Phosphorylation of Plant Eukaryotic Initiation Factor-2 by the Plant-encoded Double-stranded RNA-dependent Protein Kinase, pPKR, and Inhibition of Protein Synthesis in Vitro *

Regulation of protein synthesis by eukaryotic initia- tion factor-2 (cid:97) (eIF-2 (cid:97) ) phosphorylation is a highly conserved phenomenon in eukaryotes that occurs in re- sponse to various stress conditions. Protein kinases capable of phosphorylating eIF-2 (cid:97) have been characterized from mammals and yeast. However, the phenome- non of eIF2- (cid:97) -mediated regulation of protein synthesis and the presence of an eIF-2 (cid:97) kinase has not been dem-onstrated in higher plants. We show that plant eIF-2 (cid:97) (peIF-2 (cid:97) ) and mammalian eIF-2 (cid:97) (meIF-2 (cid:97) ) are phosphorylated similarly by both the double-stranded RNA- binding kinase, pPKR, present in plant ribosome salt wash fractions and the meIF-2 (cid:97) kinase, PKR. By several criteria, phosphorylation of peIF-2 (cid:97) is directly correlated with pPKR protein and autophosphorylation lev- els. Significantly, pPKR is capable of specifically phosphorylating Ser 51 in a synthetic eIF-2 (cid:97) peptide, a key characteristic of the eIF-2 (cid:97) kinase family. Taken to-gether, these data support the concept that pPKR is a member of the eIF-2 (cid:97) kinase family. In addition,

Regulation of protein synthesis by eukaryotic initiation factor-2␣ (eIF-2␣) phosphorylation is a highly conserved phenomenon in eukaryotes that occurs in response to various stress conditions. Protein kinases capable of phosphorylating eIF-2␣ have been characterized from mammals and yeast. However, the phenomenon of eIF2-␣-mediated regulation of protein synthesis and the presence of an eIF-2␣ kinase has not been demonstrated in higher plants. We show that plant eIF-2␣ (peIF-2␣) and mammalian eIF-2␣ (meIF-2␣) are phosphorylated similarly by both the double-stranded RNAbinding kinase, pPKR, present in plant ribosome salt wash fractions and the meIF-2␣ kinase, PKR. By several criteria, phosphorylation of peIF-2␣ is directly correlated with pPKR protein and autophosphorylation levels. Significantly, pPKR is capable of specifically phosphorylating Ser 51 in a synthetic eIF-2␣ peptide, a key characteristic of the eIF-2␣ kinase family. Taken together, these data support the concept that pPKR is a member of the eIF-2␣ kinase family. In addition, the inhibition of brome mosaic virus RNA in vitro translation in wheat germ lysates by the addition of doublestranded RNA, phosphorylated peIF-2␣, meIF-2␣, or activated human PKR suggests that plant protein synthesis may be regulated via phosphorylation of eIF-2␣.
The initiation phase in eukaryotic protein synthesis is characterized by complex interactions between numerous initiation factors, ribosomal subunits, nucleotides, and Met-tRNA i . Framework events in initiation-reinitiation involve ternary complex (eIF-2⅐GTP⅐tRNA i Met ) binding to free 40 S ribosomal subunits to form a 48 S preinitiation complex. The subsequent binding of mRNA and 60 S ribosomal subunits is dependent upon GTP hydrolysis, yielding a viable initiation complex and releasing eIF-2⅐GDP 1 (1,2). In order for another round of initiation to begin, GDP must be exchanged with GTP. This guanine nucleotide exchange is catalyzed by eIF-2B. It is generally believed that phosphorylation of the ␣ subunit of eIF-2 on Ser 51 stabilizes the eIF-2⅐GDP⅐eIF-2B complex, effectively sequestering eIF-2B. This elegant regulatory mechanism is induced by various stresses including virus infection (3), heat shock (4), and deprivation of amino acids (5) or heme (6).
The eIF-2␣ kinases comprise a specific protein kinase subfamily (7). They include the heme-deficient kinase, HRI (6,8); the interferon-induced, double-stranded RNA (dsRNA)dependent kinase, mPKR (9, 10); and yeast GCN2 (11,12). Although their noncatalytic regions differ significantly, all contain inviolate amino acid domains associated with Ser/Thr kinase activity and eIF-2␣ phosphorylation (7). Collectively, the mammalian interferon-induced dsRNA-dependent protein kinases are termed PKR or mPKR. Low levels (1-10 g/ml) of dsRNA or appropriately structured single-stranded RNA (ss-RNA) bind to and activate mPKR via autophosphorylation, whereas higher levels have an inhibitory effect (7,13). Recently a plant-encoded analog of mPKR has been identified and characterized from monocot and dicot tissues (14,15). Plant PKR, pPKR, is an M r 68,000 -70,000 Ser/Thr kinase present in both cytosolic and ribosome-associated fractions. Although it is constitutively expressed in healthy cells, pPKR protein and phosphorylation levels are significantly stimulated by dsRNA, select polyanions, or virus/viroid infection (16,17). 2 The regulation of plant protein translation by eIF-2␣ phosphorylation has not been demonstrated. Indeed, there is considerable confusion in the literature regarding the phosphorylation of peIF-2, in large part due to discrepancies in subunit identification (18 -21). meIF-2 is composed of three subunits, the M r 36,000 ␣ subunit, the M r 38,000 ␤ subunit, and the M r 52,000 ␥ subunit (22). Plant eIF-2 (peIF-2) has been purified and is also composed of three subunits of M r 38,000, M r 42,000 (doublet), and M r 50,000 (18 -20, 23-26). Several studies have indicated that kinases that phosphorylate the ␣-subunit of meIF-2 also phosphorylate the M r 42,000 subunit of peIF-2, although often this subunit has been designated as eIF-2␤ (18,19,21,27). It has not been conclusively shown that phosphorylation of the M r 42,000 subunit affects translational activity in wheat germ systems. Previous failure to detect an effect of phosphorylation in vitro (28) or in vivo (29), the lower dissociation constant of eIF-2 for GDP in plants (20) compared with mammals, and the apparent absence of plant recycling factor activity (eIF-2B) have led to the general assumption that plants do not use phosphorylation of eIF-2␣ as a means to regulate protein translation (20). However, the wheat germ M r 42,000 subunit has now been conclusively identified as the ␣-subunit and the M r 38,000 as the ␤-subunit by molecular cloning and cDNA sequencing. 3 Further, Arabidopsis thaliana-expressed sequence tags (EST, GenBank TM accession numbers T45955 and T22066) that encode functional equivalents of meIF-2␣ contain the inviolate phosphorylation sites corresponding to Ser 51 . In addition, an A. thaliana EST (Gen-Bank TM accession number T44879) and a rice EST (Gen-Bank TM accession number D25052) appear to encode the functional equivalents of yeast GCN3 and GCD2, respectively, that are subunits of recycling factor eIF-2B. Thus, it appears highly likely that plants contain the necessary factors for an eIF-2␣ phosphorylation pathway.
Here we show that the plant-encoded pPKR is capable of phosphorylating the ␣ subunit of both plant and mammalian eIF-2. In addition, in vitro protein synthesis in wheat germ lysates is specifically inhibited by phosphorylated eIF-2, dsRNA, an inducer of pPKR activity, or activated mPKR.

EXPERIMENTAL PROCEDURES
Materials-Plant eIF-2 was purified according to Ref. 30. Purified human eIF-2 was supplied by W. Merrick (Case Western Reserve University, Cleveland, OH) and N. Gupta (University of Nebraska, Lincoln, NE). Poly(rI)⅐poly(rC)-agarose was made according to Langland et al. (31). Wheat germ in vitro translation systems were obtained from Promega (Madison, WI) and Ambion (Austin, TX). Anti-pPKR serum was prepared according to Langland et al. (15), antiserum to a conserved dsRNA binding domain (that recognizes pPKR) (15) was supplied by B. Jacobs (Arizona State University, Tempe, AZ), and the monoclonal antibody to mPKR was supplied by A. Hovanessian (Institut Pasteur, Paris, France (32). The peptide corresponding to residues 45-56 of meIF-2␣ was synthesized according to the method of Mellor and Proud (33). All other reagents and chemicals were purchased from Sigma unless otherwise indicated.
Barley (Hordeum vulgare L. "Steptoe") was grown in the dark for 7 days at 25°C. Leaves were homogenized in 150 mM PIPES, pH 7.5, 50 mM EDTA, 5 mM dithiothreitol, 100 mM KCl, 1 g/ml aprotinin, 1 g/ml leupeptin, and 2 mM phenylmethylsulfonyl fluoride. Raw wheat germ was purchased locally. Plant extracts were prepared by brief homogenization at 4°C in a Waring blendor followed by filtering through four layers of cheesecloth. Following centrifugation at 30,000 ϫ g for 30 min, the supernatant was used to prepare S-100 (cytosolic) and ribosomal salt wash fractions (15). Fractions were dialyzed overnight in 20 mM HEPES, pH 7.5, 100 mM NaCl, 5 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride. Extracts from HeLa cells were prepared as described previously (34). Ribosome salt wash-associated mammalian PKR was partially purified using previously described fast protein liquid chromatography Mono S and Mono Q chromatographic methods (34).
Partial Purification of pPKR-Poly(rI)⅐poly(rC)-agarose affinity purification was done according to a modification of Langland et al. (31). Tissue fractions were incubated for 5 min at 4°C with poly(rA) (50 g/ml) prior to the addition of prewashed poly(rI)⅐poly(rC)-agarose and incubation for 30 min on ice. Following centrifugation at 100 ϫ g for 2 min, the matrix was washed three times at 4°C with 100 mM KCl, 10% glycerol, 20 mM HEPES, pH 7.5, 5 mM MgOAc, 5 mM MnCl 2 , 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 mM dithiothreitol. Competition assays were performed by incubation of the extracts with 500 g of poly(rI)⅐poly(rC), poly(rA), or poly(rU) (10-fold excess to the amount of dsRNA present on the resin) prior to the addition of poly(rI)⅐poly(rC)-agarose.
Analysis of Phosphorylated Products-In assays involving mPKR, pPKR, and eIF-2␣ phosphorylation, proteins were separated by standard SDS-PAGE followed by autoradiography (15). A Tricine gel system (16.5% T, 6% C separating gel; 10% T, 3% C spacer gel; 4% T, 3% C stacking gel) according to Schagger and von Jagow (35) was used to assay for phosphorylation of the synthetic eIF-2␣ peptide. After electrophoresis, the gel was fixed in 10% trichloroacetic acid, and phos-phopeptides were visualized by autoradiography. V-8 protease-digested eIF-2␣ peptide products were analyzed by electrophoretic thin layer chromatography according to a modification of Mellor and Proud (33). Following in vitro phosphorylation, the reaction supernatant was removed and brought to 100 mM Tris, pH 6.8. V-8 protease was added at a concentration of 10 g/ml followed by incubation at 30°C for 8 h. The samples were dried and resuspended in 6% (v/v) formic acid, 1.25% (v/v) acetic acid, 0.25% (v/v) pyridine followed by electrophoretic thin layer chromatography separation (using Silica gel G) in the same buffer at 400 V for 2 h. The peptides were visualized by autoradiography.
Immunoclearing Analysis-For immunoprecipitation clearing experiments, maximal clearing of pPKR from extracts was achieved by incubating ribosome salt wash fractions (0.5 ml) overnight at 4°C with anti-PKR sera (containing a mixture of dsRNA-binding domain antiserum (1:10 dilution), monoclonal human PKR antiserum (1:100 dilution), and anti-pPKR serum (1:10 dilution). Separate fractions were similarly incubated with preimmune serum. Staphylococcus aureus cells were then added, and incubation continued for 1 h at 4°C. The bacterial cells were pelleted, and the supernatant was used in standard poly(rI)⅐poly(rC)-agarose binding assays.
In Vitro Translation-Protein synthesis assays were done at 25°C for 60 min in standard wheat germ lysate reaction mixtures according to the manufacturer's directions using BMV RNA. Aliquots were denatured in SDS-sample buffer and separated by SDS-PAGE. Where indicated, poly(rI)⅐poly(rC), poly(rA), or purified mPKR were added to the translation reaction prior to the addition of BMV RNA.
In vitro translation assays to evaluate the effect of phosphorylated eIF-2 on protein synthesis were modified as follows: mPKR or buffer (20 mM HEPES, pH 7.5, 10% glycerol, 100 mM KCl, 5 mM MnCl 2 , 5 mM MgOAc, 14 mM ␤-mercaptoethanol) was incubated with poly(rI)⅐ poly(rC)-agarose and thoroughly washed in 1 M KCl buffer followed by a 10 mM KCl wash (31) prior to the addition of phosphorylation mix alone or plant or mammalian eIF-2 (3 g) in phosphorylation mix. Phosphorylation mix contained 100 M unlabeled ATP, 5 mM MnCl 2 , 5 mM MgOAc, 20 mM HEPES, pH 7.5, and 10 mM KCl. After incubation for 20 min at 30°C, the resin containing the bound mPKR or the resin incubated with buffer was pelleted, and the supernatant was removed and brought to 10 mM EDTA. This solution was added to a wheat germ translation mix manufactured by Ambion, and BMV RNA translation assays were performed according to the manufacturer's directions.

RESULTS AND DISCUSSION
Specific kinases present in animals and yeast phosphorylate eIF-2␣, resulting in translational regulation (3). A corresponding eIF-2␣ phosphorylation pathway has not been identified in plants, although a plant analog of the mammalian eIF-2␣ kinase, PKR, has been characterized (14 -16). Fig. 1A demonstrates that a kinase present in the ribosomal salt wash fractions from barley leaves is capable of specific in vitro phosphorylation of the M r 36,000 ␣ subunit of meIF-2 and the M r 42,000 doublet of peIF-2, now termed peIF-2␣ (lanes B and C, respectively). Phosphorylation of peIF-2␣ by the plant kinase appears better than phosphorylation of meIF-2␣, given that equal protein amounts were used. Purified human PKR likewise phosphorylates meIF-2␣ and peIF-2␣ (Fig. 1B, lanes B  and C) (36). These results suggest that peIF-2␣ may contain a domain similar to the phosphorylation domain of meIF-2␣. Supporting this conclusion is the existence of an Arabidopsis EST (T45955) having 83% identity at the amino acid level with the phosphorylation domain of human eIF-2␣ corresponding to residues 45-56. Further this plant EST has 100% identity with a domain (residues 72-90) that is highly conserved between eIF-2␣ and K3L. K3L is a vaccinia virus-encoded protein that is a competitive inhibitor of PKR-mediated phosphorylation of eIF-2␣ (37). The labeled bands at M r 48,000 and 55,000 in lane A are degradation products of pPKR based upon peptide mapping (data not shown), and it is likely that proteolytic activity not controllable by the inhibitors used in extract preparation is also responsible for the apparent degradation in the eIF-2 preparations observed in Fig. 1A, lanes B and C. The fact that plant kinase fractions were incubated with dsRNA-agarose and thoroughly washed prior to in vitro phosphorylation assays suggests that the responsible kinase is a dsRNA binding protein. In order to establish the dsRNAdependent phosphorylation of peIF-2␣ and meIF-2␣, excess soluble dsRNA was added to plant extracts prior to binding to dsRNA-agarose. Phosphorylation of peIF-2␣ and meIF-2␣ was specifically decreased in the presence of competitor dsRNA (Fig. 2, lanes B and F, respectively). The only dsRNA-binding protein kinase yet identified from plants is pPKR (17). As determined by immunoblotting according to Ref. 15, dsRNA competition decreased pPKR protein levels binding to dsRNAagarose (data not shown). This decrease in pPKR protein corresponded to a decrease in phosphorylation of the M r 68,000 -70,000 pPKR (lanes B and F) and was directly correlated with eIF-2␣ phosphorylation levels. Preincubation with 500 g of soluble ssRNA as poly(rA) or poly(rU) had no effect on the phosphorylation levels of either pPKR or eIF-2␣ (data not shown). Similarly, dsRNA competition specifically decreased mPKR protein binding to dsRNA-agarose, resulting in decreased phosphorylation of mPKR, peIF-2␣ (lane D) and meIF-2␣ (lane H). These data are consistent with the demonstration that mPKR binding of dsRNA induces two phosphorylation activities, autophosphorylation and eIF-2␣ phosphorylation (7,13), and indicate a similar mechanism for pPKR. Several phosphorylated bands are noted in Fig. 2 that do not decrease with dsRNA competition, suggesting that a contaminating dsRNA-independent kinase may be present in the plant extract. Therefore, the role of pPKR in eIF-2␣ phosphorylation was further evaluated by immunoclearing experiments. Ribosome salt wash extracts from barley leaves were incubated with preimmune serum or anti-PKR sera (consisting of a mixture of anti-pPKR and anti-dsRNA domain polyclonal sera and monoclonal antiserum to mPKR) followed by immunoprecipitation. Supernatants were then subjected to dsRNA-agarose purification and in vitro phosphorylation in the presence or absence of peIF-2 or meIF-2. pPKR protein levels (data not shown) and pPKR phosphorylation levels decreased in pPKR-immunocleared extracts compared with fractions incubated with preimmune serum (Fig. 3, compare lanes B and C). The observed decrease in both peIF-2␣ and meIF-2␣ phosphorylation levels is directly correlated with decreased pPKR levels from immunoclearing (lanes B and C).
To further define the role of pPKR in eIF-2␣ phosphorylation, dsRNA-agarose-purified ribosomal salt wash fractions from barley containing pPKR were incubated with 2-aminopurine, a potent inhibitor of PKR (38). The inhibition of meIF-2␣ and peIF-2␣ phosphorylation by barley and HeLa cell ribosomal salt wash fractions in the presence of 2-aminopurine was directly correlated with inhibition of pPKR and mPKR phosphorylation (data not shown).
The eIF-2␣ kinases specifically phosphorylate Ser 51 of meIF-2␣ (1, 2, 7). To confirm pPKR as a member of the eIF-2␣ kinase family, the ability of pPKR to specifically phosphorylate Ser 51 of eIF-2␣ was evaluated in in vitro phosphorylation reactions using a synthetic eIF-2␣ peptide substrate. A peptide substrate with the sequence ILLSELSRRRIR corresponding to residues 45-56 of meIF-2␣ (33) was synthesized and subjected to in vitro phosphorylation in the presence of either purified mPKR or partially purified pPKR. Mellor and Proud (33) reported that only eIF-2␣ kinases are capable of phosphorylating the residue corresponding to Ser 51 in this peptide in the absence of Ca 2ϩ and phosphatidylserine. Both mPKR and a dsRNA-binding kinase present in the partially purified plant extract from barley tissues phosphorylated the peptide substrate (Fig. 4A, lanes A and C, respectively). The M r 68,000 -70,000 pPKR band was visible in lane C only after prolonged exposure. The level of peptide phosphorylation decreased when purified mPKR or plant extracts were incubated with soluble dsRNA (lanes B and D) but not ssRNA (lanes A and C) prior to dsRNA-agarose binding and in vitro phosphorylation. The identity of pPKR as the kinase responsible for peptide phosphorylation was also indicated by immunoclearing experiments. Immunoabsorption of plant extracts with specific anti-pPKR sera prior to dsRNA-agarose binding resulted in decreased pPKR phosphorylation levels and a corresponding decrease in peptide phosphorylation (Fig. 4B, lanes A and B).
The synthetic eIF-2␣ peptide contains two Ser residues cor-  (lanes A, B, E, and F) or partially purified HeLa mPKR (lanes C, D, G, and H) were incubated with poly(rI)⅐poly(rC)-agarose followed by the addition of [␥-32 P]ATP and plant eIF-2 (lanes A-D) or mammalian eIF-2 (lanes E-H). In lanes B, D, F, and H, extracts were incubated with 500 g of soluble poly(rI)⅐poly(rC) prior to incubation with poly(rI)⅐poly(rC)-agarose. Radiolabeled proteins were eluted, separated by SDS-PAGE, and visualized by autoradiography. The positions of molecular mass markers (in kDa), PKR, peIF-2, and meIF-2 are shown. responding to positions 48 and 51 of meIF-2␣. These are separated by a single cleavage site for V8 protease between residues corresponding to Glu 49 and Leu 50 . The positively charged fragment (LSRRRIR) containing Ser 51 can be separated from the negatively charged fragment containing Ser 48 (ILLSE) by electrophoretic thin layer chromatography (33). When the peptide was in vitro phosphorylated in the presence of [␥-32 ]P)ATP and mPKR or the plant dsRNA-binding kinase followed by V8 digestion and peptide separation, results show that the residue corresponding to Ser 51 of the eIF-2␣ was specifically phosphorylated by mPKR and the plant dsRNA binding kinase (data not shown).
Taken together, these results demonstrate that the M r 42,000 doublet that is the ␣ subunit of plant eIF-2 and the M r 36,000 ␣ subunit of meIF-2 can be specifically phosphorylated in vitro by the M r 68,000 -70,000 plant-encoded dsRNAdependent protein kinase, pPKR, as well as the eIF-2␣ kinase, mPKR.
Since phosphorylation of eIF-2␣ by mPKR inhibits protein synthesis and pPKR is capable of phosphorylating plant eIF-2␣, we were interested in determining if plant protein synthesis is inhibited in the presence of phosphorylated eIF-2. Phosphorylation of meIF-2 and peIF-2 was accomplished by incubation with mPKR bound to poly(rI)⅐poly(rC)-agarose in the presence of unlabeled ATP and phosphorylation buffer. Following incubation, mPKR bound to poly(rI)⅐poly(rC)-agarose was pelleted, and the supernatant containing the phosphorylated meIF-2 or peIF-2 was added to translation reactions. Controls consisted of 1) phosphorylation mix (containing only buffer and ATP), 2) phosphorylation mix containing unphosphorylated peIF-2, 3) phosphorylation mix containing unphosphorylated meIF-2, and 4) phosphorylation mix alone incubated previously with mPKR bound to poly(rI)⅐poly(rC)agarose. Controls 1-3 were incubated with poly(rI)⅐poly(rC)agarose in the absence of mPKR. Fig. 5 shows that in vitro  E and F, respectively).
The observations that dsRNA inhibits in vitro translation in rabbit reticulocyte lysates via a mPKR-mediated mechanism (39) and that pPKR is constitutively present in wheat germ (15) encouraged us to determine if dsRNA addition to wheat germ would similarly inhibit protein synthesis. In the absence of exogenous dsRNA, BMV RNA is efficiently translated in vitro in wheat germ lysates (Fig. 6, lane A). The addition of dsRNA to wheat germ in vitro translation lysates inhibited translation of BMV RNA in a concentration-dependent manner (lanes B-E). In rabbit reticulocyte lysates, 0.1 g/ml dsRNA is sufficient to totally inhibit protein synthesis (39), whereas in Fig.  6A, 10 g/ml dsRNA was required to significantly inhibit protein synthesis in wheat germ lysates. However, this is consistent with the dsRNA levels necessary to activate pPKR in vitro and in vivo in plant protoplasts (40). As shown in Fig. 6B, the specificity of inhibition is indicated by the fact that translation was not inhibited by ssRNA. It should be noted that we found significant variation between wheat germ lysate preparations in terms of dsRNA levels required for inhibition, with several showing no response to dsRNA (data not shown). This effect, we believe, is due to varying levels of a PKR inhibitor present in the lysates.
We next determined if exogenous mPKR is capable of inhibiting in vitro translation in wheat germ lysates. In the absence of mPKR, translation of BMV RNA was comparable with Fig. 6, lane A. The addition of purified mPKR alone decreased in vitro translation of BMV RNA in wheat germ lysates (compare Fig.  7, lane A with Fig. 6, lane A). Only the major BMV translation products of M r 94,000 and M r 20,000 were detectable. This is likely due to the activation of mPKR that is typically observed during the purification process. However, BMV translation in wheat germ lysates was completely inhibited in the presence of mPKR and low dsRNA concentrations (0.1 g/ml) (Fig. 7, lane  B). This concentration of dsRNA was much lower than that required in Fig. 6, suggesting interaction with mPKR rather than pPKR. Furthermore, this is the optimal dsRNA concentration required for mPKR shutdown of rabbit reticulocyte translation. The addition of ssRNA or buffer alone had no effect on translation (data not shown).
The findings that plant protein translation is inhibited by phosphorylated eIF-2, activated mPKR, or dsRNA indicates that components necessary for the in vivo regulation of protein synthesis via an eIF-2␣ phosphorylation mechanism are present in plants similarly to other eukaryotes. The dsRNAdependent inhibition of protein synthesis data is in apparent conflict with previous findings (20,41). However, we believe that the existence of a plant-encoded M r 67,000 (p67) inhibitor of pPKR and eIF-2␣ phosphorylation is, in part, responsible to the lack of a consistent dsRNA-mediated inhibition of protein translation in wheat germ lysates. 2 The mammalian analog of p67 is tightly associated with meIF-2 and inhibits phosphorylation (and hence, activation) of eIF-2␣ kinases and eIF-2␣ phosphorylation (42). In separate studies, we found that mammalian p67 inhibited meIF-2␣ and peIF-2␣ phosphorylation in the presence of either purified mPKR or partially purified plant fractions containing pPKR. Phosphorylation of mPKR and pPKR in the presence of p67 was correspondingly decreased. Based upon Western blotting using monoclonal antiserum, we determined that the plant-encoded p67 inhibitor is temporally regulated during plant growth and development. 2 Differential activation of the plant p67 analog may account for variations in dsRNA sensitivity between wheat germ lysate preparations and the apparent difficulty in characterizing the plant eIF-2␣ phosphorylation pathway. These results suggest that plants have a previously unappreciated mechanism to regulate cellular response to various environmental signals.