Cloning and expression of human deoxyhypusine synthase cDNA. Structure-function studies with the recombinant enzyme and mutant proteins.

Deoxyhypusine synthase catalyzes the first step in the post-translational formation of hypusine (N-(4-amino-2-hydroxybutyl)lysine). cDNA clones encoding deoxyhypusine synthase were isolated from a human HeLa cell library. Full-length cDNA clones encoding a 369-amino acid protein (calculated molecular mass of 40,970 Da) and a shorter cDNA clone that would potentially encode a protein with an internal deletion of 56 amino acids (Asp-Ser) were isolated. The deduced amino acid sequence of the human enzyme shows a high degree of identity to that of yeast deoxyhypusine synthase and to the known sequences of tryptic peptides from the rat and Neurospora crassa enzymes. The recombinant enzyme formed upon expression in Escherichia coli effectively catalyzed deoxyhypusine synthesis. Variant human recombinant proteins with (i) a truncation of 48 or 97 NH-terminal amino acids, (ii) a truncation of 39 COOH-terminal amino acids, or (iii) an internal deletion (Asp-Ser) were inactive. A chimeric protein consisting of the complete human sequence and 16 amino acids of the yeast sequence (Gln-Asn, not present in the human enzyme) inserted between Glu and Gln exhibited moderate activity.

Deoxyhypusine synthase catalyzes the first step in the post-translational formation of hypusine (N ⑀ -(4-amino-2hydroxybutyl)lysine). cDNA clones encoding deoxyhypusine synthase were isolated from a human HeLa cell library. Full-length cDNA clones encoding a 369-amino acid protein (calculated molecular mass of 40,970 Da) and a shorter cDNA clone that would potentially encode a protein with an internal deletion of 56 amino acids (Asp 262 -Ser 317 ) were isolated. The deduced amino acid sequence of the human enzyme shows a high degree of identity to that of yeast deoxyhypusine synthase and to the known sequences of tryptic peptides from the rat and Neurospora crassa enzymes. The recombinant enzyme formed upon expression in Escherichia coli effectively catalyzed deoxyhypusine synthesis. Variant human recombinant proteins with (i) a truncation of 48 or 97 NH 2

-terminal amino acids, (ii) a truncation of 39 COOH-terminal amino acids, or (iii) an internal deletion (Asp 262 -Ser 317 ) were inactive. A chimeric protein consisting of the complete human sequence and 16 amino acids of the yeast sequence (Gln 197 -Asn 212 , not present in the human enzyme) inserted between Glu 193 and Gln 194 exhibited moderate activity.
The unusual amino acid hypusine (N ⑀ -(4-amino-2-hydroxybutyl)lysine) is formed post-translationally in a single cellular protein, eukaryotic translation initiation factor 5A (eIF-5A) 1 (1)(2)(3). Hypusine biosynthesis involves two enzymatic steps. In the first step, deoxyhypusine synthase catalyzes the NAD-dependent transfer of the butylamine moiety of spermidine to the ⑀-amino group of a specific lysine residue of the eIF-5A precursor protein to form the intermediate deoxyhypusine (N ⑀ -(4aminobutyl)lysine) residue (1, 4 -7). In the second step, mediated by deoxyhypusine hydroxylase, the conversion of the deoxyhypusine residue to the hypusine residue (8) completes eIF-5A maturation.
Hypusine is found in all eukaryotes and in some archaebacteria, but not in eubacteria (1,9). The amino acid sequence of the hypusine-containing protein, eIF-5A, is highly conserved, especially in the region surrounding the hypusine residue, sug-gesting an important fundamental function for this protein throughout eukaryotic evolution. In spite of its in vitro activity in the stimulation of methionyl puromycin synthesis and tentative identification as an initiation factor in eukaryotic protein synthesis (10,11), its true cellular function is unknown (12,13). However, hypusine and eIF-5A appear to be vital for cell proliferation in eukaryotes (12). The correlation between a reduction in hypusine formation and growth arrest in spermidine-depleted mammalian cells (14) and the requirements for expression of eIF-5A precursor protein and its hypusine modification for yeast viability (15,16) strongly support this notion. Furthermore, inhibitors of the hypusine biosynthetic enzymes exert antiproliferative effects (17)(18)(19). In this regard, deoxyhypusine synthase presents a potential target for intervention in cell proliferation. A detailed knowledge of the amino acid sequence and properties of this enzyme should contribute to attaining more specific inhibitors of this enzyme and perhaps to better control of hyperproliferative diseases.
We have recently purified deoxyhypusine synthase from rat testis, determined the amino acid sequences of several tryptic peptides, and raised polyclonal antibodies to this enzyme (20). Cross-reactivity of anti-rat deoxyhypusine synthase with the human enzyme (from HeLa cells) was utilized to facilitate the cloning of cDNAs encoding human deoxyhypusine synthase.
In this study, we report the molecular cloning, expression, purification, and characterization of human recombinant deoxyhypusine synthase. Studies of deletion and insertion mutations, designed on the basis of a comparison of the amino acid sequences of the human and yeast (21) enzymes, provide insight into the structure-function relationship of the enzyme. [1, H]Spermidine HCl (15 Ci/mmol) was purchased from DuPont NEN. Oligonucleotide primers were synthesized by the Midland Certified Reagent Co. The pET-11a expression vector and the host Escherichia coli B strain BL21(DE3) competent cells were from Novagen; T4 DNA ligase and the Superscript preamplification system for singlestrand cDNA synthesis were from Life Technologies, Inc.; AmpliTaq DNA polymerase was from Perkin-Elmer; restriction enzymes were from Boehringer Mannheim; and precast polyacrylamide gels and widerange protein standards (Mark 12) were from Novex. ec-eIF-5A was purified from E. coli lysates after overexpression of the human eIF-5A cDNA as described previously (22). Total RNA from 144 JAR, a human trophoblast cell line (American Type Culture Collection), was kindly provided by Dr. Janet Kerr (NIDR) and total RNA from human epithelial cell lines by Dr. W. Andrew Yeudall (Laboratory of Cellular Development and Oncology, NIDR).

Methods
Deoxyhypusine Synthase Assay-The enzyme activity was measured as described previously (7,20). A typical reaction mixture contained, in a total volume of 20 l, 0.2 M glycine NaOH buffer, pH 9.5, 1 mM dithiothreitol, 25 g of bovine serum albumin, 1 mM NAD, 7 M (2 Ci) [1, H]spermidine, 10 M ec-eIF-5A, and enzyme. Incubations were at 37°C for 60 min. The [ 3 H]deoxyhypusine formed was measured after its * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EMBL Data Bank with accession number(s) L39068.
Immunopurification of Rabbit Polyclonal Antibodies-Rabbit antisera against purified rat testis deoxyhypusine synthase were prepared as described previously (20). The polyclonal antibody was purified by affinity adsorption to the immobilized antigen according to a published procedure (24). Briefly, rat testis enzyme (ϳ100 g) was separated by SDS-PAGE and then electrophoretically transferred to an Immobilon (polyvinylidene difluoride) membrane (Millipore Co., Bedford, MA). The 41-kDa enzyme band was excised, and the strip was incubated with antisera. After washing the strip with 0.1 M potassium phosphate buffer, pH 4.5, 1.0 M NaCl, the specifically bound antibodies were eluted in a small volume of 0.1 M glycine HCl buffer, pH 3.0, and immediately neutralized with 1.0 N NaOH.
Cloning and Sequencing of Human cDNAs Encoding Deoxyhypusine Synthase-1 ϫ 10 6 plaques from a HeLa cell cDNA library in the Uni-Zap XR vector (Stratagene) were initially screened using the purified rabbit polyclonal antibody against rat deoxyhypusine synthase. Three immunopositive clones were detected. Nucleotide sequence analysis of their cDNA inserts was carried out by the dideoxy chain termination method (Sequenase Version 2, U. S. Biochemical Corp.) and the cycle sequencing method (SequiTherm, Epicentre Technologies Corp.). One clone among these, designated hDS-1a, was identified as a tentative deoxyhypusine synthase clone based on the close similarity of its deduced amino acid sequence to those of tryptic peptides of the rat enzyme. Because hDS-1a was not a full-length clone, 1 ϫ 10 6 plaques from the same library were rescreened using radiolabeled hDS-1a insert DNA as a hybridization probe. This probe (683 base pairs) was obtained by PCR using two primers, Pr5-2 and Pr3-3 (see Tables I and  II), and hDS-1a cDNA as template. Twenty-five positive clones, including the three independent clones hDS-10, hDS-13, and hDS-20, each encoding full-length human deoxyhypusine synthase, were isolated.
Construction of a Human Deoxyhypusine Synthase Expression Plasmid-Full-length recombinant enzyme was produced by PCR amplification of the complete coding sequence and its expression in E. coli. On the basis of the nucleotide sequence of hDS-10, PCR primers (Pr5-1 and Pr3-1; see Tables I and II) representing the 5Ј-and 3Ј-end regions of the open reading frame were synthesized with a built-in NdeI site (5Ј-end primer) and a BamHI site (3Ј-end primer) (underlined in Table I) to facilitate cloning into the expression vector pET-11a. PCR was performed in a thermal cycler with a Perkin-Elmer PCR kit using hDS-10 plasmid DNA as template. The conditions for PCR were as follows: 94°C for 5 min, denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and an extension reaction at 74°C for 2 min for 35 cycles and a final extension reaction at 74°C for 6 min. The PCR product was digested with NdeI and BamHI, ligated to linearized pET-11a, and used to transform E. coli BL21(DE3) cells. The selected transformants were grown in LB medium supplemented with 50 g/ml ampicillin. When the cell density reached an optical density of 0.6 at 600 nm, isopropyl-1thio-␤-D-galactopyranoside was added to a final concentration of 1 mM, and the cells were harvested by centrifugation 4 h later.
Construction of Deletion or Insertion Mutant Subclones of Human Deoxyhypusine Synthase cDNA and Their Expression in E. coli-Dele-tion and insertion mutants of human deoxyhypusine synthase cDNA were generated by a method similar to that used for the full-length enzyme. The nucleotide sequences of all the primers are listed in Table  I; primer pairs and templates used to construct each subclone are indicated in Table II. The mutant subclones with a truncated 5Ј-terminus, hDS m4 and hDS m5, were generated using 5Ј-end primers that hybridized to internal sites (the coding sequences for amino acids 49 -56 and 98 -105, respectively) and Pr3-1, and the subclone with a truncated 3Ј-terminus, hDS m6, was generated using Pr5-1 and Pr3-2, as indicated in Table II. A mutant cDNA with an internal deletion identical to that in hDS-1a was constructed by a two-step PCR procedure. In the first step, hDS m1 and hDS m2 cDNAs were generated in two separate reactions (Table II). The PCR products of the expected size were isolated from an agarose gel, and the mixture of the two amplified fragments was used as template in the second PCR step to construct hDS m3. The mutant subclone hDS m9, with the insertion of a sequence corresponding to Gln 197 -Asn 212 of yeast deoxyhypusine synthase, was derived by a similar approach. In the first step, hDS m7 and hDS m8 cDNAs were formed in two separate PCRs using primers containing partial nucleotide sequences of the yeast cDNA. In the second step, hDS m9 cDNA was amplified using a mixture of hDS m7 and hDS m8 cDNAs as template. After cloning PCR products into pET-11a, the expression of altered recombinant proteins was carried out as described above.
Purification of Human Recombinant Deoxyhypusine Synthase-Isopropyl-1-thio-␤-D-galactopyranoside-induced E. coli cells (20 g) were resuspended in 120 ml of buffer A (50 mM Tris acetate, pH 6.8, 0.1 mM EDTA, 1 mM dithiothreitol, and 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride) and sonicated for a 2-min interval at a setting of 70 watts, and the supernatant was collected after centrifugation for 30 min at 15,000 ϫ g. The supernatant was applied to a Fast Flow Q-Sepharose column (5 ϫ 6 cm; Pharmacia Biotech Inc.) previously equilibrated with buffer A; the column was washed with 100 ml of buffer A; and the proteins were eluted with a stepwise gradient (100 ml each) of 0.1, 0.2, 0.3, 0.4, and 0.5 M KCl in buffer A. The fractions containing enzyme activity, which was eluted between 0.3 and 0.4 M KCl, were pooled and concentrated by precipitation with ammonium sulfate (40 -80% saturation). After dialysis against buffer A, the enzyme fraction was applied to a Mono Q HR 10/10 column equilibrated with buffer A. The adsorbed enzyme was eluted with a 120-ml linear gradient of 0 -0.5 M KCl in buffer A. PAGE analysis of the purified recombinant enzyme showed a single band in the presence of SDS.

Isolation and Characterization of Human cDNAs Encoding
Deoxyhypusine Synthase-In the initial screening of the HeLa cell cDNA library with rabbit polyclonal antibodies against rat deoxyhypusine synthase, one positive clone, hDS-1a, was identified (Fig. 1). The deduced amino acid sequence of hDS-1a showed regions of identity to four tryptic peptides derived from the rat enzyme (Fig. 2, underlined sequences), but no sequence matching a fifth peptide (NGADYAVYINTAQE) was present. Of 25 positive clones isolated upon rescreening the library using a PCR-generated fragment of the hDS-1a insert as a   II Construction of full-length and altered deoxyhypusine synthase cDNAs by PCR and the activities of the recombinant proteins Each subclone was constructed from PCR amplification products using the template cDNAs and the sets of primers listed; the expression of recombinant proteins and determination of their activities were as described under "Experimental Procedures." In the case of 5Ј-terminal deletions, methionine was artificially introduced as the NH 2 -terminal amino acid (hDS m4-and hDS m5-encoded proteins). In the schematic representation, human coding sequences are black boxes, the deleted regions are white boxes, and the inserted yeast deoxyhypusine coding sequences are depicted by hatched boxes. The numbers indicate the amino acid residue number in the human coding sequence. A schematic diagram of the location of the primers in relation to the coding sequence of hDS-13 ( Fig. 1)  hybridization probe, three independent clones (hDS-10, hDS-13, and hDS-20) were found to contain complete and identical open reading frames predicted to encode a protein of 369 amino acid residues. The nucleotide sequence and the deduced amino acid sequence of one such representative clone, hDS-13, are shown in Fig. 1. Upon comparison of the sequence of hDS-13 with that of hDS-1a, it is evident that, in hDS-1a, there is an internal deletion of 168 base pairs corresponding to a deletion of 56 amino acids (Asp 262 -Ser 317 ) near the COOH terminus, in addition to a truncation of the 5Ј-terminal region (166 base pairs, 43 NH 2 -terminal amino acids). Except for the deletions underlined in Fig. 1, the sequences of the two clones are identical. To see if any other clone might have undergone the same internal deletion, PCR was performed using other positive clones as templates and the two primers, Pr5-5 and Pr3-3 (Table I), that flank the region of interest. The PCR results, taken together with analysis of the nucleotide sequences of this region in other clones, indicate that this deletion is unique for clone hDS-1a.
A comparison of the primary structure of human deoxyhypusine synthase with those of other species is shown in Fig. 2. The deduced amino acid sequence of the human enzyme is 58% identical and 73% similar to that of the yeast enzyme (21). Three regions of high sequence identity exist, encompassing residues 98 -188, 201-248, and 277-330 of the human enzyme. In addition, the amino acid sequences determined for several tryptic peptides from the rat testis enzyme (20) or the Neurospora crassa enzyme (25) also are highly similar to the corresponding regions of the human and yeast enzymes. It is interesting to note that the human enzyme is shorter than the yeast enzyme. No counterpart sequence for 16 amino acids (Gln 197 -Asn 212 ) of the yeast enzyme is present in the human enzyme.
Northern Blot Analysis-As shown in Fig. 3, one major transcript of ϳ1.4 kilobases was detected in RNA isolated from HeLa cells, human trophoblast cells (Fig. 3A), normal human epidermal keratinocytes (NHEK), or cell lines derived from head and neck tumors and dysplasias (Fig. 3B). Although deoxyhypusine synthase mRNA was detected in all the cell lines examined, its level appeared to be elevated in HeLa cells and in some of the tumor cell lines.
Purification of Human Recombinant Deoxyhypusine Synthase-Human deoxyhypusine synthase was expressed in E. coli BL21(DE3) cells in a time-dependent manner for up to 4 h after isopropyl-1-thio-␤-D-galactopyranoside addition (data not shown). The enzyme activity was purified by a simple four-step protocol, summarized in Table III. After chromatography on Mono Q HR 10/10 (Fig. 4A), one major protein band, at ϳ41 kDa, was detected upon SDS-PAGE of the peak fractions (Fig.  4B). The specific activity of human recombinant deoxyhypusine synthase (0.9 ϫ 10 6 units/mg of protein) was similar to that of the yeast recombinant enzyme (21) and of the enzyme purified from rat testis (20).
Enzymatic Properties-The catalysis of deoxyhypusine formation in the eIF-5A precursor, from spermidine in the presence of NAD, by human recombinant protein defines it as deoxyhypusine synthase. Like the rat (7,20) and yeast (21) enzymes, the human recombinant enzyme also catalyzes the cleavage of spermidine to diaminopropane and ⌬ 1 -pyrroline in the absence of the eIF-5A precursor (data not shown). The strict requirement of the human enzyme for NAD is also char- acteristic of the rat and yeast enzymes. As with these enzymes, other nucleotides, including NADH, NADPH, FAD, and FMN, did not substitute for NAD with human deoxyhypusine synthase. Interestingly, the K m values for NAD (Table IV) are strikingly different for human, rat, and yeast enzymes, varying as much as 150-fold, whereas the K m values for spermidine and ec-eIF-5A are not very different (Table IV). The K m value for NAD for the human recombinant enzyme was ϳ4.8 M, 8-fold lower than that for the rat enzyme and 150-fold lower than that for the yeast recombinant deoxyhypusine synthase.
Deletion and Insertion Mutations-Since the initial clone, hDS-1a, had an internal deletion corresponding to the loss of 56 amino acids (amino acids 262-317), we designed a mutant clone, hDS m3, with this deletion but with intact 5Ј-and 3Јterminal coding sequences (Table II) in order to assess the importance of this region. The lysate of E. coli cells overexpressing protein from this clone was found to display no enzyme activity (Table II). Lysates of E. coli cells overexpressing recombinant proteins with truncation of either 48 or 97 amino acids from the NH 2 terminus or of 39 amino acids from the COOH terminus also exhibited no activity (Table II). In an effort to determine the basis of the drastic difference in the K m value for NAD for the human and yeast enzymes, we generated a mutant subclone, hDS m9, with the partial nucleotide sequence corresponding to Gln 197 -Asn 212 of the yeast deoxyhypusine synthase inserted into the human cDNA (Table II). After expression of this clone, the resultant chimeric protein, isolated by ion-exchange chromatography on Mono Q HR 10/10 (data not shown), displayed a specific activity (ϳ6 ϫ 10 4 units/mg of protein) significantly lower than that of the wildtype enzyme. Interestingly, the K m value determined for NAD (ϳ7 M) for this aberrant enzyme was similar to the value for the wild-type human enzyme, rather than that for the yeast enzyme.

DISCUSSION
Hypusine is ubiquitous in all eukaryotic species, and its biosynthetic pathway appears to be highly conserved. This study shows that the human enzyme shares similar properties with deoxyhypusine synthases from diverse species, e.g. rat (7, FIG. 2. Comparison of amino acid sequences of deoxyhypusine synthases from different species. The amino acid sequences deduced from human and yeast cDNAs are compared with the partial amino acid sequences determined for tryptic peptides from the rat testis enzyme (20) (underlined) and the Neurospora enzyme (25) (italic). To allow for maximal alignment of the human and yeast sequences, gaps have been introduced (. . . . ). The amino acid residue numbers for the human and yeast sequences are indicated on the right. The vertical lines between residues denote identity, while the dots indicate similarity (generated by the program GCG "BestFit" (Genetics Computer Group, Madison, WI) using the local homology algorithm of Smith and Waterman (26)). 20), N. crassa (25), and Saccharomyces cerevisiae (21), in the catalysis of a complex post-translational modification reaction. Primary structure is highly conserved in the major portion of human and yeast deoxyhypusine synthase molecules, i.e between Thr 98 and Ile 330 of the human sequence. The known sequences of peptides from the rat (20) and Neurospora (25) enzymes also show a marked similarity to the primary structures of the human and yeast enzymes in this region. The maintenance of protein structure in these enzymes may have been mandated for the recognition of their three substrates, spermidine, NAD, and especially the highly conserved eIF-5A precursor, and, in turn, may have served to preserve a vital cellular function of eIF-5A. Despite the strict specificity of deoxyhypusine synthases from all species for this single cellular protein, cross-species reactivities between the enzymes and the eIF-5A precursor proteins are observed. As previously shown (21) for the yeast and rat enzymes, human deoxyhypusine synthase also catalyzes the modification of heterologous protein substrates, namely eIF-5A precursors from Chinese hamster ovary cells and yeast (data not shown).
It is intriguing that the initial clone, hDS-1a, isolated by means of immunological screening, has an internal deletion that results in the loss of 56 amino acids (Asp 262 -Ser 317 ). It was unclear whether this in-frame deletion reflects natural cellular processing of the deoxyhypusine synthase gene transcript or whether it arose as an artifact of expression library construction. Since no other clone with the same deletion was detected,  (27). Both prehybridization and hybridization were conducted at 42°C in 6 ϫ SSC containing 50% formamide, 5 ϫ Denhardt's solution, 100 g/ml yeast tRNA, and 0.1% SDS. After overnight hybridization with a cDNA probe, which was generated by PCR of hDS-1a cDNA using primers Pr5-2 and Pr3-5 and labeled with 32 P, the membranes were washed at 23°C for 10 -20 min three times with 2 ϫ SSC containing 0.1% SDS and once with 1 ϫ SSC containing 0.1% SDS and then washed at 55-60°C for 20 min three times with 0.2 ϫ SSC containing 0.1% SDS. In A, 144 JAR is a human trophoblast cell line. In B, T 45 (28), HN 30 (29), HN 22 (29), HN 8 (29), and SCC 25 (30) are derived from oral squamous cell carcinomas; DOK (31) from premalignant keratinocytes; and NHEK (Clonetics Corp.) from normal human epidermal keratinocytes (source or reference given in parentheses). Kb, kilobases.

TABLE III
Purification of recombinant human deoxyhypusine synthase Purification was from 20 g of E. coli cells that had expressed deoxyhypusine synthase, as described under "Experimental Procedures." One unit is defined as the amount of enzyme that catalyzes the formation of 1 pmol/h deoxyhypusine (equivalent to 1.67 ϫ 10 Ϫ8 mol/min).
Step  Table III was   human, rat, and yeast In the case of the human and yeast enzymes, highly purified recombinant enzyme from the respective cDNA expressed in E. coli was used. The rat enzyme was purified from testis (20). Human eIF-5A precursor protein (ec-eIF-5A) was used as the substrate for all determinations. Previously published values are given for the rat (7,17,20) and yeast (21)  we sought to determine the existence of a natural mRNA(s) with this deletion. Using different sets of primers flanking the deletion region, PCR amplification of a HeLa cDNA mixture (HeLa QuickClone cDNA (CLONTECH) or cDNA prepared by a reverse transcriptase reaction of total RNA from HeLa cells) as template thus far has provided no clear indication of the existence of a natural transcript with the corresponding deletion (data not shown). Determination of the genomic structure of deoxyhypusine synthase might offer insight as to whether such a transcript can be generated by alternative splicing. A mutant protein expressed from hDS m3, lacking Asp 262 -Ser 317 (corresponding to the internal coding region missing in hDS-1a), displayed no detectable deoxyhypusine synthase activity. This lack of activity is not surprising since the deletion removed a highly conserved region of the protein (Fig. 2) and strongly indicates that this region is critical for enzyme activity. The mutant proteins expressed from clones hDS m4, hDS m5, and hDS m6 (Table II), with truncations of 48 or 97 NH 2terminal amino acids (Met 1 -Ala 48 or Met 1 -Cys 97 ) or of 39 COOH-terminal amino acids (Asp 333 -Asp 369 ), respectively, were also devoid of enzymatic activity, suggesting that, although not highly conserved, these regions also contain amino acid residues important for the proper conformation of the enzyme, necessary for binding substrates and/or for catalysis.
The most interesting difference between the primary structures of the human and yeast enzymes is the absence of a human sequence matching Gln 197 -Asn 212 in the yeast enzyme. Although this gap (indicated by . . . . in Fig. 2) is located in the middle of a highly conserved region, the human recombinant deoxyhypusine synthase exhibits as high a specific activity as the yeast enzyme. A chimeric protein with the 16-amino acid yeast sequence inserted between Glu 193 and Gln 194 of the human enzyme exhibited moderate activity (Table II). Since the insertion of this partial yeast sequence into the human enzyme did not result in a dramatic increase in its K m value for NAD, one cannot attribute the large differences in affinities toward NAD specifically to this region of the molecule. On the other hand, it is interesting that this insertion does have a significantly detrimental effect on the human enzyme activity.
Evidence for a vital role of hypusine in eukaryotic cell proliferation has led to proposed targeting of the catalytic steps in hypusine production as potential methods for antiproliferative therapy (1,12,18,19,32). Promising results on the intervention in growth of mammalian cells, including various human tumor cell lines, have been obtained with inhibitors developed against the rat testis deoxyhypusine synthase (17,19). Although some details concerning the steps in the biosynthesis of deoxyhypusine have been revealed (1), little is known about the precise mechanism of the reaction. The availability of a human cDNA clone and its encoded protein should enable us to determine structural features of the active site and of the substrate binding domains as well as the residues involved in catalysis. A better understanding of the enzyme structure and the reaction mechanism will facilitate the development of potent and specific inhibitors of the human enzyme and pave the way to improved means of cellular regulation through the inhibition of hypusine biosynthesis.