Chapter 26

Biosynthesis of Proline

THOMAS LEISINGER

INTRODUCTION

During the 1960s and 1970s, when studies on the biochemistry and regulation of amino acid biosyntheses flourished, detailed information on proline biosynthesis accumulated rather slowly. This was due to the difficulties encountered in developing sensitive and specific assays for the first two enzymes of this pathway and to the fact that the enzyme levels in the system did not respond to changes in the concentration of proline in the growth medium. Proline biosynthesis in enterobacteria was therefore not suitable for studies on gene expression, and the main challenges in the system were the elucidation of the enzymology and the formal proof of a pathway in which reactions and intermediates were already generally accepted on the basis of circumstantial evidence.

The pathway from glutamate via glutamate γ-semialdehyde (GSA) and its spontaneous cyclization product, Δ ¹-pyrroline-5-carboxylate (P5C), to proline (Fig. 1) was first proposed by Vogel and Davis (74) and was based on the observation that the accumulation of P5C by certain Escherichia coli proline auxotrophs permitted growth of another type of proline auxotroph. Isotope dilution experiments showed that unlabeled GSA decreased the incorporation of [^l4C]glutamate into proline (75) and thus supported the proposed route. A further important stage in the development of the model for proline biosynthesis was the demonstration that feedback regulation of the pathway by proline is at some stage between glutamate and GSA (1). This conversion had been proposed to involve an activated glutamate such as γ-glutamyl phosphate. Baich (5) provided evidence for a proline-inhibitable enzyme activity catalyzing the ATP-dependent phosphorylation of glutamic acid. An E. coli mutant resistant to inhibition by a proline analog, 3,4-dehydroproline, in vivo contained a γ-glutamyl kinase with markedly lowered sensitivity to proline. This finding supported the existence of a γ-glutamyl kinase specifically involved in the first step of proline biosynthesis and has led to the formulation of the pathway in its present form (Fig. 1). The enzymes of this pathway have also been demonstrated in other prokaryotes such as Serratia marcescens (71), Pseudomonas aeruginosa (46, 47, 66), and Brevibacterium flavum. Proline formation from glutamate is thought to occur in plants and in a variety of animal tissues. However, in eukaryotes, there is also a second route for proline synthesis that leads from ornithine via P5C to proline (1).

Fig. 1Proline biosynthetic pathway. The coordinates for the gene positions are those of the E. coli physical map.

Source:

An extensive review by Adams and Frank (1) on the metabolism of proline and the hydroxyprolines in microbes and higher organisms summarizes the state of the art in 1979. The enzymology and genetics of proline biosynthesis in E. coli were last reviewed in the first edition of this book (50). In the meantime, a very modest amount of new information on the biosynthetic pathway has accumulated, and interest in proline has shifted to its roles as an osmoprotectant and as a carbon and nitrogen source. This chapter thus will recapitulate the established findings on the biosynthetic pathway as well as consider new data on proline metabolism which are of relevance for osmoregulation (see chapter 77) and regulation of proline utilization.

THE PATHWAY

γ-Glutamyl Kinase

γ-Glutamyl kinase is postulated to catalyze the ATP-dependent phosphorylation of the γ-carboxy group of l-glutamic acid. Three types of experimental difficulties inherent to the system have hampered studies on this reaction in crude extracts and the purification of the enzyme.

First, the extreme lability of γ-glutamyl phosphate and its tendency to cyclize to 5-oxopyrrolidine-2-carboxylate (70) have prevented the direct identification of the product of the first enzymatic step in any experimental system. However, the enzyme-dependent formation of γ-glutamyl hydroxamate and P_i from glutamate, ATP, Mg²⁺, and hydroxylamine (5) is compatible with γ-glutamyl phosphate being the first intermediate of the pathway. When the GSA dehydrogenase reaction, the second step in the pathway, is examined with homogeneous enzyme in the reverse of its biosynthetic direction, the end product is 5-oxopyrrolidine-2-carboxylic acid. Since the latter compound arises from γ-glutamyl phosphate, this observation is in accordance with the scheme in Fig. 1, in which γ-glutamyl phosphate is depicted as the activated intermediate between the first and the second biosynthetic steps (36). Considerations on the lability of free γ-glutamyl phosphate have led to the notion that the compound exists in vivo as an enzyme-bound intermediate. Recent work with the glutamate analog cis-cycloglutamate supports the formation of γ-glutamyl phosphate in the first enzymatic step and suggests that this intermediate reacts with a moiety (such as a thiol) on GSA dehydrogenase to form a γ-glutamyl-enzyme complex (67).

Second, the proline-specific γ-glutamyl kinase catalyzes the same activation reaction of glutamate as glutamine synthetase. Although the two activities can be differentiated on the basis of their sensitivities to proline (5, 39), the presence of both enzymes in crude extracts of E. coli makes the assay of the first enzyme of proline biosynthesis rather imprecise. With wild-type extracts and the classical assay, which is based on measuring the formation of γ-glutamyl hydroxamate from glutamate, ATP, and hydroxylamine, blank values due to glutamine synthetase activity are up to 10-fold higher than the activity due to the proline-specific γ-glutamyl kinase.

The third obstacle to an understanding of γ-glutamyl kinase was its association with the second enzyme of proline biosynthesis. Smith et al. (68) have shown that to be enzymatically active in the hydroxamate assay, γ-glutamyl kinase must be associated with GSA dehydrogenase. This association is not required when the activity of γ-glutamyl kinase is assayed by monitoring the glutamate-dependent formation of P_i from ATP, which relies on the rapid hydrolyis in solution of γ-glutamyl phosphate to glutamate and P_i (67, 72). The existence of an enzyme complex catalyzing sequential reactions in proline biosynthesis and ensuring the direct transfer of an unstable intermediate had been suggested before as a result of gel filtration studies with crude extracts, and it appeared that the complex is very labile in vitro (39). In this context, it is interesting that in the plant Vigna aconitifolia, the first two steps of proline biosynthesis are catalyzed by the bifunctional enzyme P5C synthetase. The two enzymatic domains of this protein correspond to the ProB and ProA proteins of E. coli, and each contains a leucine zipper, which may facilitate inter- or intramolecular interaction (42).

It is now clear that early attempts to purify γ-glutamyl kinase were unsuccessful because they led to the dissociation of the two enzymes with a consequent loss of hydroxamate-forming activity of the first enzyme. Purification of γ-glutamyl kinase from E. coli (68) became possible when a strain containing the proB and proA genes on a multicopy expression vector (25) yielded starting material of high specific activity and when a coupled enzyme assay based on the NADPH-dependent reduction of γ-glutamyl phosphate by GSA dehydrogenase was used throughout the purification procedure. In this assay, a large excess of highly purified GSA dehydrogenase over γ-glutamyl kinase satisfied the requirement of γ-glutamyl kinase for the second enzyme in the pathway. γ-Glutamyl kinase appears to be a hexamer composed of identical subunits. The activity of the enzyme depends in a nonhyperbolic fashion on the glutamate concentration, and depending on the source of enzyme, 15 to 40 mM l-glutamate was found to be necessary for half-maximal activity. l-Proline at 0.007 mM (0.06 mM [67]) or 3,4-dehydroproline at 0.5 mM was needed to inhibit the enzyme activity by 50% (68). Among a range of analogs of glutamate, α-methyl-dl-glutamate and γ-methyl-dl-glutamate exhibited minor activity as substrates. Only cis-cycloglutamate (cis-1-amino-1,3-dicarboxycyclohexane) was a good substrate for γ-glutamyl kinase, displaying hyperbolic saturation kinetics with a K_m value of 1 mM (67).

GSA Dehydrogenase

GSA dehydrogenase, the second enzyme in the pathway, catalyzes the in vivo NADPH-dependent reduction of γ-glutamyl phosphate to GSA. Since γ-glutamyl phosphate is not available as a substrate in vitro, enzyme activity is measured in the reverse direction by the NADP- and phosphate-dependent oxidation of GSA. The actual substrate used in this assay is P5C, a rather labile compound, which is in equilibrium with the straight-chain GSA that probably serves as the true substrate of the enzyme. dl-P5C is most conveniently prepared by periodate oxidation of δ-hydroxylysine (76). An enzymatic method for the preparation of l-P5C from l-ornithine with a partially purified preparation of ornithine 5-aminotransferase from P. aeruginosa has also been described (34).

GSA dehydrogenase from E. coli was first purified from cell material containing a ColE1-proBA plasmid, and 1,200-fold purification was needed to obtain a homogeneous product (37). A more favorable situation was created by using cells carrying expression vectors with the proBA region for enzyme purification. A 450-fold amplification of GSA dehydrogenase with a correspondingly lowered purification factor was achieved (25, 67, 68). Different values for the molecular weight of the native protein have been obtained, but overall they suggest a hexameric subunit arrangement. Studies with the pure enzyme showed that the reaction is highly specific for dl-P5C and NADP but that a number of divalent anions can substitute for phosphate. The kinetics of the reaction are consistent with a rapid-equilibrium, random-order mechanism (38).

P5C Reductase

The final step in proline biosynthesis, the NADPH-linked reduction of P5C to proline, is catalyzed by P5C reductase. In crude extracts of E. coli cells, this enzyme interferes with the assay for GSA dehydrogenase activity: both enzymes use P5C as a substrate, and NADPH, a product of the GSA dehydrogenase reaction, serves as a cosubstrate for P5C reductase. To assign the proA and proB loci to their respective enzymes, a method for the rapid removal of P5C reductase from cell extracts was devised. It permitted the establishment of the gene-enzyme relationships in a number of proA and proB mutants (35).

P5C reductase from an E. coli wild-type strain was enriched 200-fold to give a partially purified preparation that was sufficiently free of competing enzyme activities to be used in kinetic studies (62). The enzyme exhibited Michaelis-Menten kinetics, with K_m values of 0.15 and 0.03 mM for dl-P5C and NADPH, respectively. Proline and NADP, the end products of the reaction, act as competitive inhibitors with approximate K_i values of 15.0 and 0.6 mM, respectively. P5C reductase was later purified to homogeneity from an E. coli strain harboring the expression plasmid pGW7 proC. The 200-fold increase in P5C reductase activity relative to the wild-type level that was obtained in the plasmid-bearing strain greatly facilitated enzyme purification. The homogeneous enzyme preparation was used to determine the amino- and carboxy-terminal amino acid sequences but was not characterized with respect to its catalytic properties (26).

Proline Auxotrophs

Two largely unexplained observations have been made that indicate that under certain conditions, a deficiency in proline synthesis does increase the survival rate relative to the prototrophic wild type or to mutants blocked in other biosynthetic pathways. They are mentioned here because they may be useful in the isolation of proline auxotrophs. 4-Nitropyridine 1-oxide is bactericidal for E. coli wild-type and proC cells but not for proB or proA mutants. This effect, the biochemical basis of which is unknown, can be conveniently used for the selection of proA or proB mutants (43). Preferential recovery of all three types of proline auxotrophs is also observed upon penicillin enrichment of a mutagenized population. The viable counts of resting cells of most types of auxotrophs are reduced by penicillin treatment, whereas resting cells of proline-requiring mutants for unknown reasons are completely insensitive to the antibiotic (61).

Second Route to GSA

The biosynthesis of arginine from glutamate is initiated by an N acetylation that is thought to prevent the internal cyclization of the N-acetylglutamate semialdehyde that is formed in the third biosynthetic step (Fig. 2). Activation of N-acetylglutamate and the reduction of N-acetylglutamyl phosphate in the arginine pathway are analogous to the first and the second steps of proline biosynthesis. The similarity of the early intermediates in the two pathways forms the basis for the phenotypic suppression of mutations in proB or proA by mutations in the arginine pathway. This phenomenon has been observed in E. coli (7, 44) and in Salmonella typhimurium (official designation, Salmonella enterica serovar Typhimurium) (48).

Fig. 2The relationship of proline biosynthesis to the pathways for arginine biosynthesis and proline catabolism. argA to argE are the structural genes specifying steps 1 to 5 of arginine biosynthesis. putA specifies a bifunctional dehydrogenase that oxidizes proline to P5C (proline dehydrogenase activity) and P5C to glutamate (P5C dehydrogenase activity).

Source:

argD mutants accumulate N-acetylglutamate γ-semialdehyde, which is converted into GSA by N-acetylornithine deacetylase, the product of argE catalyzing the fifth step in arginine biosynthesis (Fig. 2). Such mutants are able to grow in the absence of arginine because some nonspecific aminotransferases substitute for the missing argD enzyme. In proB (or proA) argD mutants, the block in proline biosynthesis is bypassed since the first three enzymes of the arginine pathway and N-acetylornithine deacetylase lead to the formation of GSA. Proline biosynthesis in these double mutants is no longer feedback inhibited by proline but is now under repression and feedback control by arginine. The double mutants excrete proline on minimal medium and are resistant to the proline analogs 3,4-dehydroproline and azetidine-2-carboxylate (7). The interrelationship between arginine and proline biosynthesis has been used for the selection of mutants containing feedback-resistant N-acetylglutamate synthase, the first enzyme of arginine biosynthesis (28). A nonrepressible proB (or proA) argD argR mutant grows slowly on minimal medium with arginine. As a result of the partial inhibition of N-acetylglutamate synthase by arginine, it does not excrete proline. Cultivation of the triple mutant on minimal medium with arginine enriches for faster-growing quadruple mutants containing a feedback-resistant N-acetylglutamate synthase. They can be recognized as proline excretors on minimal agar with arginine.

Organization of the pro Genes

The sequences of the E. coli genes encoding the three polypeptides engaged in proline biosynthesis have been determined, and the sequences of genes for proline transport and proline degradation have also been determined (Table 1). References to the literature describing the mapping of the genes of proline biosynthesis in E. coli are given by Bachmann (4). A physical map of the chromosomal segment extending from approximately 5.0 to 12.5 min and covering the pro genes has been prepared (33) and corresponds well with the locations of proBA at 5.8 min and of proC at 8.9 min on the genetic map (4). In S. typhimurium, the three pro genes are arranged in a similar fashion at the same position of the linkage map (65). It was established that in this organism, as in E. coli, the proB gene codes for γ-glutamyl kinase and proA codes for GSA dehydrogenase (52).

Table 1Genes and gene products of proline metabolism in E. coli and S. typhimurium

The E. coli proBA genes have been cloned from ColE1 hybrid plasmids from the collection of Clarke and Carbon (17) or an F' proBA plasmid as the initial material (25, 52). They are encoded on a 3.0-kb PstI DNA fragment whose nucleotide sequence has been determined (25). The sequences of the proline biosynthesis genes of S. typhimurium have not been determined. However, complementation of transposon insertions into the chromosome of S. typhimurium revealed a strong polar effect of proB on proA, suggesting that the proB and proA genes of this organism are organized in an operon that is transcribed from proB to proA (52).

Transcription studies have not been carried out, but the following features of the nucleotide sequence of the E. coli proBA region are in accordance with the concept of an operon. (i) The DNA sequence upstream from the proB coding region contains various domains of dyad symmetry and a putative ribosome binding site. (ii) The C terminus of proB is separated by 14 bp from the N terminus of proA. There are a single translation termination codon and two ribosome binding sites within this short intercistronic region. (iii) A putative transcription termination structure at the 3' end of the proA gene can be identified, whereas there is no indication for a terminator at the 3' end of the proB gene. Support for an operonic structure of proBA comes from a comparative study on these genes in the related enterobacterium Serratia marcescens. In this organism, the proBA DNA sequence encodes proteins exhibiting 88 and 74% sequence identity with γ-glutamyl kinase and GSA dehydrogenase, respectively, of E. coli. A transcription start site downstream of a region similar to the E. coli σ ⁷⁰ consensus promoter was mapped in front of proB, and the size of the intercistronic region as well as the presence of a rho-independent terminator downstream of proA corresponded with the organization in E. coli (56).

The E. coli proC gene has been located on a 1.1-kb HincII-HaeII fragment of a λ transducing phage, with its HincII site originating in the λ vector. Determination of its nucleotide sequence (26) revealed an open reading frame for a polypeptide with a molecular weight of 28,112. This value correlates well with the molecular weight of the P5C reductase subunit that was determined by protein purification. Analysis of the amino-terminal sequence of purified P5C reductase supports the notion that translation starts at the proposed site, which is also preceded by a ribosome binding site. However, the fact that 94 nucleotides on the noncoding strand of the lacA structural gene have been found to be identical to nucleotides –119 to –25, preceding the translation start of proC, casts some doubt on the authenticity of the putative proC promoter region and suggests that a segment of lacA DNA was mistakenly incorporated into the proC region during construction of the proC clones used in the sequence studies (40).

REGULATION

Control of Proline Synthesis

Feedback inhibition of γ-glutamyl kinase was early recognized to be the major control mechanism for proline biosynthesis in E. coli. It was first demonstrated in resting cells of a proC mutant of E. coli by comparing the conversion rates of glutamate into GSA in the presence and absence of proline (6, 70). Later the sensitivity of the first enzyme to proline was also observed in crude extracts (5, 39). Studies with purified γ-glutamyl kinase (68) suggest that the activity of this enzyme is modulated not only by proline but also by glutamate and by ADP. The sigmoidal saturation kinetics of γ-glutamyl kinase with respect to glutamate lead to its relative insensitivity to small changes in the concentration of this substrate, thereby ensuring a constant rate of proline biosynthesis at varying intracellular concentrations of glutamate. ADP, a competitive inhibitor with respect to ATP, inhibits the enzyme under conditions of energy depletion and thus may lead to the tuning of proline biosynthesis with the energy charge of the cell. Proline decreases the affinity of the enzyme for glutamate at low or intermediate concentrations of this metabolite. In a mutant with abolished proline sensitivity, the modulation of enzyme activity by glutamate and ADP is conserved. Since this mutant excretes proline, the end product of the pathway is thought to be the most important effector for pathway control.

Two independently isolated E. coli mutants resistant to feedback inhibition by proline (68, 72) and a similar mutant of Serratia marcescens (56) have been characterized. These strains carry single base changes in proB which render their γ-glutamyl kinases between 100- to 700-fold less sensitive to proline than the wild-type enzymes. The amino acid substitutions resulting from the base changes were located at amino acid position 107, 117, or 143 of the γ-glutamyl kinase, thus indicating that single changes in different amino acid residues provoke decreased sensitivity to allosteric feedback inhibition (22, 57, 64).

The question of whether the synthesis of the proline enzymes in enterobacteria is subject to repression by the end product of the pathway has been addressed in several laboratories (8, 10, 18, 35, 41, 56, 62). Enzyme levels of cells grown on media without proline or under proline starvation showed no significant differences from those of cells grown on media with proline excess. Absence of end product repression in the system was also observed with proB-, proA-, or proC-lacZ fusions. The available evidence thus indicates that none of the three proline biosynthetic genes is subject to significant expression control in response to proline limitation or excess.

Overproduction of Proline

Various proline analogs have been shown to interfere at different stages with protein synthesis in E. coli (Table 2). Azetidine-2-carboxylate, 3,4-dehydroproline, and thiazolidine-4-carboxylate, analogs inhibiting γ-glutamyl kinase but supporting protein synthesis, have proven useful in the genetic analysis of the proline system (18). Resistance to these agents is based either on a deficiency in proline uptake or on proline overproduction due to defective pathway control. Mutants with resistance of the latter type are easily recognized as proline excretors. In all cases, the corresponding mutations are closely linked to proB (18, 19). This observation suggests that proline overproduction is the consequence of decreased sensitivity to proline inhibition of γ-glutamyl kinase and confirms the notion that feedback inhibition is the major mechanism for the control of proline synthesis.

Table 2Effects of proline analogs on E. coli

Preliminary data on the use of a genetically engineered E. coli strain for the overproduction of l-proline are available. A yield of 27 g of proline per liter with a 40% conversion of glucose to proline is reported (8). This productivity is achieved by a strain that is unable to degrade proline because of a chromosomal deletion of putA, the gene coding for proline dehydrogenase/P5C dehydrogenase. In addition, it carries a ColE1-based vector with the cloned proBA operon as well as the proC gene. Furthermore, the proB gene on this plasmid specifies a feedback-resistant γ-glutamyl kinase. One might suppose that such a high yield of proline would be possible only in a strain that had an enhanced potential for forming glutamate, the precursor of proline, but details on whether this was an obstacle that had to be overcome in the strain have not been reported.

Proline-overproducing derivatives of a putA Serratia marcescens strain have been obtained by selection of mutants resistant to proline analogs. One of these mutants is reported to produce 60 g of proline per liter. Transductional and DNA sequence analysis revealed that this strain carries at least four mutations that together enable a high rate of proline synthesis: putA, which prevents proline degradation; an unknown mutation leading to an increased level of glutamate dehydrogenase; a base change in proB rendering γ-glutamyl kinase 700-fold less sensitive to proline than the wild-type enzyme; and a base change in the spacer region of the proBA promoter which increases the transcriptional activity of this operon fourfold (57).

PROLINE TRANSPORT

In E. coli and S. typhimurium, three independent transport systems mediate the transport of proline. The PutP system is essential if proline is to be utilized as a carbon or nitrogen source, whereas the three-component ProU and the one-component ProP systems are required for the accumulation of exogenous proline as an osmoprotectant under conditions of osmotic stress (Table 1). The latter two transporters also mediate the uptake of glycinebetaine (N,N,N-trimethylglycine), a second important osmoprotectant in bacteria (23) (see chapter 77).

Proline auxotrophy can be satisfied not only through the activity of PutP, the major proline permease, but also by the osmoregulated transporters encoded by the proU operon and the proP gene. This has been shown by the analysis of mutants deleted for the genes encoding proline transport systems. For example, a proline auxotrophic E. coli construct deleted for the structural genes of the three transport systems [Δ(proC, putP, proP, proU)] required 200 mM l-proline for growth on LB medium. Introduction of a plasmid carrying the cloned proP gene in this strain restored the ability to grow in minimal medium with 25 μM l-proline (24). Similarly, the functionality of the ProU system in transporting proline for biosynthesis was demonstrated in a proline auxotrophic S. typhimurium mutant lacking the PutP and ProP transporters. This strain was able to grow with 100 μM l-proline in a medium containing 0.3 mM sodium chloride. Much higher concentrations of proline were needed for growth in a medium of low osmolarity. In contrast, recruitment of the ProP transporter for the uptake of proline for biosynthesis did not require high osmolarity (20). This is explained by the fact that expression of the ProU system is induced several hundred-fold by growth in high-osmolarity media whereas there is only a threefold stimulation of proP transcription upon osmotic stress (27, 69).

The proline utilization (put) operon in E. coli and S. typhimurium consists of two divergently transcribed genes, putP and putA, separated by the 420-bp put control region. putP encodes a 54-kDa integral membrane protein (54, 55) mediating the Na⁺ symport of proline(15). The putA gene codes for a 132-kDa protein (2) that has been purified from S. typhimurium (53) and E. coli (11). It associates with the inner surface of the cytoplasmic membrane and has two enzymatic activities—proline dehydrogenase and P5C dehydrogenase—which catalyze the oxidation of proline to glutamate for use as the sole carbon, nitrogen, or energy source (Fig. 1). The put operon is induced 10- to 20-fold by proline. A recent model proposes that the S. typhimurium PutA protein autogenously regulates expression of the put operon by specifically binding to DNA in the put control region. Thus, in the absence of proline, PutA is thought to be located in the cytoplasm and to act as a transcriptional repressor. In the presence of proline, it associates with the membrane and assumes its function in the oxidation of proline to glutamate (58).

To avoid a futile cycle of proline biosynthesis and degradation, synthesis and activity of the PutP and PutA proteins must be tuned to both the proline biosynthetic pool and the pool of proline accumulated for osmoprotection. For S. typhimurium, it has been shown that proline does not need to be transported through PutP to induce the put operon. Endogenously synthesized proline also induces the operon, but only at concentrations greater than that of the normal biosynthetic proline pool. Degradation by the PutA protein of the high-proline pools established under conditions of hyperosmotic stress is apparently prevented by the fact that the proline dehydrogenase activity of PutA is severely inhibited under high-salt conditions (29).

PROLYL-tRNA SYNTHETASE

Prolyl-tRNA synthetase of E. coli exists in vitro as an inactive monomer or as an active dimer that is stabilized by ATP or tRNA. A subunit molecular weight of 47,000 has been demonstrated by centrifugation in density gradients (49), and the molecular weight deduced from the DNA sequence of the proS gene amounts to 63,701 (30). The K_m values of prolyl-tRNA synthetase are 3 × 10^–5 M for ATP and 2.3 × 10^–4 for proline (60). The enzyme exhibits a broad specificity and esterifies the proline analogs l-thiazolidine-4-carboxylate, allo-4-hydroxy-l-proline, and l-azetidine-2-carboxylate. Open-chain compounds having structural similarities with portions of the proline molecule, such as N-methyl-, N-ethyl-, and N-propylglycine and N-methyl-l-alanine, are also esterified to tRNA^Pro (59).

proS, the structural gene of prolyl-tRNA synthetase, was cloned, and its sequence was determined as one of the last of the 20 aminoacyl-tRNA synthetase genes of E. coli. It was isolated by complementing a temperature-sensitive mutant defective in proS (9). The open reading frame encoding prolyl-tRNA synthetase is preceded by a σ ⁷⁰ promoter and followed by a rho-independent termination signal. The amino acid sequence of the enzyme shows homologies with the sequences of threonyl-tRNA synthetase (44.8% similarity) and seryl-tRNA synthetase (37.2% similarity). This comparison was instrumental for the discovery of three conserved motifs which subsequently were also detected in some of the other aminoacyl-tRNA synthetases and thus led to the definition of class II synthetases (30).

Prolyl-tRNA synthetase belongs to the group of seven synthetases that are transiently derepressed under conditions of starvation for the cognate amino acid. In the case of proline, derepression is three- to fourfold (3). The molecular mechanisms underlying the derepression effect have been studied for some synthetases (chapter 91) but not for prolyl-tRNA synthetase.