Peptides reproducing the phosphoacceptor sites of pp60c-src as substrates for TPK-IIB, a splenic tyrosine kinase devoid of autophosphorylation activity.

TPK-IIB, a spleen tyrosine protein kinase devoid of autophosphorylation activity (Brunati, A. M., and Pinna, L. A. (1988) Eur. J. Biochem. 172, 451-457), has been purified to near homogeneity and assayed for its ability to phosphorylate the synthetic peptides EDNEYTA and EPQYQPA reproducing the two conserved phosphoacceptor sites of pp60c-src (Tyr-416 and Tyr-527). While EPQYQPA was phosphorylated with low efficiency (Km = 16.7 mM, Kcat = 14.4), EDNEYTA is an excellent substrate displaying a Km value of 58 microM and a Kcat value of 31.2. The single substitution, in the latter peptide, of the glutamic acid adjacent to the tyrosine by alanine to give EDNAYTA caused a 6-fold increase in the Km. The positive influence on the phosphorylation of the acidic residues at -3 and -4 relative to the tyrosine is indicated by comparison of the kinetic constants for peptides EDAAYAA (Kcat = 4.6, Km 0.325 mM) and QNAAYAA (Kcat 2.4, Km 1.7 mM). Furthermore, when residues in the peptide NEYTA were replaced by alanine, the phosphorylation of the peptides NAYTA and AAYAA, was almost negligible (in terms of Kcat/Km ratio). However, AEYTA, NEYAA and AEYAA were still phosphorylated, albeit less efficiently than NEYTA. The probability that these peptides will adopt a beta-turn is EDNAYTA = EDNEYTA, NAYTA greater than NEYTA, and no predicted beta-turn for AEYTA, NEYAA, and AEYAA. Therefore these results support the concept that an amino-terminal acidic residue(s) is strictly required by TPK-IIB, irrespective of peptide conformation, although a beta-turn may enhance the phosphorylation of those peptides that satisfy this requirement. Two other spleen tyrosine kinases, TPK-I/lyn and TPK-III, both related to the src family, also have a far greater preference for the peptide EDNEYTA over EPQYQPA. However, they can be distinguished from TPK-IIB by their lower affinity for the peptides EDNEYTA and NEYTA and by their different specificity towards the substituted derivatives of NEYTA. TPK-I/lyn, accepts most of the substitutions that are detrimental to TPK-IIB, the triply substituted peptide AAYAA being actually preferred over the parent peptide NEYTA. The substitution of glutamic acid by alanine is also tolerated by TPK-III, although, in contrast to TPK-IIB, the phosphorylation efficiency is drastically decreased by the substitution of the asparagine at position -2.(ABSTRACT TRUNCATED AT 250 WORDS)

mM) and QNAAYAA (Kc,, 2.4, K , 1.7 mM). Furthermore, when residues in the peptide NEYTA were replaced by alanine, the phosphorylation of the peptides NAYTA and &Y&, was almost negligible (in terms of Kc,JK, ratio). However AEYTA, NEYAA and AE-YAA were still phosphorylated, albeit less efficieztly than NEYTA. The probability that these peptides will adopt a ,&turn is EDNAYTA = EDNEYTA, NAYTA > NEYTA, and no predictedb-turn for AEYTA, NEYAA, and AEYAA. Therefore these results support the concept that an amino-terminal acidic residue(s) is strictly required by TPK-IIB, irrespective of peptide conformation, although a b-turn may enhance the phosphorylation of those peptides that satisfy this requirement.
Two other spleen tyrosine kinases, TPK-I/Zyn and TPK-111, both related to the src family, also have a far greater preference for the peptide EDNEYTA over EPQYQPA. However, they can be distinguished from TPK-IIB by their lower affinity for the peptides ED-NEYTA and NEYTA and by their different specificity towards the substituted derivatives of NEYTA. TPK-I/Zyn, accepts most of the substitutions that are detrimental to TPK-IIB, the triply substituted peptide YAA being actually preferred over the parent peptide NEYTA. The substitution of glutamic acid by alanine is also tolerated by TPK-111, although, in contrast to TPK-IIB, the phosphorylation efficiency is drastically decreased by the substitution of the asparagine at po-* This work was supported by Italian MURST and CNR (Target project on Biotechnology and Bioinstrumentation). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. sition -2. Taken together these data would indicate that the highly conserved phosphoacceptor site, homologous to pp60"-"'" Tyr-416, is optimally configured for the specificity requirements of TPK-IIB, suggesting that TPK-IIB or tyrosine protein kinases with similar specificity might be involved in the phosphorylation of the members of the src family.
All the tyrosine protein kinases encoded by cellular genes of the src family contain two major phosphoacceptor sites which are homologous to Tyr-416 and Tyr-527 of pp60""" (reviewed in (1)). The former represents the main autophosphorylation site in vitro, and its phosphorylation correlates with increased kinase activity (2). The latter, which is close to the carboxyl terminus and is absent in the oncogenic forms (v-src), appears to be responsible for the down-regulation of the kinase itself (1, 3). Recently a brain TPK' uncapable of autophosphorylation has been reported to specifically phosphorylate in vitro the carboxyl-terminal phosphoacceptor site of pp60""" (4). Such an unusual lack of autophosphorylation activity is also shared by TPK-IIB (5), a spleen enzyme which can be resolved by chromatographic procedures from three other forms of tyrosine kinase two of which (TPK-I and TPK-IIA) are closely related to each other. While TPK-I and -1IA are immunologically indistinguishable from the product of the lyn oncogene (6), which is a member of the src family (7), TPK-IIB does not cross-react with any of the monospecific antibodies against TPKs of the src family tested so far. Such behavior and its inability to undergo autophosphorylation argue against any close relationship of TPK-IIB to the src family, whose members invariably include a highly conserved autophosphorylation site (see Ref. 1). These findings prompted us to undertake a study aimed at assessing whether TPK-IIB, like the brain "c-src-kinase" (4), might be involved in the phosphorylation and regulation of the TPKs belonging to the src family. Here we present data concerning the phosphorylation of peptides reproducing the two phosphoacceptor sites of the src proteins, and supporting the concept that Tyr-416 but not Tyr-527 is an excellent target for TPK-IIB. were synthesized by solid-phase technique from Fmoc amino acids using a manual synthesizer (Model Biolinx 4175, LKB). The sidechain functional groups of glutamic acid, aspartic acid, threonine, and tyrosine were blocked using acid-labile t-butyl groups. Syntheses were performed in continuous flow on a 0.1-nmol scale using the Fmoc amino acid active esters derived from pentafluorophenol or 3,4dihydro-3-hydroxy-4-oxo-1,2,3-benzotriazine. The synthesis with pentafluorophenol esters was carried out in the presence of N-hydroxybenzotriazole as a catalyst. Synthetic peptides were cleaved from the resin and side chain protecting groups removed using 95% aqueous trifluoroacetic acid for 1 or 2 h. Reagents were evaporated under low vacuum. The residue was dissolved in water and lyophilized. Crude peptides were purified by HPLC on a Waters Delta-Pak CIR 300A column (7.8 X 300 mm) using a Perkin-Elmer 410 LC B10 HPLC apparatus. Elution was performed with linear gradient from 0.1% aqueous trifluoroacetic acid to 30% acetonitrile containing 0.08% trifluoroacetic acid in 25 min a t 3 ml/min with the eluent monitored at 215 nm. The purity of the peptides (95% or more) was checked by analytical HPLC on a reversed-phase Waters Delta-Pak CIS 300A 15-pm (3.9 mm X 30 cm) column and by amino acid analysis. Peptides AEYAA and AAYAA (8), prepared by the traditional method in solution, were kindly provided by Professor F. Marchiori, Department of Organic Chemistry, University of Padova. Monoclonal m AS 327 anti-src antibodies were kindly provided by Dr. J. S. Brugge (State University, New York). Anti-cst-1 antibodies recognizing pp60"""', p 5 P , and pp62'"'' (12) and anti-SEEP antibodies raised against the conserved 330-345 segment of c-src were kindly provided by Dr. R. Kypta (EMBL, Heidelberg, Federal Republic of Germany).
The tyrosine protein kinases conventionally termed TPK-I, TPK-IIB, and TPK-I11 were isolated and partially purified from the particulate fraction of beef spleen essentially by the three-step procedure previously applied to rat spleen (5) and routinally assayed using poly(Glu,Tyr)4:1 as substrate (5). TPK-I is immunologically indistinguishable from the tyrosine protein kinase expressed by the lyn oncogene (6) and will be thereafter termed here TPK-I/lyn. TPK-I11 is also related to the src family as it cross-reacts with anti-cst-1 and anti-SEEP antibodies. Its precise identity however is still undetermined.
TPK-IIB has been further purified by gel filtration through a 2-X 80-cm Sephacryl S200 column equilibrated and eluted with buffer A (5) including 0.5 M NaC1, followed by HPLC on Mono Q H R 5/5 column connected to a Perkin-Elmer 410 LC B10 HPLC apparatus. The column, equilibrated with 20 mM Tris-HC1 (pH 7.5), 10% glycerol, and 15 mM 2-mercaptoethanol was washed with 15 ml of the same buffer and eluted at a flow rate of 0.8 ml/min with a linear NaCl gradient (0-0.5 M). For analytical purposes the fraction with highest tyrosine kinase activity was resubmitted to Mono Q HPLC applying a discontinuous gradient (see Fig. 1). A single sharp peak of absorbance at 280 nm overlapped by tyrosine kinase activity was eluted at 0.18 M NaCI.
Phosphorylation of peptides was performed by incubation at 30 "C in 50 pl of a medium containing 50 mM Tris-HC1, pH 7.5, 10 mM MnC12, 10 p~ sodium vanadate, 20 PM [y-:"P]ATP (specific activity, 1000 cpm/pmol), and 2-10 units of tyrosine protein kinase; in the case of TPK-I/lyn 2 M NaCl was also added to the incubation medium as an activator (6). Unless otherwise indicated the peptide concentration was 2 mM and the incubation time 10 min. The reaction was stopped by adding acetic acid (final concentration, 30%), and 32P incorporated into the peptide was evaluated by combining ion-exchange (9) and isobutanol-benzene extraction (10) as detailed previously (11).
Autophosphorylation was performed by 10-min incubation in the same medium described above except for the absence of NaCl and any peptide substrate. The reaction was started by adding ["PIATP after 10-min preincubation and was stopped by boiling in 2% SDS. '"P incorporated was evaluated by 11% SDS-PAGE and autoradiography as described previously (5).

RESULTS
The spleen tyrosine protein kinase termed TPK-IIB was first characterized by its unique lack of autophosphorylation and its remarkable inhibition by heparin (5). Recently TPK-IIB has been shown not to cross-react with various monospe-cific antibodies raised against tyrosine kinases of the src family, namely those expressed by kk, hck, lyn, and fyn (6).
TPK-IIB also fails to cross-react with anti-c-src antibodies (m AS 327), with anti-cst-1 antibodies which recognize pp62c-yes, pp60"-"", and p 5 P , and with anti-SEEP antibodies raised against the 330-345 segment of pp60c-"", which is highly conserved in all the tyrosine kinases of the src family (not shown).
After the heparin sepharose chromatography to separate TPK-IIB from TPK-IIA (5), TPK-IIB underwent three additional purification steps as shown in Table I. On Mono Q HPLC the final preparation exhibits an individual sharp protein peak coinciding with tyrosine kinase activity (Fig. 1). This peak gives rise to a single prominent protein band of the expected M , (52,000) upon SDS-PAGE ( Fig. 1, inset). The identification of this band as the tyrosine kinase itself is also consistent with the high specific activity of the final preparation of TPK-IIB which is comparable with the activity exhibited by other highly purified preparation of tyrosine kinases (13, 14 and references cited therein). Such a high activity would hardly be compatible with the single prominent band merely representing a contaminating component.
As pointed out previously, TPK-IIB does not undergo any detectable autophosphorylation (5). Its lack of autophosphorylation sites has been now corroborated by its failure to crossreact with antiphosphotyrosine monoclonal antibodies ( Fig.  2): no phosphotyrosine signal could be detected with TPK-IIB, with or without preincubation in the autophosphorylation medium. On similar blots, lysates from jurkat cells expressing high levels of p56lCk, a tyrosine kinase of the src family (l), resulted in strong positive signal, of the expected molecular weight. The same signal is evident with spleen TPK-I/lyn, another src-related tyrosine kinase capable of autophosphorylation (5,6). The specificity of the reaction was established by preincubation of the antibody with phosphotyrosine which totally eliminated the signal.
This finding clearly confirms that TPK-IIB is devoid of phosphotyrosyl sites, a quite unusual feature, distinguishing it from all the known tyrosine kinases of the src-family. TPK-IIB nevertheless phosphorylates synthetic peptides reproducing the two main tyrosyl phosphoacceptor sites that are conserved in all the cellular members of the src family. As shown in Table I1 the peptide EDNEYTA, reproducing the sequence around pp60"-"" Tyr-416, is an especially good substrate, particularly by virtue of its K,,, value (58 p~) , which is one of the lowest ever reported for peptides of comparable size used as substrates for tyrosine kinases (e.g. see Ref. 15). The phosphorylation efficiency of EDNEYTA by TPK-IIB is more than 30 times higher than that of angiotensin 11, a widely employed substrate of tyrosine protein kinases, while the peptide EPQYQPA, reproducing the down-regulation site corresponding to pp60""" Tyr-527, is a much poorer substrate: its K,,, value is more than two orders of magnitude higher and its Kc,, is about two-fold lower than those of EDNEYTA (Table 11).
In order to obtain a better insight into the particular susceptibility of the peptide EDNEYTA to phosphorylation by TPK-IIB, a number of derivatives have been synthesized and analyzed for their kinetic parameters (Table 11). It is evident that, except for the conservative replacement of Glu for Asp at position -3 relative to tyrosine, all the other substitutions tested are unfavorable, giving rise to peptides that invariably display higher K,,, and sometimes also lower Kc,, values. The replacement of the glutamic acid adjacent to tyrosine is especially unfavorable. It is more detrimental however in the pentapeptide NAYTA than in the heptapep- Purification of TPK-IIR from bovine spleen The isolation of'the particulate fraction from 3.5 kg of spleen from freshly slaughtered beef and its extraction with lri Nonidet P-10 were performed essentially as previously described (5). The crude extract was resolved into four fractions (TPK-I, TPK-IIA, TPK-IIB, and TPK-111) by combining DEAE-Sepharose and heparin-Sepharose chromatographies, as in a previous study (5). TPK-IIB was further purified by phosphocellulose chromatography ( 5 ) and by Sephacryl S-200 gel filtration and Mono Q HPLC as described under "Materials and Methods." The activity in the crude extract and after the first purification step is underestimated due to the presence of tyrosine protein phosphatase activity not completely inhibited by the vanadate present in the kinase assay. This may also account for the higher than 100% apparent recovery of activity after the DEAE-Sepharose and heparin-Sepharose steps if the activities of' the other tyrosine kinase fractions, TPK-I/TPK-111 and TPK-IIA, resolved by DEAE-Sepharose, and heparin-Sepharose, respectively (5) , .  Table  1) was subjected to a second Mono Q HPLC by applying a discontinuous gradient (dotted line). Proteins were automatically recorded a t 280 nm (solid line). 0.4-ml fractions were collected and 10-pl aliquots were assayed for tyrosine kinase activity tide EDNAYTA, suggesting that the additional acidic residues a t position -3 and -4, present only in the heptapeptide, may exert a positive influence, reinforcing the effect of glutamic acid at position -1. Consistent with this hypothesis ED-NEYTA, EDNAYTA and EDAAYAA are much better substrates than NEYTA, NAYTA, and AAYAA, respectively, and EDAAYAA is phosphorylated more efficiently than QNAAYAA. The striking superiority of EDNEYTA over its triply substituted EDAAYAA derivative, however, highlights the crucial relevance of the amino acids nearer to the tyrosyl residue. The comparative analysis of the individual monosubstituted derivatives of NEYTA, namely, AEYTA, NAYTA, and NEYAA clearly indicates the most harmful substitution to be that of the glutamic acid adjacent to the amino-terminal side of tyrosine, NAYTA exhibiting a 10-fold higher K , and a more than 8-fold lower Kc,, than the parent peptide NEYTA. It should be noted that the propensity to adopt a &turn conformation is similar for NAYTA and NEYTA and for EDNAYTA and EDNEYTA, despite their sharply different susceptibility to phosphorylation. Apparently therefore an acidic side chain adjacent to the amino-terminal side of tyrosine acts as a powerful specificity determinant independent of its ability to confer a &turn conformation.
On the other hand the unfavorable effect of substituting the neutral residues at positions -2 and/or +1 with alanine could be interpreted in terms of conformational alterations, since the singly and doubly substituted derivatives AEYTA, N E Y W a n d AEYAA have lost the predicted P-turn conformation of the parent pentapeptide NEYTA (see Table 11). The same effect could account for the decreased efficiency of EDAAYAA compared with EDNAYTA.
The excellent phosphorylation efficiency of EDNEYTA by TPK-IIB, by virtue of its low K,, is especially remarkable if  Chou and Fasman (20) for all the quartets of amino acids including tyrosine. Only the highest value for each peptide is reported and the corresponding predicted &turn is shown if the p, value is higher than the cut-off value of p , = 0.75.10" (20).
Peptides Kc., Apparent K , K..,/K,,, p  Kinetic constants of src-derived peptides for partially purified tyrosine kinases TPK-I/lyn and TPK-111 Vmnx (expressed as nmol. min-lmg-l) and apparent K,,, values were determined as detailed in Table I1 for TPK-IIB. In the case of TPK-I/lyn and TPK-111, however, Kc., values could not be calculated since the enzyme preparations were only partially purified. compared with tyrosine kinases of the src family, whose K,,, values for src-derived peptides very similar to EDNEYTA have been reported to lay in the millimolar range (reviewed in Ref. 15). Also consistent with this observation is the fact that the src-related spleen tyrosine kinase termed TPK-I/lyn exhibits a 3 . 8 -m~ K,,, for EDNEYTA (Table HI), a value which compares quite well with the values of 6.25 and 5.0 mM reported for the phosphorylation of the peptides EDNEY-TARQG and IEDNEYTARQ by pp60'."" (16) and ~5 6 "~ (17), respectively. Moreover, those substitutions that decrease the phosphorylation of the parent pentapeptide by TPK-IIB are favorable to TPK-I/lyn whose phosphorylation efficiency is actually higher than the triply substituted derivative AAYAA than it is with NEYTA (Table 111).

Vmmx
TPK-111, another spleen tyrosine kinase ( 5 ) immunologically related with the src family for being recognized by anticst-1 and anti-SEEP antibodies* displays a peptide substrate A. M. Brunati, R. Kypta, A. Donella-Deana, and L. A. Pinna, unpublished data. 2 of src Peptides 17801 specificity distinct from those of either TPK-IIB and TPK-I/lyn. It is reminiscent of TPK-IIB in that the parentage of pentapeptide is preferred over the substituted derivatives (Table 111). The replacement of the asparagine a t position -2 however is much more detrimental than that of the glutamic acid adjacent to the tyrosyl residue (Table 111).
Any more direct and exhaustive comparison of the catalytic efficiencies of the three spleen tyrosine kinases considered here was hampered by the fact that TPK-I/lyn and TPK-I11 are still only partly purified, so that their Kcat values could not be calculated nor compared with those of TPK-IIB.
The possibility that the lower K,,, values of TPK-IIR might merely reflect its higher degree of purification was ruled out by the fact that identical values were obtained when the kinetic experiments were performed with TPK-IIB after heparin sepharose, the degree of purification of which is comparable to that of TPK-I/lyn and TPK-I11 (not shown).
In order to assess the actual ability of TPK-IIB to phosphorylate tyrosine kinases of the src family, TPK-I/lyn, whose autophosphorylation site is exactly reproduced by the eptapeptide EDNEYTA, was incubated with TPK-IIB in the absence and presence of heparin, which is a powerful inhibitor of TPK-IIB (5). while it stimulates TPK-I/lyn activity (5.6). As shown in Fig. 3A TPK-IIB promotes an increased radiolabeling of TPK-I/lyn which is suppressed by heparin. The "P-peptide maps obtained from TPK-I/lyn either autophosphorylated or phosphorylated in the presence of TPK-IIR are identical (Fig. 3 B ) suggesting that the same phosphoacceptor site(s) are involved. This would indicate that the sequence EDNEYTA, is preferentially affected by TPK-IIR even if it is included into the parent proteins.

DISCUSSION
A somewhat paradoxical outcome of this work is that peptides reproducing the highly conserved autophosphorylation site shared by all TPKs of the src family (Tyr-416 of pp60c"'r) are more efficiently phosphorylated by a spleen T P K presumably unrelated to the src family (TPK-IIB) than they are by members of the src family such as pp60""" (161, p56""  2 kI)a). (18), and by two src-related spleen TPKs, namely TPK-I/lyn and TPK-111. These peptides are excellent substrates for TPK-IIB, displaying K , values in the micromolar range. The fact that substitution of any of the amino acids surrounding tyrosine decreases their ability to serve as substrates for TPK-IIB suggests that all the features of the main phosphoacceptor site of the src family contribute to the specificity for this kinase. The membership of this tyrosine kinase to the src family, though not completely ruled out, is rendered extremely unlikely by its lack of autophosphorylation activity, a property shared by all the src-related TPKs known so far and by its markedly different peptide substrate specificity. Furthermore, the modifications that decrease the phosphorylation of synthetic peptides by TPK-IIB either do not affect or even improve the phosphorylation of the same peptides by TPK-Illyn. Such a behavior, albeit somewhat paradoxical, was not totally unexpected considering that amino-terminal acidic residues (which are invariably found at the src autophosphorylation sites) are not required for the phosphorylation of synthetic peptides by TPKs related to pp60""" (16,19), and may even exert an unfavorable effect (8).
In this connection it would be interesting to establish the identity of TPK-111, which is also a member of the src family according to its reactivity with anti-cst-1 and anti-SEEP antibodies raised against conserved segments of src protein kinases. Its substrate specificity however is significantly distinct from that of TPK-I/lyn. The finding that TPK-I11 is not recognized by a variety of monospecific antibodies raised against the products of c-src, lck, hck, lyn, fyn ( 6 ) , and yes' oncogenes, increases the probability that it might be identical or very closely related to the last member of the src family, namely fgr. Interestingly the fgr tyrosine kinase differentiates for having the least conserved autophosphorylation site among the members of the src family (see Ref. 1). This might correlate with the distinct site specificity of TPK-111.
It is possible that the great susceptibility of EDNEYTA and NEYTA to phosphorylation by TPK-IIB might partially derive from their adopting a @-turn conformation, which is predicted by the method of Chou and Fasman (20). However, the probability of adopting a @-turn conformation is either the same or even higher for EDNAYTA and NAYTA, which are much worse substrates than EDNEYTA and NEYTA, respectively. It is reasonable therefore to conclude that the favorable effect of glutamic acid at position -1 is accounted for predominantly by the acidic nature of its side chain, rather than by any effect of conformation. The intrinsic importance of the acidic nature of the residue adjacent to the aminoterminal side of tyrosine is also corroborated by the very poor susceptibility to phosphorylation by TPK-IIB of EPQYQPA, despite its high probability of adopting a @-turn conformation (see Table 11). The additional finding that the triply substituted heptapeptide EDMYAA, although a much poorer substrate than the parent peptide, is nevertheless phosphorylated more efficiently than its neutral derivative QNAAYAA, discloses the favorable role of the acidic residue(s) a t position -3 and/or -4 as well.
Nevertheless, it is possible that a P-turn conformation might improve the suitability of phosphoacceptor sites that already fulfill the minimum structural requirements. This would account for the decreased phosphorylation efficiency of the peptides AEYTA, NEYAA, and AEYAA which have lost the predicted /?-turn conformation of NEYTA.
Taken together these observations would suggest that TPK-IIB or other TPK(s) with similar site specificity are involved in the phosphorylation of the src products in uiuo. It should be recalled in this connection that although pp60""" Tyr-416 (and the homologous tyrosines of the other src-TPKs) undergo in vitro autophosphorylation, no incontrovertible evidence is available that their in viuo phosphorylation invariably occurs by the same mechanism. Rather, the possibility is still open that heterologous phosphorylation of Tyr-416 by a distinct kinase could contribute to the activation of p p 6 0 (see Ref. 1). The inter-, rather than intramolecular mechanism of in vitro autophosphorylation (21) would be consistent with this hypothesis, assuming that a kinase with higher affinity for the phosphorylation site (like TPK-IIB) is present. On the other hand, it is also possible that the autophosphorylation efficiency of the src-related tyrosine kinase is increased either by conformational features inherent in the tertiary structure of the kinase itself or by so far unidentified endogenous effectors. In such a case Tyr-416 phosphorylation/activation by an heterologous kinase may still become relevant whenever the cellular src-TPKs have an intrinsically low activity, because of down-regulation by concomitant Tyr-527 phosphorylation. In any event the hypothesis that the autophosphorylation site(s) of src tyrosine kinases might be targeted by TPK-IIB has been corroborated, at least in vitro, by showing that TPK-IIB promotes an increased phosphorylation of TPK-I/lyn, giving rise to the same "P-peptide map obtained with autophosphorylated TPK-I/lyn.
In this respect TPK-IIB seems to be different from the brain T P K reported to specifically phosphorylate pp60""" a t its carboxyl-terminal site (Tyr-527) (4) despite sharing the same inability to perform autophosphorylation, a most unusual property among tyrosine kinases. TPK-IIB actually displays a very low phosphorylation efficiency toward the synthetic peptide EPQYQPA reproducing the sequence around Tyr-527. It should be noted however that such a peptide is an extremely poor substrate for TPK-I/lyn too (Table 11), consistent with the concept that the carboxylterminal tyrosine is not an autophosphorylation site but, rather, a target for another TPK(s) involved in down-regulation of src-TPKs.
Although TPK-IIB poorly phosphorylates the peptide EPQYQPA, it is still conceivable, that its efficiency might increase once this sequence is included in a more extended protein domain. Considering, however, the opposite regulatory functions of Tyr-416 and Tyr-527 it seems unlikely that they might be targets for the same kinase. All the data obtained with synthetic peptides support the concept that TPK-IIB displays a remarkable affinity for the site including Tyr-416 but not for the one including Tyr-527. A similar preference for EDNEYTA over EPQYQPA is shared by the other spleen tyrosine protein kinases characterized SO far. In this respect the synthetic peptide EPQYQPA could prove a suitable substrate for monitoring and detecting the tyrosine protein kinase(s) which are able to down-regulate the TPKs of the src family by phosphorylating their carboxyl-terminal site.