Printed in USA - PDF document

1 Printed in U.S.A. Molecular Cloning and
Printed in U.S.A. Molecular Cloning and DNA Sequence of ZucE, the Gene Encoding the Lactose-specific Enzyme II of the Phosphotransferase System of Lactobacillus casei EVIDENCE THAT A CYSTEINE RESIDUE IS ESSENTIAL FOR SUGAR PHOSPHORYLATION* (Received for publication, July 24, 1990) Carl-Alfred Alpert+ and Bruce M. Chassys From the Laboratory for Microbial Ecology, National Institute for Dental Research, National Institutes of Health, Bethesda, Maryland 20892 The gene coding for the lactose-specific Enzyme II of the Lactobacillus casei phosphoenolpyruvate-de- “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Supported by a grant from Deutsche Forschungsgemeinschaft. Present address: Geseilschaft fiir Biotechnologische F&&hung, Mas- cheroder Wee 1, 3300 Braunschwein. Federal Reoublic of Germanv. § To whom correspndence should be addressed: Dept. of Food Science, ABL 103, 1302 W. Pennsylvania Ave., University of Illinois, Urbana, IL 61801. ’ The abbreviations used are: PTS, phosphoenolpyruvate sugar- phosphotransferase svstem: PEP. uhosDhoenoluvruvate: EI. Enzvme i of the PTS; HPr, heat stable p;~tein;~FIII’“‘, &tose-&e&c fa”ctor III or enzyme III of the PTS: EII’“‘. lac@, gene coding for FIII’“‘; lacG, gene coding for P-P-gal; SDS, sodium dodecyl sulfate; SDS- PAGE, SDS-polyacrylamide gel electrophoresis. 1968; McKay, 1982; McKay, 1983; Premi et al., 1972; Chassy et al., 1978). The overall reaction for the phosphorylation and transport of lactose can be written as follows (Stein et al., 1974, Schrecker et al., 1975): Mg’+ PEP + Enzyme I - P - Enzyme I + Pyruvate P - Enzyme I + HPr -+ Enzyme I + P - HPr of EIILmL.,,,i and Robillard, 1988; Erni and Zanolari, 1986; Schnetz et al., 1987). The Enzymes II from Gram-positive organisms have proven m

2 ore resistant to characterization. Purif
ore resistant to characterization. Purification of EII’“’ from S. aureus resulted in complete loss of enzymatic activity. The purified protein was used to raise antisera, but little biochemical insight into mechanism and structure was gained (Schafer et al., 1981). The gene encoding the EIIlaC from S. aureus has recently been isolated by molecular cloning and its DNA sequence reported (Breidt et al., 1987). Little is known about the secondary or tertiary structure of the pro- tein, about the Cloning and of the EII’“‘-The previously re- ported sequence of the 1acF gene was followed by a putative p-independent transcription terminator (Alpert and Chassy, 1988) which was interpreted as possibly signaling the end of a transcriptional unit. The previously published sequence of the region 5’ to the 1acG (P-p-gal) gene contained a partial URF (Porter and Chassy, 1988); in L. casei 1acG is followed immediately by 1acF. However, the lactose-specific genes of S. aureus occur in the order lucF-lacE-1acG (Breidt et al., 1987). An identical order has recently been reported to occur in Lactococcus lactis (de Vos et al., 1990). We wanted to evaluate the presence of a similar arrangement in L. casei, since it appeared likely that lacE would precede, rather than follow, lucF in L. casei. Therefore, of this paper (including “Materials and Methods”) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press. - URF E II ‘X P-p-gal ---- 1 I URF F Ill ‘kc 1 kb plZ616 pU617 FIG. 1. Molecular cloning and sequencing strategy for the EIIlaCL.e~aei. The restriction endonuclease cutting site map on the line depicts the region of pLZ64 which encodes the

3 lactose-PTS genes (Alpert and Chassy, 1
lactose-PTS genes (Alpert and Chassy, 1988; Porter and Chassy, 1988). The reading frames encoded by this fragment are shown below the map. For clarity, the region encoding EII’“L.,,~, discussed in this paper, is also shown. Molecular cloning of the EcoRI-PstI fragment in pUC18 resulted in pLZ618, cloning of the PstI-Hind111 fragment in pUC18 gave rise to pLZ617 which was used in the experiments reported in this paper. The bold arrows shown above the expanded representation of the P&I-Hind111 fragment in the lower portion of the figure represent sequences derived from subclones; the bold arrows below the line represent tides as primers for sequencing reactions. The sequences represented by bold arrows were used in the compilation of the composite sequence of the region. frame directly to the start codon in the first reading frame. A similar arrangement is also found in a fifth case, where FIII’SC~.,,,; is preceded by another ATG codon, forming the sequence ATGATG. The importance of this finding is not yet clear, since only a few genes from this species are known. However, this motif (ATGA) also occurs at the start codons of the three lactose specific genes in S. aureus (Breidt et al., 1987). Remarkably, initiation codons followed at the 3’ end by the base A have been shown to increase E. coli ribosome binding to I RBS and start sequences of six characterized genes of the species Lactobacilh.5 The abbreviations used are: DHFR, dihydrofolate reductase; His-Decarb, histidine decarboxylase; D-Hit-DH, D- 2-hydroxy-isocaproate dehydrogenase. Sequence around start RBS sequence Bases between RBS and start Ref. EII’“” AAGCGCTCAGGAGGAAAAGACTC ATGA 9 This paper P-@-gal CACTTAACAGGAGGTTATTAAGCA ATGA 10 Porter and Chassy (1988) FIII’“’ GATTGAATCGGAGGGAAA ATGATG” 7 Alpert and Chassy (1988) DHFR A

4 TACTCAAAGGAGGGGTCTCGA ATGA 8 Andrews e
TACTCAAAGGAGGGGTCTCGA ATGA 8 Andrews et al. (1985) His-Decarb. TTTTTATTAGGAGGTCTAATT ATGT 7 Vanderslice et al. (1986) D-Hit-DH AATTGGAAAGGAAGTTTAACAC ATGA 8 Lerch et al. (1989) a The actual start codon is presumably (based on 70% represents a highly hydro- phobic polypeptide, while the remaining 30% COOH-terminal region constitutes a more hydrophilic domain (Figs. 2 3). This is in agreement with the analysis of the EII’“’ of S. aureus and L. la&s which have hydrophobic NHz-terminal regions of similar size and calculated molecular masses of 62,688 and 61,562 Da, respectively (Breidt et al., 1987; de Vos et al., 1990). In EII’aC~.caTer, seven potential a-helical membrane-spanning domains of hydropathy index � 1.3 and length � 18 residues were observed (Fig. 3). Further study will be necessary to determine the actual conformation of these regions and the mechanism of insertion of EII’aC~.eaqer into the membrane. The present model of the lac-PTS operon in L. casei con- sists of the lacE-lacG-lacF genes, followed by a 22564 LacE and the Active Center Cysteine of EII’““L.,,,,, FIG. 2. DNA sequence of lacE. The DNA sequence of the PstI- Hind111 fragment in pLZ617 from bases 1 to 2100 is shown. The sequence beyond base 1828 has been published previously (Porter and Chassy, 1988). An ORF, lacE, starts with ATG at position 349 runs larities (Chassy and Alpert, 1989). The only proteins related I d 20 b ‘lb0 40 FIN. 3. Hydropathy analysis of EII’“‘,C(I,p, by the method of Kyte and Doolittle (1982). The NH,-terminal Lanes l-7 show assays of membranes isolated from mutants His”“+Arg”“, His”X+ArgZ;“, His.,“‘~Arg.L”~, His.V’:+Arg.““, HiS”‘7-Leu’l”, HipI* L Leu”“, and His”“‘-Leu.&#

5 147;‘, respectively. TABLE II Polyp
147;‘, respectively. TABLE II Polypeptlde sequence .sLmLlarLtLes between the finzymesI1’“’ Sequences IdentIty [‘/o] Conservative [%I L. casei/S. aureus tantly related to the EII“” from S. aureus and L. lactis, while these two proteins appeared to be more similar to each other than to the L. casei protein (Table II and Vos et al., 1990). This finding was evidenced by the higher percentage of iden- of EII?~.ca.sei L.casei s. aureus S.lacti.3 Consensus L.casei S.tl"G5"S S.lactis consensus L.casei 1 * . ..MNKV... FDKLKPVFEA IAANKYISAI RDGFIACMPI MTMMQKLIAQ IEKGKPFFEK LSRNIYLRAI RDGFISAMF'V . ..MHKLIEL IEKGKPFFEK ISRNIYLRAI RDGFIAGMPV ---M-K---- --K-Kp-FE- ---N-Y--AI mGFI--m- 61 WPDNVTNTLM VAYNYSMGLL ALFVAGTTAK NLTDSKNLEL WDKGMEAILM KPYNYTMGLV AFLVAGTTAK SLTDSFNRKL TPYSYSMGIL AFFVGGTTAK ALTDSKNRDL W-------m --y-y-M&- A--V-GTTM -LTDS-N--L 121 * ILSILPLKTG VDL.TYMGTQ GLICAYIVGL IVFNIYYVCI S. aureus FLASDPAKDG GFLSAFMGTK GLLTAFLSAF VTVIVYNFCV LMAAEPAKEG GFLTAFMGTK GLLTAFIAAF VTVNVYKVCV Consensus -----p-K-G emL---MGT- GLeeA----- -----y--C- 181 L. casei KDLIPMGLSV TAFWLFGVGF KAATGTVLPR WIIQVLSPLF S. aureus KDLIPFSAVI IILYALDLCI RNSFKSNVAE GILKLFEPLF S.lactis KDLIPFTVSV VLLYGLELLV KGTLGVTVAE SIGTLIAPLF Consensus m~Ip----- _--_______ --____-___ -I-----pLF FIG. 6. Alignment of the ELI’“” Proteins from L. casei, S. aureus, and L. lactis (Breidt et al., 1987; DeVos et al.. 1990). The sequences were repetitively aligned with the GAP program of the GCG program package until no further gaps were introduced. + marks histine residues 267,270,457, and in the EII’BcL CU-‘PI. * indicates cysteine residues L.casei S. aureus S.lactis Consensus L.casei S. aureus S.lactis Consensus L.casei S.aureus S.lactis Consensus L.casei S. aureus S.lactis Consensus L. casei S. aureus S.lactis Consensus L. casei * +

6 t FCGVQGPSIV QPAWPIMIA NTAANLQQYQ AGQHVS
t FCGVQGPSIV QPAWPIMIA NTAANLQQYQ AGQHVSHVLA FVGIHGPSIV EPAIAAITYA NIEANFKLLQ AGEHADKIIT FVGIHGPSIV EPAIAAITYA NIDVNLHLIQ AGQHADKVIT F-G--GpSIV -PA---I--A N--N----Q AG-H---m-- 300 MNTMDYVMNF GGTGATLVVP SGTQMFIVTF GGTGATLDVP SGTQMFIATM GGTGATLIVP --T------- GGTGATL-VP 301 * 360 FIMLFAARSA QLKAVGKAAF VPCTFGVNEP VLFGMF'IIMN PMLFIPFLAT PIVNVCLFKF FMFMWMTKSK RNKAIGRASV VPTFFGVNEP ILFGAPLVLN PVFFIPFVLA PIVNVWIFKL FLFMWICKSD RNRAIGRASV VPTFFGVNEP ILFGAPIVLN PIFFVPFIFA PIVNVWIFKF F-------S- ---A-G-A-- "p--FGmEP NFNTAKA... L-EE------ -____--___ ---___---_ ------A--- + t DA....THVL PETAPSAHGE DSILAASGVS .DDAAKASNI DAVLGKADVA KEDVAANNNI D-------V- _---__---- 481 * AYFKQNEVDV LVLCAGGGTS GILANALNKL SKERGLKLSA T.. . .EQTNV LVLCAGGGTS GLLANALNKA AEEYHVPVKA T....KETNV LVLCAGGGTS GLLANALNKA AAFXNVPVKA ---------V LVLCAGGGTS G-LANALNK- --E------A 540 AARAYGQDMD LIKDMNMVIL AAGGYGAHMD IMKEYQLIIL AAGGYGAHRE MLPEFDLVIL &A--yG---- APQMESMKGN LKKITDKYGV KLVTTTGRQY IELTNNGDMA 590 LDFVESNL.. 22565 60 IIFSSIFMMV AYVPNAWGFY ILFSSIFLLI AYVPNIFGFK ILFSSIFILI AYVE'NAWGFH I-FSSIF--- AYVPN--GF- 120 PKTNQINPVA VIVASEISFV ESTNQINFIS TMQAAMCGFL PATNQINFLS TMLASMVGFL -TNQIN-- ---A----F- 180 KNNVTIKLPE QVPGNIAQSF KRNITIKMPK EVPPNISQVF KNNVTIRMPE DVPPNISQVF K-N-TI--P- -VP-NI-Q-F 240 QASDSYLGLA LIAGAMAFFW TAADGWIGYT IIFGAFALFW SAADGYLGIT LIFGAYAFFW -A-D---G-- -I-GA-A-FW S.aureus APQVASNYED IKQDTDRLGI KLAKTQGAEY IKLTRDGQAA LDFVQQQFEN S.lactis APQVASNFDD MKAETDKLGI KLVKTEGAQY IKLTRDGQGA LAFVQQQFD. Consensus ApQ--s--s- -K--T,,-+- KL--T-G--y I-LT--G--A L-FV------ tical amino acids in the sequences of these proteins when sources. It may also be that all Enzymes11 do not share a compared to the of the Active Center Phosphorylated Amino phospho-p-D-glucosidase encoded by the E. coli bgl (p-gluco- Acid Residue-The cytoplasmic PTS proteins have been quite side) operon has been noted (Porter and Chass

7 y, 1988). It has extensively studied. Th
y, 1988). It has extensively studied. Their phosphoryl group-carrying residues recently been reported that Factor III”’ (for cellobiose and have been identified as either N-l or N-3 phosphohistidines arbutin) appears to share an ancestral relationship with Fac- by NMR studies and by isolation and sequencing of 22566 LacE and the Active Center Cysteine of EII’ac~,.,,,, TABLE III Histidine replacement mutants of EII”’ The codons for histidine were mutagenized to obtain proteins with a single amino acid replacement/polypeptide. * mutants obtained but assayed, + mutants assayed and positive for lactose phosphoryl- ation. Ala Arg Asn CYS Gln GlU GUY Ile Leu LYS Pro Thr . + . + . 1234567891011 FIG. 7. Autoradiograph of paper chromatographic analysis of the activity of cysteine mutants. Lanes I through 7 show assays of membranes prepared from mutants, where Cys”‘, CYS’“~, Cy8”, Cy$:,“, ,Tys:,,“, Cys:M”, and Cys4X:’ were replaced with a serine residue. Lanes 8 show the assays of TABLE IV Lactose-phosphorylation by mutant EII’“’ proteins Washed E. call membranes from cells carrying plasmids coding for EII’“‘,, ,(,\(., proteins with the indicated amino acid replacing one cys- teine/polypeptide were assayed for [‘%]lactose-phosphorylation ac- tivity (see “Materials and Methods”). + lactose phosphorylation was observed, - no lactose phosphorylation was obs&ved. Ser + -0 Thr + Pro + Ala + His - Tyr - “Five independently obtained clones from two different experi- ments were assayed. EII m”““‘L”’ actually contains two phosphorylation sites (Pas et al., 1988a, 1988b). According to a proposal by Saier et al., (1985) one of these aureus and S. carnosus (Reiche et al., 1988). The presence of the phosphohistidine,

8 sequence similarities of the phosphoryl
sequence similarities of the phosphorylated peptides, and III= 419 Val Ile Ala Q.4 Ile Ser $Zi& Ala II"'9 408 Ala Il.3 Asp Ala Q.f, Ile Thr Arg Leu II@= 417 As" LEU Asp Ala &a Ile Thr Arg LB" 11gvt 454 Ala Gin Ala m Asp Phe Tie Pro FIG. 8. Alignment of the proposed active centers of EII1ac~.e~e~ and EIIm”g,,u with those proposed for other En- zymes II by Pas and Robillard (Pas and Robillard, 1988a; Erni et al., 1987; Bramley and Kornberg, 1987; Rogers et al., 1988; Erni and Zanolari, 1986; Yamada and Saier, 1987). Identical residues extended sequence similarities of the EIII”“““““’ from S. au- reus and S. carnosus to the COOH-terminal part of the EIImannito’ from E. coli, all support the theory of an FIII domain being contained in the E11”“““‘““‘. The second phosphorylation site was concluded to be the site at which the actual transfer of the phosphoryl group from the protein to the sugar takes place. The phosphoryl group- carrying residue in the corresponding isolated peptide was identified as a cysteine, implying for the first time that an amino acid other than histidine functions in the sugar phorylations by the PTS (Pas and Robillard, 1988). Obviously, it would be important to determine whether order to identify possible active center residues, the three known up with corresponding histidine residues in the other two Enzymes II’“‘. HisZG7 is the last residue at the end of a region of scattered identities reaching from position 218 to position 274 in the alignment,. Neither HisZ7”, His4”7, nor His4”7 fall into regions of similarity to the other proteins. Similarly, of the 7 cysteine residues in EII’RC,,,,,,i, there is only one in the EII’“’ alignment that aligns lanes 1-4). In addition, we assayed a s