was passaged through JM107MA2, while modified Xvir was passaged through JM107MA2 [pPvuMI.9] and is thus protected against restriction by the PvulI REase. ND, not determined. c When added, IPTG was present at a concentration of 1 mM.
197 I n c o n t r o l experiments, using varying c o n c e n t r a t i o n s of peptide solvent ( b u t n o peptide) in the diluent, the extent of r e n a t u r a t i o n was consistently 2 0 - 2 5 %
(recovered/
i n p u t activity units, assayed with pBluescriptKS D N A ) . N
m
W h e n R ' P v u I I was first d e n a t u r e d a n d then allowed to
".i~ 1 MM D peptidc - ' . . ' ~ ' ~
r e n a t u r e with WS.PvuII added to the diluent, the peptide decreased r e n a t u r a b l e activity in a d o s e - d e p e n d e n t
#M WS'Pwll
m a n n e r (Fig. 3). These results are expressed as the disappearance of u n c u t D N A , so a reduction in E N a s e activity appears as a rise in the slope (to a less-negative value). The results of these experiments indicate that the inhibi%
S
10
15
20
25
30
35
40
tory effects of W ~ ' P v u l I are half-maximal at approx.
Time, rain. -8o f I:
°
.
.
.
I
.
10 n M , a n d s a t u r a t i n g at 0.5 pM. The control ' D ' peptide, which like W . P v u I I is quite acidic, has n o i n h i b i t o r y effect
.
_~ooF /
o
,
of ENase m o n o m e r in these experiments (assuming 100%
V
~oer~ ~OC Ere,
o'~ .
-140
,.y.
-160
at tested c o n c e n t r a t i o n s below 1.5 ~tM. The c o n c e n t r a t i o n
o - - w~.Pval
o
o=,,e
[
.
.
~
m o n o m e r i z a t i o n in the urea) was just u n d e r 4 n M . The m a x i m a l extent of i n h i b i t i o n observed was 2-3-fold
..................
( 7 - 1 2 % recovery of activity, vs. 2 0 - 2 5 % recovery in the solvent controls). It is i m p o r t a n t to note that we have n o t yet d e t e r m i n e d the m e a n s by which WS.PvuI1 inhibits r e n a t u r a t i o n of
-180 I
-200
I
o.s
I
~
~.s
I
I
I
2
2.s
3
R.PvuI]. It r e m a i n s possible that W~.PvuII was acting in some way other t h a n by p r e v e n t i n g dimer reassociation.
[peptlde], C
-2000
]
,
•
,
i
.
.
.
.
i
•
,
•
,
i
•
,
,
,
i
,
,
,
,
(e) A rationale for W'PvulI as a subunit association inhibitor
1 ./~,
'~' ,.-., -4000
We have yet to directly d e m o n s t r a t e b i n d i n g between
I'1
~, "8 '~" -6ooo
]
W . P v u I I a n d a m o n o m e r of R.PvuII. Nonetheless, what
might be the physiological rationale for a protein that could inhibit R ' P v u I I s u b u n i t association? First, the pro-
.i.a "8 ~ '~ -~ooo
,000o "~ - 12000
posed regulatory goal for both C-PvuII a n d W . P v u I I is to delay the a p p e a r a n c e of active R.PvuII dimers. It would be c o u n t e r p r o d u c t i v e for a n R M S 2 to produce a potent i n h i b i t o r of ENase activity unless its synthesis was
(I
I: .....
ooo- , , '","
soo
I ooo
i soo
'ooo
carefully timed, but the t r a n s c r i p t i o n data (Vijeseurier
[W s" P ~ II], nM
Fig. 3. Effectof W'.PvuII on recovery of R.PvulI activity during renaturation. R.PvuII (40 units; 0.27 pmol) was added to a 10 111(final vol.) denaturation mix containing 7 M urea/10% DMSO/22 mM KHPO4 (pH 7.0). After 1 min at 55'C, 130 lal of reaction mix was added (3-5 lag pBR322 or pBluescript DNA; final concentrations of 100lag BSA per ml/20 mM Tris-HCl/5 mM MgCI2/50 mM KC1, pH 7.4), including various concentrations of of either W'.PvuII, the control 'D' peptide, or the peptide solvent alone (50% ethanol/0.1% acetic acid). The tubes were incubated at 37°C, and 20-lalportions of the reaction were removed at various times for electrophoresis on 1 or 1.5% agarose gels in TAE buffer. The 'D' peptide has the aa sequence QGDLVRKLKEEKAPEIDIKKAVAELKTRKKILEDKE. Control reactions included equivalent amounts of peptide solvent. Photographic negatives of the ethidium-stained gels were subjected to scanning densitometry, and intensities of the uncleaved substrate were plotted vs. time for each concentration of peptide. The slopes resulting from linear regression were plotted vs. the concentration of peptide added. Note that an increase in slope (less negative)is associated with reduced ENase activity. (A) Measuring R.PvuII activity following renaturation in the
presence or absence of Ws.PvulI: sample data. The slopes and correlation coefficients resulting from linear regression were: - 152 (0.99) for the solvent control, -154 (0.98) for D peptide and -108 (0.99) for W"PvuII. These slopes, and those from the other tested peptide concentrations, were used to generate the graph in (B). (B) Effect of WS.Pvu[I and a control peptide on the recovery of R.Pvull activity. The rate of disappearance of the uncut plasmid pBR322 (one PvulI site) is shown plotted against the various concentrations of either the Ws.PvuII peptide, or the control 'D' peptide. In this experiment, a recovery/densitometry standard (linearized pBluescript SK) was added to each quenched reaction before loading the gel, and band densities were corrected accordingly. (C) Effect of W'.PvuII on the recovery of R.PvuII activity. The rate of disappearance of uncut pBluescriptKS (two PvuIl sites) is shown vs. the various concentrations of Ws.PvuII peptide. The vertical bars reflect the correlation coefficientsfrom linear regressionscomparable to those illustrated in (A) [bar length=( 1-R)*lslopel, where R is the correlation coefficient and IslopeJ refers to the absolute value].
198 et al., 1993) indicate that p v u l I W continues to be transcribed even after the PvuII R M S 2 is established. O u r current understanding of PvuII ENase expression is that the long transcript which includes pvullC and p v u H W continues t h r o u g h pvuHR (Vijeseurier et al., 1993). The ENase appears to be poorly expressed from this long transcript, however. A p r o m o t e r closer to the E N a s e gene is p r o p o s e d to become active only after accumulation of C.PvuII (Tao and Blumenthal, 1992). While translation of pvulIR from the long transcript is apparently poor, there m a y be e n o u g h ENase m o n o m e r made for an inhibitory peptide to be useful by blocking dimerization. In this way, p v u l I W and pvulIC would be acting in parallel to delay ENase accumulation until the host D N A could be methylated. The crystal structures for several ENases have been solved, and while the enzymes have very little actual aa similarity, they do share some structural similarities: the catalytic d o m a i n is structurally conserved a m o n g ENases as different as PvuIl, EcoRI, EcoRV and BamHI (see Cheng et al., 1994). In contrast, the subunit interface domains responsible for dimerization are distinctly different from one another. It is thus n o t e w o r t h y that W.PvuII resembles the sequence of, and m a y act at, a portion of the structurally conserved catalytic d o m a i n of R.PvuII which happens to lie at the subunit interface.
(f) Conclusions (1) We have found an O R F in the PvuII RMS2, and are interested in it for three reasons. First, the O R F is definitely transcribed, and has g o o d signals for translation initiation. Second, it has a h o m o l o g in at least one other R M S 2 (Sinai). Third, its aa sequence resembles that of a region of R.PvulI. (2) Ws'PvuII did not appear to inhibit the activity of intact R.PvulI dimers in vitro, but overexpression in vivo led to a fivefold decrease in apparent restriction. (3) W h e n R.PvulI was first denatured and then allowed to renature in the presence of Ws.PvuII, the peptide appeared to decrease renaturable" activity in a concentration-dependent manner. The degree of inhibition was half maximal at a peptide concentration of a b o u t 10 nM, under the conditions used. (4) In order to clearly establish a role for W.PvulI, we must demonstrate physical interaction between W.PvuII and R.PvulI, and demonstrate that W.PvulI is synthesized from its native expression elements in E. coli and P. vulgaris.
ACKNOWLEDGEMENTS We are grateful to X i a o d o n g Cheng (Cold Spring H a r b o r L a b o r a t o r y ) for providing the R.PvuII structure
coordinates, and to C a t h y D r e n n a n and Eric F a u m a n (University of Michigan) for help with the p r o g r a m X - P L O R and with structure plots. David D i g n a m (Medical College of Ohio) kindly provided the ' D ' peptide, as well as m u c h helpful advice. The C D spectra were obtained with the generous help of Michael O g a w a and Alexei Gretchikhine (Bowling Green State University). This report is based u p o n work supported by the N a t i o n a l Science F o u n d a t i o n under grant D M B 9205248. G.M.A. presented this work at the Third International W o r k s h o p on Biological D N A Methylation, and her travel to that meeting was made possible, in part, by generous support from New England BioLabs.
REFERENCES Athanasiadis, A., Gregoriu, M., Thanos, D., Kokkinidis, M. and Papamatheakis, J.: Complete nucleotide sequence of the PvulI restriction enzyme gene from Proteus vulgaris. Nucleic Acids Res. 18 (1990) 6434. Athanasiadis, A., Vlassi, M., Kotsifaki, D., Tucker, P.A., Wilson, K.S. and Kokkinidis, M.: Crystal structure of PvuII endonuclease reveals extensive structural homologies to EcoRV. Struct. Biol. 1 (1994) 469-475. Bab6, L.M., Ros6, J. and Craik, C.S.: Synthetic "interface" peptides alter dimeric assembly of the HIV 1 and 2 proteases. Protein Sci. 1 (1992) 1244 1263. Blumenthal, R.M.: The PvulI restriction-modification system: cloning, characterization and use in revealing in E. coli barrier to certain methylases or methylated DNAs. Elsevier/North-Holland, Amsterdam, 1987. Blumenthal, R.M., Gregory, S.A. and Cooperider, J.S.: Cloning of a restriction-modification system from Proteus vulgaris and its use in analyzing a methylase-sensitive phenotype in Escherichia coll. J. Bacteriol. 164 (1985) 501-509. Brooks, J.E., Nathan, P.D., Landry, D., Sznyter, L.A., Waite-Rees, P., Ives, C.L., Moran, L.S., Slatko, B.E. and Benner, J.S.: Characterization of the cloned BamHI restriction modification system: its nucleotide sequence, properties of the methylase, and expression in heterologous hosts. Nucleic Acids Res. 19 (1991) 841 850. Br0nger, A.T.: X-PLOR, vet. 3.1: a system for x-ray crystallography and NMR. Yale University, New Haven, CT, 1992. Calvin\Koons, M.D. and Blumenthal, R.M,: Characterization of pPvu1, the autonomous plasmid from Proteus vulgaris that carries the genes for the PvulI restriction-modification system. Gene 157 (1995) 73-79. Cheng, X., Balendiran, K., Schildkraut, I. and Anderson, J.E.: Structure of PvuII endonuclease with cognate DNA. EMBO J. 13 (1994) 3927-3935. D'Arcy, A., Brown, R.S., Zabeau, M., Van Resandt, R.W. and Winkler, F.K.: Purification and crystallization of the EcoRV restriction endonuclease. J. Biol. Chem. 260 (1985) 1987-1990. Gingeras, T.R., Greenough, L., Schildkraut, I. and Roberts, R.J.: Two new restriction endonucleases from Proteus vulgaris. Nucleic Acids Res. 9 (1981) 4525-4536. Krogsrud, R.L., Welchner, E., Scouten, E. and Liuzzi, M.: A solid-phase assay for the binding of peptidic subunit association inhibitors to the Herpes simplex virus ribonucleotide reductase large subunit. Anal. Biochem. 213 (1993) 386-394.
199 Luke, P.A., McCallum, S.A. and Halford, S.E.: The EcoRV restriction endonuclease. Gene Amplif. Anal. 5 (1987) 185-207. Mark, K. and Studier, F.W.: Purification of the gene 0.3 protein of bacteriophage T7, an inhibitor of the DNA restriction system of Escherichia coli. J. Biol. Chem. 256 (1981) 3573 3578. Minton, N.P.: Improved plasmid vectors for the isolation of translational lac gene fusions. Gene 31 (1984) 269-273. Modrich, P.A. and Zabel, D.: EcoRI endonuclease: physical and catalytic properties of the homogeneous enzyme. J. Biol. Chem. 251 (1976) 5866 5874. Nardone, G. and Chirikjian, J.G.: The enzymes of the BamH l restriction-modification system. Gene Amplif. Anal. 5 (1987) 147-184. Perry, K.M., Pookanjanatavip, M., Zhao, J., Santi, D.V. and Stroud, R.M.: Reversible dissociation and unfolding of the dimeric protein thymidylate synthase. Protein Sci. 1 (1992) 796-800. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Press, Cold Spring Harbor, NY, 1989. Tao, T. and Blumenthal, R.M.: Sequence and subunit structure of restriction endonuclease PvuII. ASM Abstr. 91 (1991) 175.
Tao, T. and Blumenthal, R.M.: Sequence and characterization of pvuHR, the PvuII endonuclease gene, and of pvullC, its regulatory gene. J. Bacteriol. 174 (1992) 3395-3398. Tao, T., Bourne, J.C. and Blumenthal, R.M.: A family of regulatory genes associated with type II restriction-modification systems. J. Bacteriol. 173 (1991) 1367-1375. Tao, T., Walter, J., Brennan, K.J., Cotterman, M.M. and Blumenthal, R.M.: Sequence, internal homology and high-level expression of the gene for a DNA-(cytosine N4)-methyltransferase, M.PvuII. Nucleic Acids Res. 17 (1989) 4161 4175. Vijeseurier, R.M., Blumenthal, R.M. and Dunbar, J.C.: Regulation of type II restriction-modification systems: the pvulIC gene appears to code for a transcriptional activator of pvullR. Abstr. FASEB Suture. Res. Conf. Restriction Endonucleases and Modification Methyltransferases: Structures and Mechanisms, Saxtons River, VT, 1993, p. 38. Wagner, S. and Green, M.R.: HTLV-1 Tax protein stimulation of DNA binding of bZIP proteins by enhancing dimerization. Science 262 (1993) 395 399.
193
GENE 08493
Gene pvuHW: a possible modulator of PvulI endonuclease subunit association* (Protein-protein interaction; restriction-modification system; regulation)
Gail M. Adams and Robert M. Blumenthal Department of Microbiology, Medical College of Ohio, Toledo, OH 43699-0008, USA Received by A.S. Bhagwat: 5 August 1994; Accepted: 30 September 1994; Received at publishers: 6 October 1994
SUMMARY
The PvulI restriction-modification system has been found to contain three genes which code for a DNA methyltransferase (MTase), a restriction endonuclease (ENase) and a small protein required for expression of the ENase-encoding gene. In addition, there is a small open reading frame (ORF) within and opposite to the MTase-encoding gene. The region containing this ORF is transcribed, and the ORF has an excellent Shine-Dalgarno sequence with an ATA start codon. A closely related ORF is present in the Sinai system. The 28-amino-acid (aa) predicted peptide from the PvulI ORF resembles a region of the PvulI ENase at the dimer interface. We have cloned this ORF, giving it an ATG start codon and putting it under the control of an inducible promoter: induction leads to a slight but significant decrease in restriction of bacteriophage ~. We also have obtained the 28-aa synthetic peptide, and are exploring the possibility that it modulates ENase subunit association. While this peptide has no detectible effect on dimeric PvuI1 ENase, it inhibits renaturation of urea-denatured ENase in a concentration-dependent manner. The ORF may represent an additional safeguard during establishment of the PvulI restriction-modification system in a new host cell, helping to delay the appearance of active ENase dimers, while the MTase accumulates and protects the host chromosome.
INTRODUCTION
Type-II R-M systems (RMS2s) are widespread among bacteria; and many RMS2 genes appear to be horizontally exchanged on plasmids. The expression of these Correspondence to: Dr. R.M. Blumenthal, Department of Microbiology, Medical College of Ohio, Toledo, OH 43699-0008, USA. Tel. (1-419) 381-5422; Fax ( 1-419) 381-3002; e-mail: [email protected] * Presented at the Third New England BioLabs Workshop on Biological DNA Modification, Vilnius, Lithuania, 22-28 May 1994. Abbreviations: aa, amino acid(s); BSA, bovine serum albumin; DMSO, dimethylsulfoxide; MTase (M.), DNA methyltransferase; ENase (R.), restriction endonuclease; HIV, human immunodeficiency virus; nt, nucleotide(s); ORF, open reading frame; pvullC and C.PvuII, gene and protein for the activator of pvullR expression; pvullM, gene encoding M.PvuII; pvulIR, gene encoding R.PvuII; pvuHW and W.PvuII, gene and peptide for the ORF opposite pvulIM; WS.PvuII, the synthetic W.PvuII peptide; RMS2, type-II R-M; R-M, restriction-modification system; TAE, 0.04 M Tris-acetate/0.002 M EDTA pH 7,5-7.8
SSDI 0378-1119(94)00704-7
RMS2 genes must be tightly controlled: methylation of the host chromosome must precede appearance of active ENase dimers. This control is especially important when an RMS2 is first introduced into a new host, but may also play an important role in recoveries from starvation or similar stresses. The PvuII RMS2 was isolated from the Gram- bacterium Proteus vulgaris (Gingeras et al., 1981), and its genes cloned from their native location on a plasmid (Blumenthal et al., 1985; Blumenthal, 1987; Calvin\Koons and Blumenthal, 1995). Three genes have been identified: pvulIM codes for the N4-methylcytosine DNA MTase (Tao et al., 1989), pvulIR codes for the homodimeric ENase (Athanasiadis et al., 1990; Tao and Blumenthal, 1991; 1992) and pvulIC specifies a regulatory protein required for expression ofpvulIR (Tao et al., 1991; Tao and Blumenthal, 1992). We currently view the role of pvuHC, and of other
194 similar genes (Brooks et al., 1991; Tao et al., 1991), as providing a delay in expression of the R gene. This does not preclude the possibility that additional genes play a role in delaying the appearance of ENase activity. We are exploring the role of a fourth O R F in the PvulI system, and present evidence that it may provide an additional regulatory mechanism acting to prevent premature appearance of ENase. The product of this O R F may be an inhibitor of PvuII ENase subunit association. If so, this would be the first reported example of a naturallyoccurring enzyme inhibitory peptide that inhibits subunit association, thus acting as an 'anti-chaperonin.'
and is probably expressed, we began to investigate its possible functions. First, is pvuHW unique? Several other RMS2s are related to PvuII in genetic organization and in their use of C-proteins such as the product of pvu11C (Tao et al., 1991). If pvulIW is playing a significant role then one might expect to find homologs. In fact, we found a very similar O R F opposite the M gene in the SmaI RMS2 (Fig. 1B). These two ORFs have 11 identities over 35 positions (31% identity), with three conservative substitutions and three gaps. As the PvuII and Sinai MTases have some sequence similarity to one another, it is important to note that the two W ORFs do not lie opposite analogous regions of the two M genes.
RESULTSAND DISCUSSION
(a) A fourth ORF is present in the PvuII RMS2 We have named the new O R F pvuHW, because it lies within pvulIM but has the opposite polarity (W vs. M). W'PvuII has a predicted Mr of 3035, an estimated pI of 3.77, and its gene is preceded by a strong Shine-Dalgarno sequence and an ATA start codon (Fig. 1A). Furthermore, transcript analyses, both in E. coli clones and in the native host P. vulgaris, indicate that this O R F begins about 40 nt from the 5' end of a long mRNA molecule that extends through pvulIC and pvulIR (Vijeseurier et al., 1993). Because this O R F is transcribed, A M >
MetMetThrLeuAsnLouGlnThzMetSezSerAsnASl~4etLeuAsn> ATGATGACTTTGAATCTACAGACTATGAGTAGCAATGATATGTTGAAT TACTACTGAAACTTAGATGTCTGATACTCATCGTTACTATACAACTTA
M >
PheGlyLysLysProAlsTyzThzThrSerAsnGlySorMotTyr~lo> 32
< W
TTTGGAAAAAAACCTGCCTACACAACTAGTAATGGGTCTATGTATATA AAACCTTTTTTTGGACGGATGTGTTGATCATTACCCAGATACATATAT <***ValValLeuLeuProAspIleTyrIle
M >
16 48
96
20
GiyAspSerLeuGluLeuLeuGluSerPhoProGluGluSerIleSer> GGTGACTCATTGGAGCTATTAGAATCATTCCCAGAAGAAAGTATTAGT CCACTGAGTAACCTCGATAATCTTAGTAAGGGTCTTCTTTCATAATCA < W ProSeEGluAsnSerSerAsnSezAsl~AsnGlySozSerLeuXleLeu
48 144
M >
LeuValMotThrSerProPzoPheAlaLeuGImAzgL¥sLysGluTyr> CTGGTTATGACTAGTCCTCCCTTCGCATTACAACGAAAAAAAGAATAC GACCA~Yu~CTGATCAGGAGGGAAGCGTAATGTTGCTTTTTTTCTTATG < W ArgThr(Met)
64 192
M >
80... 240
GlyAsnLeuGluGlnHisGluTyzV&iAspTrpPheLouSerPheAla> GGAAACTTAGAACAGCATGAGTATGTGGATTGGTTTTTGTCATTTGCT CCTTTGAATCTTGTCGTACTCATACACCTAACCAAAAACAGTAAACGA
4
B W-PvuII>
MTRL-ILSSG~DSNSSNES-PIYID
II
III
•
:.
II
I
I
..... PLLW*
II
W-SmaI>MEFGILFYRRLNILflLAREAYflSLflflZPTHSLflSflWQTTPVVPDP W.SmaI>KNGSRTFPPGREPANTHNLAMARGKTAK*
Fig. 1. The W.PvulI ORF. (A) Location and sequence of the W.PvulI ORF opposite the MTase gene,pvulIM (shown in part). Numbers refer to the aa positions. Potential Shine-Dalgarnoand promoter sequences are underlined.Asterisks (***)mark a stop codon; arrowheads (< and >) indicate gene orientations. (B) Comparison of the aa sequences of pvuHW and a similar ORF from the Sinai RMS2. Short vertical lines (J)indicateidentities,colons(:) indicatehighly-conservativesubstitutions and dots (.) indicate less-conservativesubstitutions Asterisks (*) mark the stop codons.
(b) W'PvulI resembles a portion of R'PvulI In searching for such W homologs, we found that pvullW resembles a segment of the PvuII ENase (Fig. 2A). The matched sequences have 10 identities over 28 positions (36% identity), with five conservative substitutions and two gaps. The structure of R.PvuII has been determined (Athanasiadis et al., 1994; Cheng et al., 1994). The region of the structure apparently mimicked by W'PvuII is a portion of the [~-sheet cluster that constitutes the catalytic core, and a contiguous loop that comes to within 3.7 ,& of the second subunit at the interface (Fig. 2A). If this sequence similarity is relevant, then one would expect the peptide could adopt a conformation similar to the corresponding portion of the ENase. To assess this, W.PvuII side chains were substituted into one subunit of the R'PvuII dimer structure where they differed, and the remainder of that subunit was deleted. This R'PvuIIW.PvuII complex was subjected to energy minimization, with the Ca atoms immobilized, using the program X-PLOR (BrOnger, 1992). The R.PvuII-proximal portion of the resulting W.PvuII structure is shown in spacefilling form (Fig. 2B). The optimized peptide structure can fit in the original position of the second ENase subunit with no van der Waals clashes, but some docking movement would allow the peptide tyrosine to fit more closely into a hydrophobic pocket (L 24, 131, F 32) and form a hydrogen bond to the carbonyl of N 29. This is consistent with the possibility that the peptide-ENase sequence similarity is structurally relevant. Accordingly, we proceeded to test the hypothesis that W.PvuII is an inhibitor of R'PvulI.
(c) Does W.PvuII inhibit the activity of R.PvulI? Peptides that inhibit ENase activity are known (e.g., Mark and Studier, 1981), and we tested W.PvuII for this ability in two ways. First, pvuHW was PCR amplified, given an ATG start codon and put downstream from the
195
PvuIIR
Helix A Loop AB Helix B (M)SHPD~NKLL ELWPHIQEYQ DLALK~IND IFQDN~GKLL QVLLI~GL~
PvuIIW
/ /,~ 2"
----Antiparallel ~ sheet-Apparent Mg++ contact PvuIIR PvuIIW
GNDAV I Ill LILSSGND--
/.,~/ I~I
active site
J_
DNAGQEYELK
I...I SNSSNES---
H - b o n d s same His on o t h e r subunit
. L,
I.
SIN~DL~KGF STHHHMN~VI IAKYRQVPWI... .I
il
PIYIDPLLVV Y/////~////~
Region shown as spacefilling in structure B.
Fig. 2. Sequence similarity between W.PvuII and R.PvulI and possible structural implications. (A) Sequence comparison. The symbols used are described in the legend to Fig. 1. The indicated close approach (3.7 A) is between the two subunits of the native ENase dimer. Helix A-Loop AB-Helix B constitutes the primary subunit interface in the native structure (Cheng et al., 1994). This is followed in the primary sequence by part of the l-sheet cluster that includes the sidechains critical for catalysis, and then by a loop (34) that includes the close contact described above. (B) The 6-aa region indicated in (A) was examined as described in the text. Briefly, the sidechains for one ENase subunit in the dimer crystal structure (Cheng et al., 1994) were substituted with the corresponding sidechains from W.PvuII. All other aa in the substituted subunit were eliminated. This was followed by energy minimization, with the positions of the s-carbons fixed, using the program X-PLOR (Brtinger, 1992). The result is shown with the intact ENase subunit as a ribbon diagram, and the 6-aa portion of W.PvuI1 in space-filling form, As shown, the N-terminal dimerization region of the ENase is at the right, the catalytic core is in the center, and the recognition region is to the left. In an active dimer, the second subunit would occupy the bottom half of the picture with its N terminus at the upper right. The arrow points to the loop similar in sequence to that portion of W.PvuII shown in space-filling form. i n d u c i b l e lac p r o m o t e r o n a h i g h - c o p y - n u m b e r v e c t o r ( p B l u e s c r i p t ; S t r a t a g e n e , L a Jolla, CA, U S A ) . We c o n f i r m e d t h a t the s e q u e n c e of the c l o n e d i n s e r t was u n c h a n g e d f r o m the wild-type. T h i s c l o n e was p u t i n t o a n E, coli s t r a i n ( J M 1 0 7 M A 2 ) w h i c h a l r e a d y c a r r i e d a pACYC184-derived p l a s m i d b e a r i n g the p v u H M C R genes. T h e p B l u e s c r i p t p l a s m i d s h a v e a c o p y n u m b e r o v e r tenfold h i g h e r t h a n pACYC184 ( M i n t o n , 1984), so b o t h
p r o m o t e r s t r e n g t h a n d gene d o s a g e s h o u l d h a v e i n c r e a s e d the a m o u n t of W.PvulI relative to R.PvulI t h o u g h a c t u a l a m o u n t s in the cell were n o t d e t e r m i n e d . I P T G i n d u c t i o n of p v u H W d i d n o t d e t e c t i b l y r e d u c e in vitro activity w h e n R . P v u I I was a s s a y e d in cell extracts (not shown). W h e n the i n d u c e d a n d u n i n d u c e d cells d e s c r i b e d a b o v e were c o m p a r e d for the a b i l i t y to restrict b a c t e r i o p h a g e ~, a small b u t statistically-significant effect
196 was seen (Table I). I n d u c t i o n of p v u H W led to a fivefold decrease in apparent restriction, and this effect was not seen in either induction of the vector control strain or with the use of M.PvuII-modified ~. O u r second a p p r o a c h was to prepare W'PvuII synthetically (Chiron Mimotopes). Circular dichroism spectra provided no evidence that the free peptide maintains stable secondary structure, either in peptide solvent (50% ethanol/0.1% acetic acid) or in neutral 50 m M phosphate buffer with or without 20% ethylene glycol (not shown). The synthetic 28-aa peptide, which we designate WS.PvuII, was preincubated with R.PvuII and then the ENase activity was assayed by digestion of bacteriophage ~, or plasmid D N A . An equivalent a m o u n t of the peptide solvent was added to control incubations. Adding up to a 121-fold m o l a r excess (66 ~tM) of WS'PvuII did not affect in vitro activity of R.PvuII, even when the activity assay was preceded by a 24 h coincubation with W ~.PvuII at 4°C (not shown). In other experiments, W ~.PvuII was added to the reaction as well as the co-incubation, to exclude the possibility that dilution into the reaction buffer led to rapid dissociation of inhibited complexes (not shown). These experiments showed no detectible inhibition of ENase activity relative to the control incubations lacking peptide. These results have several possible interpretations, of which two seem m o s t likely. First, W.Pvull m a y normally play no role in controlling R.PvuII activity, notwithstanding the small in vivo effect shown in Table I, t h o u g h it could be the r e m n a n t of a previously-used control system. Second, W.PvuII m a y only bind to m o n o m e r s of R.PvuII and prevent their dimerization. While gel-filtration c h r o m a t o g r a p h y of R'PvuII in 0.5
M NaC1 yielded h o m o d i m e r s as the only detectible active form (Tao and Blumenthal, 1992), the subunit-subunit Ka for R.PvulI has not yet been determined. This Kd appears to be quite low for the few ENases that have been characterized for multimerization constants ( M o d r i c h and Zabel, 1976; D ' A r c y et al., 1985; Luke et al., 1987; N a r d o n e and Chirikjian, 1987). If the R.PvulI dimer dissociation constant is comparable in magnitude, then the a m o u n t and appearance rate of free m o n o m e r would be minute under the conditions used here (approx. 0.5 pM ENase).
(d) Does WS.PvulI affect subunit association of R.PvulI? Peptides that p r o m o t e subunit association are k n o w n (Wagner and Green, 1993), and peptides that inhibit enzyme subunit association have been synthesized t h o u g h not yet found to occur naturally (Bab6 et al., 1992; Perry et al., 1992; K r o g s r u d et al., 1993). Some inhibitory synthetic peptides that mimicked a large proportion of the H I V protease subunit interface actually p r o m o t e d dimer dissociation, but most of the inhibitory peptides only prevented association of m o n o m e r s (Bab~ et al., 1992). If W'PvulI is m o d u l a t i n g subunit association at all, the results described in section c suggest it is acting in this latter manner. Testing this would require a population of R.PvulI monomers. O u r a p p r o a c h was to find denaturation/renaturation conditions for R.PvulI, and we began with the urea denaturation/renaturation conditions reported for subunit association studies with the H I V protease (Bab6 et al., 1992). In contrast to the protease, R.PvulI is quite stable and appears to refold and reassociate very quickly with no special treatment when the urea is diluted out.
TABLE I In vivo effect of pvuIIW induction on restriction of bacteriophage ?~ Plasmida
)wir (unmod)b
Xvir (mod)b
1
2
-- IPTG
+ IPTGc
- IPTG
+ IPTGc
(none)
(none)
ND
pBluescript-KS
pPvuRM 184
pPvuWKS
(2.2+0.1) x 101° (1.00) (4.5 + 0.4) x 10 9 (2.1×10 -1 ) ( 1.8 + 0.0) X 101° (8.2X10 1)
ND
pPvuRM184
(8.5 +_0.2)z 109 (1.00) (3.9 + 0.2) x 105 (4.6x10 -s) (4.8 ± 0.2) × l0 s (5.6X10 5)
( 1.3 + 0.1 ) x 10s (1.4x10 5) (2.4 +_0.2) x 10 6 (2.8X10-4)
± 0.3) × 109 (1.7z10 -1 ) ( 1.3 ± 0.0) × 10l° (5.8X10 1) (3.7
a The host strain is E. coli JM107MA2 (Blumenthal et al., 1985). Plasmid pPvuRM184 is a pACYC184-based clone of the intact PvulI RMS2. The two plasmid-bearing strains differ in that one also contains pBluescript (Stratagene), while the other contains a pBluescript clone of the PCR-amplified pvullW gene under the control of Ptac-(pPvuWKS). b The numbers indicate apparent phage titers + standard error, with normalized values (relative efficiency of plating) shown in parentheses. Each determination was made in triplicate. Each strain was grown in LB medium containing carbenicillin and tetracycline, dilutions of the phage were adsorbed for 20 min, and the mix was then plated onto LB agar in )~ top agar (media are as described in Sambrook et al. 1989). Unmodified ~ vir was passaged through JM107MA2, while modified Xvir was passaged through JM107MA2 [pPvuMI.9] and is thus protected against restriction by the PvulI REase. ND, not determined. c When added, IPTG was present at a concentration of 1 mM.
197 I n c o n t r o l experiments, using varying c o n c e n t r a t i o n s of peptide solvent ( b u t n o peptide) in the diluent, the extent of r e n a t u r a t i o n was consistently 2 0 - 2 5 %
(recovered/
i n p u t activity units, assayed with pBluescriptKS D N A ) . N
m
W h e n R ' P v u I I was first d e n a t u r e d a n d then allowed to
".i~ 1 MM D peptidc - ' . . ' ~ ' ~
r e n a t u r e with WS.PvuII added to the diluent, the peptide decreased r e n a t u r a b l e activity in a d o s e - d e p e n d e n t
#M WS'Pwll
m a n n e r (Fig. 3). These results are expressed as the disappearance of u n c u t D N A , so a reduction in E N a s e activity appears as a rise in the slope (to a less-negative value). The results of these experiments indicate that the inhibi%
S
10
15
20
25
30
35
40
tory effects of W ~ ' P v u l I are half-maximal at approx.
Time, rain. -8o f I:
°
.
.
.
I
.
10 n M , a n d s a t u r a t i n g at 0.5 pM. The control ' D ' peptide, which like W . P v u I I is quite acidic, has n o i n h i b i t o r y effect
.
_~ooF /
o
,
of ENase m o n o m e r in these experiments (assuming 100%
V
~oer~ ~OC Ere,
o'~ .
-140
,.y.
-160
at tested c o n c e n t r a t i o n s below 1.5 ~tM. The c o n c e n t r a t i o n
o - - w~.Pval
o
o=,,e
[
.
.
~
m o n o m e r i z a t i o n in the urea) was just u n d e r 4 n M . The m a x i m a l extent of i n h i b i t i o n observed was 2-3-fold
..................
( 7 - 1 2 % recovery of activity, vs. 2 0 - 2 5 % recovery in the solvent controls). It is i m p o r t a n t to note that we have n o t yet d e t e r m i n e d the m e a n s by which WS.PvuI1 inhibits r e n a t u r a t i o n of
-180 I
-200
I
o.s
I
~
~.s
I
I
I
2
2.s
3
R.PvuI]. It r e m a i n s possible that W~.PvuII was acting in some way other t h a n by p r e v e n t i n g dimer reassociation.
[peptlde], C
-2000
]
,
•
,
i
.
.
.
.
i
•
,
•
,
i
•
,
,
,
i
,
,
,
,
(e) A rationale for W'PvulI as a subunit association inhibitor
1 ./~,
'~' ,.-., -4000
We have yet to directly d e m o n s t r a t e b i n d i n g between
I'1
~, "8 '~" -6ooo
]
W . P v u I I a n d a m o n o m e r of R.PvuII. Nonetheless, what
might be the physiological rationale for a protein that could inhibit R ' P v u I I s u b u n i t association? First, the pro-
.i.a "8 ~ '~ -~ooo
,000o "~ - 12000
posed regulatory goal for both C-PvuII a n d W . P v u I I is to delay the a p p e a r a n c e of active R.PvuII dimers. It would be c o u n t e r p r o d u c t i v e for a n R M S 2 to produce a potent i n h i b i t o r of ENase activity unless its synthesis was
(I
I: .....
ooo- , , '","
soo
I ooo
i soo
'ooo
carefully timed, but the t r a n s c r i p t i o n data (Vijeseurier
[W s" P ~ II], nM
Fig. 3. Effectof W'.PvuII on recovery of R.PvulI activity during renaturation. R.PvuII (40 units; 0.27 pmol) was added to a 10 111(final vol.) denaturation mix containing 7 M urea/10% DMSO/22 mM KHPO4 (pH 7.0). After 1 min at 55'C, 130 lal of reaction mix was added (3-5 lag pBR322 or pBluescript DNA; final concentrations of 100lag BSA per ml/20 mM Tris-HCl/5 mM MgCI2/50 mM KC1, pH 7.4), including various concentrations of of either W'.PvuII, the control 'D' peptide, or the peptide solvent alone (50% ethanol/0.1% acetic acid). The tubes were incubated at 37°C, and 20-lalportions of the reaction were removed at various times for electrophoresis on 1 or 1.5% agarose gels in TAE buffer. The 'D' peptide has the aa sequence QGDLVRKLKEEKAPEIDIKKAVAELKTRKKILEDKE. Control reactions included equivalent amounts of peptide solvent. Photographic negatives of the ethidium-stained gels were subjected to scanning densitometry, and intensities of the uncleaved substrate were plotted vs. time for each concentration of peptide. The slopes resulting from linear regression were plotted vs. the concentration of peptide added. Note that an increase in slope (less negative)is associated with reduced ENase activity. (A) Measuring R.PvuII activity following renaturation in the
presence or absence of Ws.PvulI: sample data. The slopes and correlation coefficients resulting from linear regression were: - 152 (0.99) for the solvent control, -154 (0.98) for D peptide and -108 (0.99) for W"PvuII. These slopes, and those from the other tested peptide concentrations, were used to generate the graph in (B). (B) Effect of WS.Pvu[I and a control peptide on the recovery of R.Pvull activity. The rate of disappearance of the uncut plasmid pBR322 (one PvulI site) is shown plotted against the various concentrations of either the Ws.PvuII peptide, or the control 'D' peptide. In this experiment, a recovery/densitometry standard (linearized pBluescript SK) was added to each quenched reaction before loading the gel, and band densities were corrected accordingly. (C) Effect of W'.PvuII on the recovery of R.PvuII activity. The rate of disappearance of uncut pBluescriptKS (two PvuIl sites) is shown vs. the various concentrations of Ws.PvuII peptide. The vertical bars reflect the correlation coefficientsfrom linear regressionscomparable to those illustrated in (A) [bar length=( 1-R)*lslopel, where R is the correlation coefficient and IslopeJ refers to the absolute value].
198 et al., 1993) indicate that p v u l I W continues to be transcribed even after the PvuII R M S 2 is established. O u r current understanding of PvuII ENase expression is that the long transcript which includes pvullC and p v u H W continues t h r o u g h pvuHR (Vijeseurier et al., 1993). The ENase appears to be poorly expressed from this long transcript, however. A p r o m o t e r closer to the E N a s e gene is p r o p o s e d to become active only after accumulation of C.PvuII (Tao and Blumenthal, 1992). While translation of pvulIR from the long transcript is apparently poor, there m a y be e n o u g h ENase m o n o m e r made for an inhibitory peptide to be useful by blocking dimerization. In this way, p v u l I W and pvulIC would be acting in parallel to delay ENase accumulation until the host D N A could be methylated. The crystal structures for several ENases have been solved, and while the enzymes have very little actual aa similarity, they do share some structural similarities: the catalytic d o m a i n is structurally conserved a m o n g ENases as different as PvuIl, EcoRI, EcoRV and BamHI (see Cheng et al., 1994). In contrast, the subunit interface domains responsible for dimerization are distinctly different from one another. It is thus n o t e w o r t h y that W.PvuII resembles the sequence of, and m a y act at, a portion of the structurally conserved catalytic d o m a i n of R.PvuII which happens to lie at the subunit interface.
(f) Conclusions (1) We have found an O R F in the PvuII RMS2, and are interested in it for three reasons. First, the O R F is definitely transcribed, and has g o o d signals for translation initiation. Second, it has a h o m o l o g in at least one other R M S 2 (Sinai). Third, its aa sequence resembles that of a region of R.PvulI. (2) Ws'PvuII did not appear to inhibit the activity of intact R.PvulI dimers in vitro, but overexpression in vivo led to a fivefold decrease in apparent restriction. (3) W h e n R.PvulI was first denatured and then allowed to renature in the presence of Ws.PvuII, the peptide appeared to decrease renaturable" activity in a concentration-dependent manner. The degree of inhibition was half maximal at a peptide concentration of a b o u t 10 nM, under the conditions used. (4) In order to clearly establish a role for W.PvulI, we must demonstrate physical interaction between W.PvuII and R.PvulI, and demonstrate that W.PvulI is synthesized from its native expression elements in E. coli and P. vulgaris.
ACKNOWLEDGEMENTS We are grateful to X i a o d o n g Cheng (Cold Spring H a r b o r L a b o r a t o r y ) for providing the R.PvuII structure
coordinates, and to C a t h y D r e n n a n and Eric F a u m a n (University of Michigan) for help with the p r o g r a m X - P L O R and with structure plots. David D i g n a m (Medical College of Ohio) kindly provided the ' D ' peptide, as well as m u c h helpful advice. The C D spectra were obtained with the generous help of Michael O g a w a and Alexei Gretchikhine (Bowling Green State University). This report is based u p o n work supported by the N a t i o n a l Science F o u n d a t i o n under grant D M B 9205248. G.M.A. presented this work at the Third International W o r k s h o p on Biological D N A Methylation, and her travel to that meeting was made possible, in part, by generous support from New England BioLabs.
REFERENCES Athanasiadis, A., Gregoriu, M., Thanos, D., Kokkinidis, M. and Papamatheakis, J.: Complete nucleotide sequence of the PvulI restriction enzyme gene from Proteus vulgaris. Nucleic Acids Res. 18 (1990) 6434. Athanasiadis, A., Vlassi, M., Kotsifaki, D., Tucker, P.A., Wilson, K.S. and Kokkinidis, M.: Crystal structure of PvuII endonuclease reveals extensive structural homologies to EcoRV. Struct. Biol. 1 (1994) 469-475. Bab6, L.M., Ros6, J. and Craik, C.S.: Synthetic "interface" peptides alter dimeric assembly of the HIV 1 and 2 proteases. Protein Sci. 1 (1992) 1244 1263. Blumenthal, R.M.: The PvulI restriction-modification system: cloning, characterization and use in revealing in E. coli barrier to certain methylases or methylated DNAs. Elsevier/North-Holland, Amsterdam, 1987. Blumenthal, R.M., Gregory, S.A. and Cooperider, J.S.: Cloning of a restriction-modification system from Proteus vulgaris and its use in analyzing a methylase-sensitive phenotype in Escherichia coll. J. Bacteriol. 164 (1985) 501-509. Brooks, J.E., Nathan, P.D., Landry, D., Sznyter, L.A., Waite-Rees, P., Ives, C.L., Moran, L.S., Slatko, B.E. and Benner, J.S.: Characterization of the cloned BamHI restriction modification system: its nucleotide sequence, properties of the methylase, and expression in heterologous hosts. Nucleic Acids Res. 19 (1991) 841 850. Br0nger, A.T.: X-PLOR, vet. 3.1: a system for x-ray crystallography and NMR. Yale University, New Haven, CT, 1992. Calvin\Koons, M.D. and Blumenthal, R.M,: Characterization of pPvu1, the autonomous plasmid from Proteus vulgaris that carries the genes for the PvulI restriction-modification system. Gene 157 (1995) 73-79. Cheng, X., Balendiran, K., Schildkraut, I. and Anderson, J.E.: Structure of PvuII endonuclease with cognate DNA. EMBO J. 13 (1994) 3927-3935. D'Arcy, A., Brown, R.S., Zabeau, M., Van Resandt, R.W. and Winkler, F.K.: Purification and crystallization of the EcoRV restriction endonuclease. J. Biol. Chem. 260 (1985) 1987-1990. Gingeras, T.R., Greenough, L., Schildkraut, I. and Roberts, R.J.: Two new restriction endonucleases from Proteus vulgaris. Nucleic Acids Res. 9 (1981) 4525-4536. Krogsrud, R.L., Welchner, E., Scouten, E. and Liuzzi, M.: A solid-phase assay for the binding of peptidic subunit association inhibitors to the Herpes simplex virus ribonucleotide reductase large subunit. Anal. Biochem. 213 (1993) 386-394.
199 Luke, P.A., McCallum, S.A. and Halford, S.E.: The EcoRV restriction endonuclease. Gene Amplif. Anal. 5 (1987) 185-207. Mark, K. and Studier, F.W.: Purification of the gene 0.3 protein of bacteriophage T7, an inhibitor of the DNA restriction system of Escherichia coli. J. Biol. Chem. 256 (1981) 3573 3578. Minton, N.P.: Improved plasmid vectors for the isolation of translational lac gene fusions. Gene 31 (1984) 269-273. Modrich, P.A. and Zabel, D.: EcoRI endonuclease: physical and catalytic properties of the homogeneous enzyme. J. Biol. Chem. 251 (1976) 5866 5874. Nardone, G. and Chirikjian, J.G.: The enzymes of the BamH l restriction-modification system. Gene Amplif. Anal. 5 (1987) 147-184. Perry, K.M., Pookanjanatavip, M., Zhao, J., Santi, D.V. and Stroud, R.M.: Reversible dissociation and unfolding of the dimeric protein thymidylate synthase. Protein Sci. 1 (1992) 796-800. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Press, Cold Spring Harbor, NY, 1989. Tao, T. and Blumenthal, R.M.: Sequence and subunit structure of restriction endonuclease PvuII. ASM Abstr. 91 (1991) 175.
Tao, T. and Blumenthal, R.M.: Sequence and characterization of pvuHR, the PvuII endonuclease gene, and of pvullC, its regulatory gene. J. Bacteriol. 174 (1992) 3395-3398. Tao, T., Bourne, J.C. and Blumenthal, R.M.: A family of regulatory genes associated with type II restriction-modification systems. J. Bacteriol. 173 (1991) 1367-1375. Tao, T., Walter, J., Brennan, K.J., Cotterman, M.M. and Blumenthal, R.M.: Sequence, internal homology and high-level expression of the gene for a DNA-(cytosine N4)-methyltransferase, M.PvuII. Nucleic Acids Res. 17 (1989) 4161 4175. Vijeseurier, R.M., Blumenthal, R.M. and Dunbar, J.C.: Regulation of type II restriction-modification systems: the pvulIC gene appears to code for a transcriptional activator of pvullR. Abstr. FASEB Suture. Res. Conf. Restriction Endonucleases and Modification Methyltransferases: Structures and Mechanisms, Saxtons River, VT, 1993, p. 38. Wagner, S. and Green, M.R.: HTLV-1 Tax protein stimulation of DNA binding of bZIP proteins by enhancing dimerization. Science 262 (1993) 395 399.