The molecular cloning, sequence and expression of the hdcB gene from Lactobacillus 30A

259

Gene, 85 (1989) 259-265 Elsevier GENE 03320

The molecular cloning, sequence and expression of the hdcB gene from Lactobacillus 30A (Recombinant DNA; histidine decarboxylase; transcriptional regulation; inducible operon; Gram + promoter)

William C. Copeland*, John D. Domena and Jon D. Robertus ClaytonFoundationBiochemicalInstituteand Departmentof Chemistty,Universityof Texas at Austin, Awtin, TX 78712 (U.S.A.) Received by S.R. Kushner: 25 June 1989 Revised: 21 August 1989 Accepted: 30 August 1989

SUMMARY

We previously cloned the structural gene ticA, which encodes the enzyme estate decarboxylase (HDC; EC 4.1.1.22), from ~crobaci~l~ 30a and found what appeared to be the start of a second gene 59 nucleotide (nt) downstream from the hdcA stop codon [Vanderslice et al., J. Biol. Chem. 32 (1986) 15186-151911. Here we report the complete nt sequence of this second gene, which we have named /z&B, and show that it encodes a 20-kDaprotein, HDCB, which was purified from Escherichiu coli. The hdcA and hdcB genes together comprise au operon, the transcription from which is shown to be increased threefold by the presence of histiclme in the &rowth medium. Western blots were used to qua&ate the rise in concentrations of both gene products during histidine induction of the Mc operon. This increase was found to be pro~~ion~ to the observed threefold increase in the concentration of the respective mRNAs. Transcription of the Adc operon in the mutant-3 strain of Lactobucilius 30a [ Recsei and Snell, Biochemistry 12 (1973) 365-3713 was shown to be constitutively 15-fold greater than in uninduced wild type cells and was unaffected by histidine. The transcription start point was defined as a guanine 73 nt 5’ to the start codon of the hdcA gene. Of the transcripts initiated at this promoter, 15 % include both h&A and hdcB sequences, the remainder terminate in the intergenic region and thus encode only hdcA.

Co~espondence to: Dr. J.D. Robertus, Department of Chemistry, University of Texas at Austin, Austin, TX 78712 (U.S.A.) Tel. f512)471-3175; Fax(512)471-86%. * Present address: Department of Pathology, Stanford University, Stanford, CA 94305 (U.S.A.) Tel. (415)725-4908.

Abbreviations: aa, amino acid(s); bp, base pair(s); buffer A, 50 mM Tris . HCl/l mM D’ITjl mM EDTA pH 8.0; DTT,

~~io~eitol; ELISA. enzyme-linked i~~osorbent assay; HDC, histidine decarboxylase (EC 4.1.1.22); J&4, gene encoding HDC (previously designated bdc); HDCB, protein encoded by the hdc3 gene; kb, kilobasefs) or 1000 bp; L30a, Lactobacillus 30a; oligo, oligodeoxyribonucleotide; ORF, open reading frame; TN buffer, 50 mM Tris *HCl pH 7.4/200 mM NaCl; tsp, transcription start point; wt, wild type.

0378-I119/89/$03.500 1989Else&r Science Publishers B.V.(Biomedicai Division)

260

INTRODUCTION

Lactobacillus 30a (L30a), isolated from the digestive tract of the horse (Rodwell, 1953), contains an enzyme, HDC, which decarboxylates histidine to yield hist~~e. The physiolo~c~ significance of this process is unknown, but it may be that histamine secretion enables the bacterium to regulate some aspect of its environment in an advantageous way. The enzymology of HDC, including active-site formation within the inactive proenzyme by an apparently autocatalytic polypeptide cleavage, is quite unusual (Recsei and Snell, 1984). Recsei and Snell(1982) demonstrated that the specific activity of the enzyme increases fourfold in response to histidine in the growth medium. They also created a mutant strain of L30a (mutant-3) by chemical mutagenesis which constitutively overproduced a mutated form of the enzyme (Recsei and Snell, 1973; 1982). To dissect the autoactivation and catalytic mechanisms of this unique enzyme, we previously cloned and sequenced its structural gene, h&l (Vanderslice et al., 1986), expressed it in E. coli (Copeland et al., 1987), and created several sitedirected mutants (Vanderslice et al., 1988). The nt sequence 5’ to the hdcA gene was compared with the 5’ region of 29 other genes cloned from Grampositive organisms (Graves and Rabinowitz, 1986). This comparison revealed that the hdcA upstream region contained elements typical of those found at positions -35 and -10 relative to tsp in a number of other genes. From these similarities, the guanine residue 73 nt 5’ to the methionine start codon was inferred to be the tsp for the hdcA gene. The start of a second ORF, called hdcB, was noted downstream from hdcA. This ORF began with an AUG located 59 nt 3’ to the end of the hdcA coding region and extended for 67 aa to the end of the restriction fragment under study (Vanderslice et al., 1986). A potential Shine-Dalgamo sequence was present in the intergenie region, but no recognizable promoter elements were found. The HDC and the presumed downstream protein were therefore proposed to be components of an operon, the transcription of which was regulated by the histidine concentration in the growth medium. We here describe the complete nt sequence of the downstream ORF, which we have pro~sion~ly designated h&B in keeping with the nomenclature of

Demerec et al. (1966). The presumed tsp of the hdc operon was confirmed using primer extension and histidine-induced transcription of the two genes has been quantitated for both the wt and mutant-3 strains. The HDCB protein has been produced in E. co&, purified and used to generate ~lyclon~ antiserum which, along with antibodies to HDC, has been used to measure the change in concentrations of the two proteins in L30a during histidine induction.

MATERIALS AND METHODS

(a) Standard DNA cloning techniques Plasmid isolation, oligo preparation, genomic library construction, colony screening, DNA sequencing and DNA hyb~d~ation were performed as previously described (Vanderslice et al., 1986; 1988; Copeland et al., 1987), unless otherwise noted. (b) Primer extension Primer extension on total RNA was performed using a slight mo~cation of the procedure of Jones et al. (1985). Six pmol of 32P-labeled oligo complementary to the message strand of hdcA gene were added to 25 pg of total RNA isolated from wt L30a in 10 ~1 of 7 mM Tris 9HCl pH 8.3/7 mM MgCl,/SO mM NaCl. This mixture was heated for 5 mm at 65 ‘C in a water bath and then allowed to anneal at room temperat~e for 25 min. After annealing, 2 ~1 of 0.5 M Tris * HCl pH 8.3/50 mM MgCl,/SO mM DTT/O.S M KCl, 2 ~1 of a mixture of the four dNTPs (20 mM each), and 400 units of Moloney murine leukemia virus reverse transcriptase (Bethesda Research Laboratories) were added and the reaction volume was adjusted to 20 ~1with sterile water. Following incubation for 1 h at 42”C, 10 fig of RNase A (Sigma Chemical Co.) was added and incubation continued for 10 mm at room temperature. The DNA was ethanol-precipitated, rinsed with 70% ethanol, dried and resuspended in 10 ~1 50% formamide, 0.1% bromophenol blue in 1 x TBE buffer (Maniatis et al., 1982). Two ~1 were loaded onto a standard 7 M ureaJ8% ~lyac~l~de sequencing gel. The same oligo primers used above

261

were annealed to the h&4 gene in an M 13 vector and used to generate a sequence ladder which was loaded in adjacent lanes of the same gel. Electrophoresis and autoradiography were performed as previously described (Vanderslice et al., 1986). (c) Expression of the RdcB gene product and protein purification To purify the hdcB gene product, 2 liters of E. coli JM105 cells (Yanisch-Perron et al., 1985) carrying pBC15 (Fig. 1) were grown for 12-14 h at 37°C in a rich medium made up of 16 g tryptone, 10 g yeast extract and 10 g NaCl per liter. Cell pellets (12 g), obtained by centrifugation (15 min, 5000 x g) were frozen in liquid nitrogen and stored at -80°C. Frozen cells were thawed and resuspended in 12 ml of buffer A then disrupted by grinding for 10 min in a pre-cooled mortar with 12 g of alumina (Sigma). The mixture was centrifuged for 10 min at 5000 x g and the supematant set aside. The pellet was resuspended in 12 ml of buffer A and ground for another 5 min followed by centrifugation for 10 min at 5000 x g. The two supematants were combined to yield the crude cellular extract (Fraction I). An equal volume of 2 y0 (w/v) streptomycin sulfate in buffer A was slowly added to Fraction I over a l-h period to precipitate nucleic acids which were then removed by centrifugation for 30 min at 19000 x g. The resulting supematant was brought to 25% saturation with solid ammonium sulfate and the precipitate removed by centrifugation for 15 min at 19000 x g. The supematant was then brought to 70 % saturation and the precipitate collected by centrifugation as above. The 70% precipitate was resuspended in buffer A and dialysed against three l-liter changes of buffer A (Fraction II). Fraction II was loaded onto a 2.5 x 30 cm column of DE 52 (Whatman) which had been equilibrated in buffer A. The column was washed with 300 ml of buffer A at 25 ml/h. A 600~ml linear gradient of O-300 mM NaCl in buffer A was applied to the column and 4-n-n fractions were collected. Samples (10 ~1) of every other fraction were subjected to denaturing gel electrophoresis as described in the legend to Fig. 4. Fractions containing a 20-kDa protein were pooled (Fraction III). Fraction III was concentrated to approx. 4 mg/ml in a stirred cell concentrator with an Amicon YM 10

1

B X

1 Gel purify 1

1 B

1 1

HC Gel purify

/ I

Fig. 1. Construction ofthe pBCl5 expression plasmid. (Top line) L30a DNA around the hdc operon. Hatched boxes, coding repions, with the N-terminal regions to the left, hdcA is on the left, hdcB is on the right. A 2-kb BamHI restriction fragment of L30a DNA isolated from a L30a genomic library using an oligo probe complementary to the hdcA/hdcB intergenic region was inserted into the BamHI site of pUC8 to yield pHDCA. Plasmid pPV7 (Copeland et al., 1987) expresses hdcA from the ZacZp promoter present on the pUC13 parent plasmid. The hdcB gene was excised from pHDCA and inserted into its native position behind the hdcA gene in pPV7 yielding pBC15. Abbreviations: B, BamHI; D, DruI; E, EcoRI; Hc, HincII; X, XbaI.

membrane. Ten mg of this material were loaded onto a 2.5 x 120 cm Sephadex G-75 column which had been equilibrated with buffer A. The column was washed with buffer A at 20 ml/h while 3-ml fractions

262

were collected. Fractions containing a 20-kDa protein were pooled and concentrated to approx. 1 mg/ml (Fraction IV). (d) Amino acid sequence determination N-terminal aa sequencing was performed on an Applied Biosystems 477A instrument. The sample consisted of 40 pg of Fraction IV which had been dialysed into water and concentrated to 50 ~1 by evaporation. (e) Growth and extraction of L3Oa for immunoquantitation of HDC and HDCB Cultures (45 ml) of wt and mutant-3 L30a (Recsei and Snell, 1973) were grown in the crude medium of Chang and Snell(1968) with or without 13 mM histidine. Growth was initiated by inoculating the 45-ml cultures with 450 ~1 of a dense overnight culture which had been grown in crude medium without histidine. The cultures were maintained at 37 ‘C for 14 h without aeration. Cells were harvested by centrifugation for 15 min at 5000 x g. Pellets were resuspended in 45 ml of H,O, then re-pelleted and resuspended in 2 ml of H,O. One ml of this material was placed in two 2-ml tubes, pelleted and frozen in liquid nitrogen. Tubes containing L30a pellets were thawed and 1 ml of 0.1-M ammonium acetate pH 4.6 was added. Cells were disrupted by shaking with 0.1~mm glass beads for three 1-min pulses in a mini bead-beater (Bio-Spec Products, Bartlesville, OK). Following centrifugation for 10 min at 3500 x g, the supematant was collected and protein measured using the Bio-Rad protein assay reagent. Electrophoresis, blotting and immunodetection were carried out as described in the legend to Fig. 4.

RESULTS AND DISCUSSION

(a) Nucleotide sequence of the hdcB gene Fig. 2 shows the nt sequence of the hdcB gene and its deduced aa sequence. Both strands were sequenced over the entire length of the gene. The ORF encodes a 174-aa protein with a calculated M, of 19883. Following the TAA stop codon is an

ICC GTT 1,,

Act

ccc cc, t,c

Ghc

Fig. 2. The nt and deduced aa sequence of HDCB. The sequence of the mRNA-like strand is shown. Within the protein-coding region, the aa are written below the nt sequence. Palindromic sequences capable of producing stem and loop structures are indicated by arrows. Sequencing was performed as described previously (Vanderslice et al., 1986).

oligo(dA) stretch and an ‘inverted repeat cbrresponding to a potential 16-bp stem and a 30-nt loop structure. Ten nt after this structure is an ATG codon followed by a short ORF extending to the end of the present sequence data. (b) Transcriptional regulation of the hdcA and hdcB genes To identify the 5’ end of the hdcA mRNA, two [32P]oligos complementary to different regions of

263

the message strand of these genes were separately annealed to total RNA isolated from L30a. These oligos were extended with reverse transcriptase and electrophoresed beside the products of sequencing reactions utilizing the same oligo primers with the h&4 gene and upstream region as a template. Autoradiography of the gels revealed a band containing 90 % of the incorporated label, the position of which corresponded to a guanine located 73 nt 5’ to the start codon for the MO4 gene (not shown). We previously proposed this guanine as the tsp based on sequence similarities between this gene and other Gram+ genes (Vanderslice et al., 1986). To determine whether the increase in HDC activity observed during histidine induction (Recsei and Snell, 1982) was due to increased transcription, total RNA was isolated from wt L30a grown in the presence or absence of 13 mM free histidine and mutant-3 L30a grown in the absence of 13 mM histidine. Northern-blot analysis of this RNA (Fig. 3) showed that histidine induction increased transcription of the hdc operon threefold in wt L30a. The amount of h&A message in uninduced mutant-3 was 15 times greater than that in uninduced wt; subsequent Northern blots showed that 13 mM histidine had no effect on the level of hdc transcripts in the mutant-3 strain (not shown). These data agree well with the results of Recsei and Snell (1982), who found that mutant-3 produced 4.5 times as much HDC protein as wt when both strains were induced with histidine. We previously noted a potential stem and loop between the h&l and h&B genes and speculated that this structure might serve as a terminator for RNA polymerase (Vanderslice et al., 1986). The two different bands seen on the Northern blots provided evidence to support this proposal (Fig. 3). The h&Bspecific probe should hybridize only to a transcript encoding both genes, while the h&A probe should bind to these same full-length transcripts as well as any shorter transcripts produced by termination in the intergenic region. As shown in Fig. 3, the h&A probe bound to two RNA bands, one of 1000 nt and one of 1800 nt, while the h&B probe identified only the 1%kb band. Quantitation of the two bands indicated that both the wt and mutant-3 strains, regardless of histidine induction, produced l/6 as many long transcripts as short transcripts. Although we have not determined the precise 3’ end of the shorter

B

A kb

I23

I23

Fig. 3. Autoradiogram of Northern blots containing total RNA isolated from histidine induced and uninduced cultures of wt and mutant-3 L30a. Lanes 1A and 1B contain total RNA isolated from histidine-induced cultures of wt L30a. Lanes 2A and 28 contain total RNA from cultures of uninduced wt, while lanes 3A and 3B contain total RNA from utiduced cultures of mutant-3 L30a. Blot A was hybridized with an oligo specific for the N terminus of the hdcA gene, blot B was hybridized with an oligo specific for the C terminus of the hdcB gene. The positions of RNA markers, in kb, are shown on the left. This autoradiogram is overexposed to show the lighter bands from the wt samples; densitometry of mutant-3 RNA bands was performed on unsaturated autoradiograms. L30a strains were grown without aeration in 250 ml ofcrude medium (Chang and Snell, 1968), plus or minus 13 mM histidine. Total RNA was isolated from these cultures (Wek and Hatfield, 1986). 2Opg of total RNA was denatured in 50% formamide/2.2 M formaldehyde at 65°C for 3 min and electrophoresed through a 2.5% formaldehyde/l % agarose gel (Foumey et al., 1988). RNA was transferred to nitrocellulose (Thomas, 1980) for 36 h, then incubated 2 h at 80°C under vacuum. Prehybridixation and hybridization with [32P]oligo probes were done under Southern-blot conditions (Maniatis et al., 1982); autoradiograph bands were quantified by scanning densitometry.

transcript, a 1-kb transcript is sufficient to encode the HDC protein. The predicted size of a transcript including both HDC and HDCB is 1.8 kb. The involvement of the putative terminator structure and/or other factors in transcriptional termination in the h&4/h&B intergenic region remains to be investigated. (c) Purification of the hdcB gene product

A purification scheme described in MATERIALS AND METHODS,

SeCtiOIl C Was

USed t0

ptllify

a

264

20-kDa protein found in extracts of E. coli cells carrying the pBCl.5 plasmid. This band was not seen in extracts of E. coli cells bearing pUC13, the parent plasmid of pBC15 (not shown). Fraction IV consisted primarily of this 20-kDa protein (Fig. 4A). The identity of this protein as HDCB was confinned by N-terminal aa sequencing. The aa sequence thus determined matches that deduced for aa 2-16 of HDCB. Rabbit polyclonal antiserum containing anti-HDCB activity was obtained and verified by Western blotting (Fig. 4B). The function of this protein is not known.

kDa

(d) Measurement of ISDC and HDCB protein levels in L3Oa

Equal amounts of protein (100 pg) from wt and mutant-3 L30a extracts were subjected to SDS-polya~ryl~de-gel el~trophoresis and then electrophoretically transferred to nitrocellulose membranes. Immunostaining and densitometry of these Western blots were used to measure the approximate relative levels of HDC and HDCB in extracts of wt and mutant-3 L30a grown with and without histidine. The amount of HDC protein measured in histidine-induced wt cells was five times that in uninduced cells, while both induced and uninduced mutant-3 cells contained tenfold more than uninduced wt. These figures agree reasonably well with the transcript data which indicated three- and 15-fold increases, respectively. The amount of HDCB protein measured in induced wt cells was 1.5 times higher than in cells grown without histidine. Mutant-3, with or without induction, produced about 2.5 times as much HDCB as the uninduced wt. The discrepancy between this figure and transcript data (2.5 vs. 10) was most likely due to errors in protein quantitation arising from the small amount of HDCB protein in uninduced wt L30a, which was near the lower limit of detection of the Junostaining technique. (e) Conclusions We have investigated the mechanism of induction of HDC activity in L30a and have found that a second gene, hdcB, encoding a 20-kDa protein of unknown function, begins 59 nt after the final codon of the McA gene. The entire h&B gene sequence is

Fig. 4. SDS-poiyac~I~de-gel electrophoresis and Western blotting of HDCB. (Panel A) 0.1% SDS-12.5% ~Iyac~l~idegel electrophoresis (Coomassie brilliant blue-stained) of HDCB samples. Lanes: 1, pre-stained h& markers (Bio-Rad); 2, Fraction II; 3, Fraction III; 4, Fraction IV. The molecular sixes of the markers are indicated on the left margin (in kDa). The anomalous migration of the HDCB band in lane 4 is due to a slight ‘smile’ in this gel and was not observed on other gels. (Panel B) A Western blot from an SDS-polyacrylamide gel showing the specificity of the anti-BDCB antiserum. Lanes: 1, pre-stained M, markers; 2, Fraction IV HDCB (1 pg); the ~munoreactive band above HDCB was due to an antigenic impurity in the HDCB preparations used for immunization; 3, crude extract from histidine-induced L30a wt (5Ong). SDS-polyacrylamide slab gels (Laemmli and Favre, 1973) were run at 200 V. After electrophoresis, proteins were transferred to nitrocelhtlose membranes (PH75, Schleicher & Schueli) in a Hoefer Scientific Instruments TE 42 Transfor unit with 25 mM Tris base/192 mM glycine/20% (v/v) methanol. Transfer voltage: O-4 min, 20 V; 4-8 min, 40 V; 8-12 min, 60 V; 12-16 min, 80 V; 16-24min, 1OOV. Membranes were agitated in the following solutions at room temperature: TN but&r plus 0.5 y0 (w/v) nonfat powdered milk (30 mm); anti-HDCB serum or purified antiHDC antibodies (Recsei and Snell, 1982) diluted l/1000 in TN plus 0.5 % milk (2 h); TN buffer (5 min); goat anti-rabbit horseradish peroxidase conjugate (Bio-Rad) diluted l/3000 in TN plus 0.5% milk (1 h); TN bull’er (10 mm); 0.5 mg/ml Bio-Rad HRP color development solution, 0.01% hydrogen peroxide in TN buffer (10 min). Intensity of stained bands was quantitated by densitometry. Rabbits were injected subcutaneously with 1 mg of Fraction IV HDCB in Freund’s complete adjuvant (Difco). Subsequent injections (0.75 mg) of HDCB in incomplete adjuvant were performed at two-week intervals. Alter live weeks, blood samples were collected every other week. Preimmune and immune sera were tested for anti-HDCB activity using standard 96-well plate ELISA protocols.

265

reported here along with the puritication of the HDCB protein. The h&4 and hdcL3genes are components of an operon transcribed from a single promoter. Two transcripts of 1 kb and 1.8 kb are produced from this promoter in vivo. The shorter transcript, which accounts for 85% of the total message transcribed, encodes only HDC while the longer one encodes both HDC and HDCB. Transcription of the hdc operon in wt L30a was increased threefold by histidine in the growth medium, while transcription in the mutant-3 strain was constitutively 15-fold greater than uninduced wt. Immunoquantitation of HDC and HDCB proteins in cell extracts yielded similar results. The mechanisms by which the hdc operon is regulated in response to histidine and the possible involvement of HDCB in histidine/histamine metabolism are the subjects of continuing investigation in this laboratory.

ACKNOWLEDGEMENTS

The nt sequence for hdcBhas been deposited in the EMBL GenBank (accession No. X13099). The wt (American Type Culture Collecton Strain No. 33222) and mutant-3 strains of L30a and anti-HDC antibodies (Recsei and Snell, 1982) were kindly provided by Dr. E. Snell of the University of Texas at Austin. We are grateful to Michael Ready for helpful discussions and to Raquelle Keegan for her help in preparing the figures. This work was supported by grants GM35989 and GM30048 from the National Institutes of Health.

REFERENCES Chang, G.W. and Snell, E.E.: Hi&line decarboxylase of La&obacillus 30a II. Purification, substrate specificity and stereospecificity. Biochemistry 7 (1968) 2005-2012.

Copeland, W.C., Vanderslice, P. and Robertus, J.D.: Expression and characterization of Lactobacillur 30a histidine decarboxylase in Escherichia coli. Protein Eng. 1 (1987) 419-423. Demerec, M., Adelberg, E.A., Clark, A.J. and Hartman, P.E.: A proposal for a uniform nomenclature in bacterial genetics. Genetics 54 (1966) 61-76. Fourney, R.M., Miyakoshi, J., Day, R.S. and Paterson, M.C.: Northern blotting: efficient staining and transfer. Focus 10 (1988) 5-7. Graves, M.C. and Rabinowitz, J.C.: In vivo and in vitro transcription of the Clostridium pasieurianum ferredoxin gene. J. Biol. Chem. 261 (1986) 11409-11415. Jones, K.A., Yamamoto, K.R. and Tjian, R.: Two distinct transcription factors bind to the HSV thymidine kinase promoter in vitro. Cell 42 (1985) 559-572. Laemmli, U.K. and Favre, M.: Maturation of the head of bacteriophage T4. J. Mol. Biol. 80 (1973) 575-599. Maniatis, T., Fritsch, E.F. and Sambrook, J.: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982. Recsei, P.A. and Snell, E.E.: Prohistidine decarboxylase from Lactobacillw 30a. A new type of zymogen. Biochemistry 12 (1973) 365-371. Recsei, P.A. and Snell, E.E.: Histidine decarboxylase of Lacrobacillus 30a. Comparative properties of wild type and mutant proenzymes and their derived enzymes. J. Biol. Chem. 257 (1982) 7196-7202. Recsei, P.A. and Snell, E.E.: Pyruvoyl enzymes. Amm. Rev. Biochem. 53 (1984) 357-387. Rodwell, A.W.: The histidme decarboxylase of a species of Lactobacillur; apparent dispensability of pyridoxal phosphate as co-enzyme. J. Gen. Microbial. 8 (1953) 233-237. Thomas, P.S.: Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA 77 (1980) 5201-5205. Vanderslice, P., Copeland, W.C. and Robertus, J.D.: Cloning and nucleotide sequence of wild type and a mutant histidine decarboxylase from Lactobacillus 30a. J. Biol. Chem. 261 (1986) 15186-15191. Vanderslice, P., Copeland, W.C. and Robertus, J.D.: Sitedirected alteration of serine-82 causes nonproductive chain cleavage in prohistidine decarboxylase. J. Biol. Chem. 263 (1988) 10583-10586. Wek, R.C. and Hatfield, G.W.: Nucleotide sequence and in vivo expression of the ilvY and iZvCgenes in Escherichiu coli K-12. J. Biol. Chem. 261 (1986) 2441-2450. Yanisch-Perron, C., Vieira, J. and Messing, J.: Improved Ml3 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33 (1985) 103-l 19.