The histidine-binding protein J is a component of histidine transport. Identification of its structural gene, hisJ.

Abstract The histidine-binding protein J, previously shown to be involved in histidine transport in Salmonella typhimurium (Ames, G. F., and Lever, J. (1970) Proc. Nat. Acad. Sci. U. S. A. 66, 1096), is shown unequivocally to be the product of the hisJ gene. A hisJ mutant with an altered J protein and a correspondingly altered histidine transport has been isolated and characterized. The J protein from this strain has an increased temperature sensitivity, besides having altered chromatographic and electrophoretic mobilities. The in vivo effect of the altered J protein is expressed as an increased temperature sensitivity of histidine transport. Our data indicate that the hisJ gene is the structural gene for the J protein and that the J protein is an obligatory component of histidine transport. As expected, there is an excellent correlation between the specificity of transport in the wild type strain and the specificity of binding of the wild type J protein for a variety of amino acids, amino acid analogues, and inhibitors.

G. F., AND LEVER, J. (1970) PYOC. Nat. Acad. Sci. U. S. A. 66, 1096), is shown unequivocally tc be the product of the hisJ gene. A hisJ mutant with an altered J protein and a correspondingly altered histidine transport has been isolated and characterized. The J protein from this strain has an increased temperature sensitivity, besides having altered chromatographic and electrophoretic mobilities. The in vivo effect of the altered J protein is expressed as an increased temperature sensitivity of histidine transport. Our data indicate that the his/ gene is the structural gene for the J protein and that the J protein is an obligatory component of histidine transport.
As expected, there is an excellent correlation between the specificity of transport in the wild type strain and the specificity of binding of the wild type J protein for a variety of amino acids, amino acid analogues, and inhibitors.
Sal~~onella fypllimuriunz transports L-hi&dine through at least five permeases with different patterns of specificity and affinity (l-3).
The histidine permease with the highest affinity (J-l' permease, wit,h a I<, of about 10-8 ?VI) has been show11 to be composed of at least two proteins, J and 1' (3), and has beeu analyzed kinet,ically, biochemirall,y, and .genet,ically. 'l'he J protein is a periplasmic histidine-binding protein, released by osmotic shock (3, 4) and coded for by the hisJ gene (3). The P protein is essential for histidine transport by the J protein and is coded for by the hid' gene (2, 3) but has not been identified biochemically.
The P protein is also necessary for the functioning of another histidine permease (the K-P permease, with a K, of about 10e7 M), which works in parallel to the J-P system (3).
We demonstrated earlier a direct correlation between the * This research was supported by United States Public IIenlth Service Grant,s AM12121 and AM12092.
1: Present address, Department of Medicine, HSB 2034, University of California, San Diego, La Jolla, California 92037. Part of these studies are taken from a thesis snbmitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biochemistry at the University of California, Berkeley.
activity of the histidille permease and the levels of the ,J protein (3). Strains with a mut&ion in the hisJ gene are defective in histidine t~ransport, and correspondingly lack the ,J protein. Strains with :I mutation iu the d/~l~ri sit,?, which is thought to be a control locus for histidine transport, have elevated histiclille transl)ort aud are correspondingly elevated in tire J Ixotein. The proof that hisJ mutations are in the struct,ural ge~~r for the J protein, rather than in a control locus for the genes of histiditre transport, was obtained by the c,h:lracterization of a temperaturesensitive mutation which simultaneously alters both the J I)rotein and transport through the JI' permease.
This mutant is described in detail in this paper together with other evidence ou the role of the .J protein in transport.
The J-P permease also transl)orts other substances (among them u-histidine, L-arginine, n-2-hydrazino-3-(4-imidazolyl)propionic acid) with lower afinity; the correlation between the specificity of transport and the binding specificity of the J protein is presented.
The purificat,ion and some of the properties of the xvild type J protein have beeu described (4).

EXPIZRIMFNTAL PROCEDURE
Chemicals-All chemica!s Iron1 commercial sources \vere of the highest purity available.
Histidine concentration was verified by amino acid analysis, and radiochemical purity was monitored by paper chromatography in the two I-propanol systems above, using radioautography for detection. Possible histidine contamination in amino acids and analogues used in the competition experiments was monitored using a Beckmau/Spinco model 121 amino acid analyzer with espanded range. Imidazole pyruvic acid (Calbiochem), I,-histidinol (Cycle Chemical Company), and n-histidylhistidine (Mann) were purified using descending preparative paper chromatography with Whatman No. 3MM paper, in a solvent system of equal parts collidine-2,4-lutidine-HZO, made 2% in diethylamine freshly before use. Compounds were eluted from the paper with water and lyophilized to dryness.
Assays-n-Histidine uptake was measured by the "growing cells" method (1) (10 to 20 pg, dry wt, of cells per ml), with the modification for the washing of the samples as already described (3). The J protein was purified, assayed, and characterized as described (3, 4). [3H]Histidine binding activity was assayed either by the dialysis or by the filtration method (3,4).
Bacterial Strains and Media-All strains were derived from S. typhimurium strain LT-2 and are listed in Table I. Conditions for growth have been described (2). Except for TA271 which was obtained from T. Klopotowski (9), all strains were constructed in this laboratory and analyzed genetically by transduction with phage P22 int-4 as described earlier (2). Growth on n-histidine (1 pmole on disc) and sensitivity to D-HIPA (in the presence of n-histidinol for histidine auxotrophs) were assayed on Petri plates by the radial streak method (2).
Genetic Analysis of Bacterial strains-Strains TA271 (dhuA1 hisF645) and TA1014 (dhuAl) have been described (3). Strain TA1768 (dhuA1 hisJ561?' hisF6.45) was obtained by mutagenizing TA271 with the frameshift mutagen, , and then selecting for a mutation in hisJ (such mutation renders the strain unable to grow on n-hi&dine, but still D-HIPA sensitive), as described previously (3). Mutation hisJ5617 so obtained was shown to be closely linked (about 98%) to dhuAl by the same methods used for other hisJ mutations (3). Strain TA1789 (dhuA1 hisJ5617 hisJ5620 hisF645; briefly mentioned in (3)) was isolated as a revertant of TA1768, which was capable of growing on n-histidine at room temperature, but not at 40".
Strains TA1771 and TA1791 were obtained by transducing the 1. Temperature sensitivity of L-histidine uptake in the hisJ revertant TA1791. A, initial rates of uptake of L-histidine (micromoles per min per g, dry wt) as a function of temperature. B, L-histidine uptake through the J component as a function of temperature.
The initial rates of uptake at each temperature by TA1771 (i.e. through the K component (3)) are subtracted from the corresponding rates of uptake by TA1014 and TA1791. The rates thus corrected (uptake through the J component) are expressed as percentages, taking the corrected rate of uptake by TA1014 as 100% at each temperature.

RESULTS
Mutant Strain with Qualitative Albration of Both J Protein and Histidine Transport-Strain TA1789 (dhuAl hisJ5617 hisJ5620 hisF645) is a revertant of TA1768 (dhuAl hisJ5617 hisF645) and was chosen for further study because of its temperaturesensitive phenotype.
It carries two mutations in hisJ, both induced by the frameshift mutagen : the initial mutation, hisJ5617, causes loss of both the J component of transport and the J protein, and hisJ5620, a mutation at a second site in the hisJ gene, restores partial J protein function.
Bs a consequence of the double mutation in the hid gene, strain TA1789 (and its prototrophic derivative TA1791) has a temperature-sensitive J component of transport which can be analyzed by measuring either L-histidine uptake or the ability to grow on n-histidine.
Concomitant!y with these altered properties of transport, TA1789 (and TA1791) produces a J protein with several altered properties including an increased temperature sensit.ivity.
Genetic Mapping-Strain TA1789 (dhuA1 hisJ5617 hisJ5620 hisF645) was shown by the following genetic tests to contain the mutation, hisJ5620, responsible for the second site reversion, in the hid gene. Mutation hisJ5620 was first shown to be cotransducible (42%) with purF1.4.5, by growing phage on TA1789 / dhuAl hisJ56l7 hi sJ5620 b&F645 The double mutation causing temperature sensitivity was then placed very close to dhdi (907, cotransducible) by the following cross. Phage grown on T,41789 was used to transduce the double mutant purF145 hisF645 to growth at 23" on n-histidine (0.3 mM), in the presence of adenine and thiamine (0.2 mM and 0.02 mM, respectively; adenine plus thiamine are required by strains containing the purF145 mutation); thus, the selection was only for dhuAl, and not for the temperature-sensitive mutation or for I'ur+. The recombinants growing on these plates were then tested for n-histidine growth at 23" and at 40": recombinants growing at both temperatures inherited only the dhdi mutation, those growing only at 23" inherited also the temperature-sensitive mutation. Ninety per cent (72 out of 80) of the u-histidine-growing recombinants were temperature sensitive; the remaining 10% grew well at either temperature, confirming the presence of dhuAl in this strain. Thus mutation hisJ56W is 42% cotransducible with purFl45 and 90$& cotransducible with dhuAl: these results firmly establish its position in the histidine permease cluster.2 2 It has been recently found (Govons and Ferro-Luxzi Ames, unpublished results) that the linkage between purF145 and the permease cluster is 2% when purF145 is the unselected marker in the transduction experiments, as opposed to a linkage of about 40'5,when Pur' is the selected marker. These data could be  which has a mutation in hisJ, is unable to grow on n-histidine at either temperature, as expected from its lack of J protein.
The significance of Fig. 2 is in the growth pattern of the temperature-sensitive strain, TM789 (dhuA1 hisJ5617 hisJ56dO hisFGQ5) ; this strain is able to grow at 17" (although not as well as the grandparent TA271), but it is completely unable to grow on n-histidine at 40". '~A1791 shows the normal activity levels expected for a dhuAl strain when grown at 30", but it has greatly decreased activity (28% residual activity) when grown at 37".

Chromatographic
Properties-The chromatographic properties of J protein from Th1791 (Jts protein) differ markedly from those of wild type J protein isolated from either LT.2 or dhu$i as illustrated in Fig. 3.4 A comparison of the elution profiles on DEAE-cellulose shows that the JtS protein from TA1791 elut,es at a higher salt molarity than the wild type J protein.
The elution pattern of this column is highly reproducible for different shock fluid preparations and for repacked columns.
The position of the wild type protein was identical in four experiments, while the JtS protein from TA1791 was eluted at a higher salt molarity in two independent preparations. The positions of acid phenyl phosphatase activity (which is partially resolvable into two peaks) and of histidine binding activity Peak III, as well as the conductivity measurements, are highly reproducible markers.
The reason for such a large difference in mobility between two proteins with the same molecular weight and isoelectric point is under investigation.
The Jt" protein presumably differs from the wild type protein in a sequence of several amino acids, because TA1791 was obtained by reversion of a frameshift mutant with ICR-191. Hydroxylapatite chromatography of wild type and TA1791 J proteins using gradient elution from 1 to 100 mM sodium phosphate (pH 7.0) also revealed a difference in mobility.
Wild type J protein was eluted at 40 mM phosphate, whereas JtS protein was eluted at 10 rnnl phosphate, using the same batch of adsorbant.
Both proteins are eluted similarly on Sephadex G-100 indicating a similar molecular weight. Dissociation Constant for H&id&-- Fig.  4 shows that the Jts protein from TM791 has a higher dissociation constant for * J protein isolated from dhuAf has been shown to be identical with that isolated from the wild type (4). Therefore the name of wild type protein refers to J protein from either dhuA1 or wild type.
These values were determined with the filtration assay at room temperature.
A difference in affinity for histidine is also observed using the dialysis assay, which is performed at 4": under these conditions the J protein from TA1791 has a KD = 2.5 PM and the wild type J protein has a KI, = 0.14 /.4M (4).
Temperature Xtabilify-The greater heat lability of the Jts protein (from TA1791) as compared with the wild type protein is shown in Fig. 5 We have had difficulty obtaining pure Jts protein because of this instability.
The J protein from either strain is stable to quick-freezing in ethanol-Dry Ice and storage at -10".
Disc Gel Etectrophoresis-Disc gel electrophoresis differentiates wild type from mutant J protein.
The pure wild type J protein gives a single band in the low p1-T system (4). h preparation of Jta prot,ein that was 80:/c pure shows only one major band which moves faster than the wild type protein.
A mixture of the two proteins gives two bands corresponding in position to each of the single bands.
Isoelecfric pH-J proteins from dhuA1 and from TM791 have the same isoelectric pEI of 5.5.
Immztnology-Shock fluid from TAl791, grown either at 30" or at 37", cross-reacts with antiserum prepared against the wild type J protein.
Shock fluid from several hisJ mutants lacking the binding protein contained no cross-reacting material. This demonstrates that the protein produced by TM791 is not an altogether different protein.
Correlation between Specificity of i n Viva Hi&dine Transport and Isolated J Protein The J protein has been definitively demonstrated to be a component of high affinity histidine transport by the use of mutants.
Consistent with this the binding properties of the J protein in vitro correlate well with the properties of the transport system in viva with respect to I)-and L-histidine, arginine, lysine, citrulline, ornithine, HIPA, imidazole pyruvic acid, and azaserine.
In Viva SpeciJicity Table III shows that L-arginine inhibits L-["HJhistidine uptake. Essentially complete inhibition is obtained at the highest arginine concentration used, which suggests that not only the J-P permease, but also the K-P permease is inhibited by arginine. Competitors of the transport system can also be identified by the inhibition of growth on n-histidine. n-Histidine is itself a much poorer substrate of the transport system than n-histidine (9). hrginine, citrulline, lysine, and ornithine completely inhibit growth of dhuA1 hisF645 on n-histidine (Fig. 6). Inhibition of growth by these compounds only occurs when n-hi&dine is the substrate because the 3 protein is an obligatory step in n-histidine 5 The discrepancy between these residual activities and those published in Reference 3 could be due to the higher purity of the proteins used here.  Pure J protein from either wild type or dhuAf was lrsed. Binding was assayed by filtration (4). All values are averages of duplicate assays.
The r,-[zH]histidine concentration was 1 FM when the additions were in lo-and 1,000.fold excess; 0.1 PM when the additions were in lO,OOO-fold excess.
All compounds tested, except n-histidine and D-HIPA, contained less than O.OlOI, contaminating histidine, either after repurification or as available commercially. L-Histidine contamination in the unlabeled n-histidine was shown to be less than 0.1% by a radioisotope dilution assay of L- they also contain a dhuA mutation (9).
&is is the case for rj-histidine, growth on irnidazole pyruvic acid is also cx)mpletely dependeut upon transport through the J protein, because introductioll of a hisJ mutation eliminates simultaneously the ability to grow on imidazole pyruvic acid and on u-histidine. We have previously demonstrated (3) that the histidine analogue HIPh is transported by the J-l' permease. .2zaserine is also transported through the elevated high affinity J-l' permease of the dhuA1 mutant.
This inhibitory analogue 1~1s bee11 demonstrated to be :I subst,rate for the aromatic permease (WOE') (1, 2). The inhibition of growth produced by azaserine on the wild type is completely reversed by any of the aromatic amino acids. However, introduction of the dhuA1 mutation causes the bacteria to acquire sensitivity to azaserine, even in the presence of an aromatic amino acid or of an aroY was grown overnight and diluted as described in the legend to Fig. 2. L-Arginine, or r,-lysine, or Lcitrulline, or uL-ornithine x-ere added at zero time where indicated. Growth was at 37".
mutation. This has been interpreted to mean that azaserine is transported through the J-1' histidine permease in sufficient amount to inhibit growth, when this system is elevated. In agreement with this interpretation, it was found that introduction of a hisP mutation in the dhuAl strain causes complete resistance to this new azaserine sensitivity, but only when an aromatic amino acid is present.
In the absence of an aromatic amino acid, transport of azaserine still occurs through the aromatic permease in :L dhu;l hisP double mutant and causes inhibition. These data allowed us to predict that the J protein would have an affinity for azuserine, despite its lack of obvious resemblance to histidine.

Xpecijkity 0s J Protein
The affinity of t,he Ilure J protein for a wide variety of amino acids and histidine nnalogues was measured by competition of the unlabeled compound with L-[3H]histidine binding ( Table  IV).
The following compounds have a significant affinity for the J protein: L-histidine, @nine, lysiue, T)-~~Il'& azaseriue, v-hietidine, citrulline, and ornithilre. X11 these compounds are known to be substrates of the high affinity histidiue permense. 50 other natural amino acid inhibited histidine binding up to lO,OOO-fold excess; cysteine consistently gave a slight stimulation of hixtidine binding; this might indicate a sulfhydryl activation of the J protein.
The relatively high affinity of the J protein for L-arginille is in agreement with the effect of L-arginine upon transport of L-an12 u-histidine (Table III and  This explains the strong inhibition of n-histidine growth (Fig. 6) by citrulline and ornitbine, which do not compete with L-histidine binding a,t I OOO-fold excess. The K. of the D-histidine-J protein complex, measured by competition of L-f3H]histidine binding, is 500 PM. The maximal number of L-hi&dine binding sites on the J protein is the same ill the presence and absence of D-h&dine. DISCUSSION .i necessnrg pereqnisito for rt biochemical approach to a study of active transport is the isolation of a presumed transport component and its unequivocal identification as an obligatory component of the transport system under study. Numerous laboratories have recelntly isolated a variety of such presumed components (reviewed in Reference 13j. .%mong these, a class of proteins called the "binding proteins," which bind a variety of small molecules and are thought to be located on the surface of gram-negative bacteria (14), has been implicated indirectly in the active transport of these molecules. The lack of direct evidence concerning t,he role of binding proteins in transport has been discussed (13).
We feel that the J protein has been shown conclusively to be a component, of histidine transport. m7e previously demonstrated (3) ii direct' correlation between the act,ivity of the histidine perrnease and the levels of the histidine-binding protein J, by showing that (a) mutation in the &~A site causes the simultaneous elevation of' the .7 protein aud of the rate of t-histidine transport; (b) mutation in the hisJ gene causes loss of the J protein and a decrease in the rate of L-histidine transport.
The ident'ity of the J protein as the J component of histidine t,ransport is now firmly established by the finding that both are coded for by a siugle structural gene. The properties of a temperature-sensitive strain, Th1791, provide this evidence. 'I'M791 was obtained as a, revertrant of a strain containing a hisJ mutation (and therefore defective in histidine transport). However, TM791 is not a t)rue revertant because it has not recovered the properties of a strain with an intact hisJ gene; the reversion is due to a second site mutation in the hisJ gene itself, as shown by the finding of an altered J protein and by the genetic analysis.
The J protein isolated from TM791 differs from the wild type J protein in temperature stability, affinity for histidine, chromatographic properties, and disc gel electrophoretic behavior.
The production of altered J protein has been correlated with the in viuo temperature sensitivity of T-%1791 ; both the L-histidine transport and the ability to grow on D-hi&dine are temperature sensitive. Both these activities require funct.ional J protein.
The possibility that the mutant protein is a new, completely different protein is excluded. The two proteins have several properties in common, although we chose to stress the differences. The temperature-sensitive mutation maps in the same place as mutatjions causing loss of the J prot,ein; both prot,eins bind histidine :~nd function in the transport of L-histidine and D-histidine; they have exactly the same molecular weight and isoelectric pH; they run in the same position in some gel electrophoresis systems; they are both dependent on hisP gene product for function; the mutant protein cross-reacts with antiserum to the wild type J prot,ein.
Many other revertant strains have been similarly obtained from a variety of hisJ-cont)aining st#rains. Preliminary experiments indicate that many of these also produce J proteins with altered properties.
It should be pointed out that the temperature sensitivity of the J protein in TA1791, as assayed in vitro by histidine binding, may not necessarily account for the in viva temperature sensitivity of growth on D-histidine.
If there is an interaction between the J protein and the 1' protein (or any other transport protein), the temperature-sensitive mutation could have affected such in L&O interaction, rather than the binding of the substrate to the J protein.
The evidence obtained from the genetic analysis that the hisJ gene is the structural gene for the J protein agrees with the excellent correlation between the properties of the wild type transport in vivo and of the wild type J protein in oitro.
IV'e presented here iuformation concerning the correlation in sperificity for substrates and analogues.
The highest affinity of botb in ~$0 transport and of the 3 protein is for L-histidine.
None of the other amino acids, added at lO,OOO-fold excess, competed with histidine binding to the J protein except for those shown in Table IV. In fact, only those compounds which have been shown t,o have a physiological effect on transport inhibit histidine binding, thus demonstratillg the correlation between the specificity of the J protein and of transport.
'I'he J protein has a relatively high affinity for ;Irginine and lysine. This is consistent with the known effects of t,hese compounds on transport in uivo; both arginine and lysine are good competitors of L-histidine uptake and D-hi&dine growth. Even though the affinity of the J protein for arginine is very good, it should be emphasized t,hat it still is considerably lower than its affinity for histidine (K. for arginine = 10 pM; KD for histidine = 0.1 PM).
The affinity of the J protein for D-histidine, citrulline, and ornithine, and for the analogues, D-IIIPR and azaserine, also correlates well with the affinity that these compounds have for the J-l' permease in vivo. As a consequence of this affinity, D-histidine, D-HIPB and azaserine can enter the cell. Therefore, D-histidine can act as growth substrate, while PHIPA and azaserine act as inhibitors.
The affinity of citrulline and ornithine for the J-P permease is shown by their inhibition of growth on D-histidine.
Recently (15) a periplasmic protein which binds hist,idine with much poorer affinity than the J protein has been isolated :mtl purified from S. typhirrrurium. This protein differs in se~~~~:tl biochemical properties from the J protein (discussed in Refrrence 4), and in all likelihood it is a different protein altogether. Moreover, no genetic characterization of this protein has been present,ed, thus rendering it difficult to draw any conclusion concerning the relationship between this proteiu and either the J protein or histidine transport.
Acknowledgments-We would like to thank Bruce N. Ames and Sydney Govons Ktistii for their advice and for innumerable stimulating discussions; I3etty I. Kirk for painfully analyzing the purity of the compounds in Table IV; C. E. Rallou for his financial support through United States Public Health Service Grant xm2121.