Structure-function studies on phytochrome. Preliminary characterization of highly purified phytochrome from Avena sativa enriched in the 124-kilodalton species.

A new protocol for the purification of phytochrome from etiolated Avena sativa L. seedlings in 15-25% yield is described. This preparation exhibits molecular properties dissimilar to those previously reported from oats. Phytochrome prepared by this method contains a dominant species having a Mr = 124,000 (70% of total) on sodium dodecyl sulfate-polyacrylamide gels, in addition to a chromoprotein species with Mr = 118,000 (30%). A blocked NH2 terminus is indicated by amino acid sequencing of the purified chromoprotein preparation. By ultraviolet-visible spectroscopy, this preparation displays a 730-nm absorption maximum for Pfr, an enhanced APfr730/APfr668 ratio under saturating red light illumination, and a reduced amount of dark reversion. Circular dichroism reveals a new absorption band for Pfr in the blue spectral region. These spectral properties and electrophoretic behavior are consistent with available information on phytochrome in vivo or in rapidly prepared extracts from oat seedlings. Thus, the molecular properties described here may more accurately reflect those of phytochrome in vivo in etiolated A. sativa shoots than those of phytochrome species isolated by previous methodologies.

immunochemical identification of phytochrome from crude extracts of etiolated oat seedlings (3) and etiolated rye seedlings (4) has shown that endogenous phytochrome in uiuo exhibits a monomer M, near 124,000. The purification of this larger, native species of phytochrome has not yet been reported. The present work was undertaken for two reasons: to develop a preparative purification method for phytochrome which minimizes the degradation encountered in previously reported procedures (2,(5)(6)(7) and to provide a preliminary characterization of the purified chromoprotein.
In this paper, we report the purification of phytochrome from etiolated oat seedlings which is enriched in the 124-kDa phytochrome species as demonstrated by NaDodS0,-PAGE.

RESULTS
Phytochrome Purification-A rapid and high yield procedure for the purification of phytochrome from etiolated oat seedlings enriched in the 124-kDa polypeptide has been realized by this work. This purification protocol has been adapted from the methods pioneered by Bolton and Pratt (13,14) and further developed by Litts (15); however, significant modifications have been made to afford a more stable and largely undegraded phytochrome product. The purification can be accomplished in 36 h and routinely yields 15 to 25% of the endogenous phytochrome in a highly purified form. Table I summarizes a representative yield and degree of purification sium phosphate buffer; 2-ME, 2-mercaptoethanol; NMM, N-methylmorpholine; Pf,, far red light-absorbing form of phytochrome; P,, red light-absorbing form of phytochrome; PEG, polyethylene glycol; PEI, polyethyleneimine; PMSF, phenylmethylsulfonyl fluoride; SAR, specific absorbance ratio which is the ratio A a A ; ; for P,; TASEG, a buffer containing 50 mM Tris base (T), 100 mM (NHJ2SOI (AS), 25% (v/v) ethylene glycol (EG), 1 mM EDTA, 1 mM PMSF, and 1.4 mM 2-mercaptoethanol (pH 7.8 at 4 'C) unless otherwise stated; TEG, a buffer containing 50 mM Tris base (T) and 25% (v/v) ethylene glycol (EG) (pH 7.8 at 4 "C). Per cent yields are relative to the recovery of phytochrome in the PEI supernatant. The PEI supernatant has been shown to contain in excess of 95% of the extractable phytochrome from etiolated oat shoots (15). Due to scatter in the PEI supernatant, the direct measurement of Am is not possible. This A& values has been calculated from the A(AA) value using a ratio of 0.92 for A&/A(AA) obtained empirically for more purified phytochrome samples.
Overall yield of HA2 Pool A + Pool B.
(based on SAR) of this phytochrome species from 1 kg of fresh oat tissue. Similar yields were obtained using 1 kg of frozen tissue or 100 g of lyophilized tissue.
In the crude extraction, the use of EPPS in the grinding buffer was found to be superior to Tris because of the reduced temperature dependence of the pH of the EPPS buffer system. This consideration was found to be more important when frozen tissue was extracted. PMSF (1-2 mM) and low concentrations of 2-mercaptoethanol (20 mM in the grinding buffer and 1.4 mM in all other buffers) were found to reduce proteolysis markedly. High concentrations of 2-mercaptoethanol (0.1 M) led to significant degradation of phytochrome. The reduced concentrations of 2-mercaptoethanol used in the present study necessitated the addition of Na2S03 in the grinding buffer to prevent phenolic oxidation problems (5,16,17). Reduced amounts of 2-mercaptoethanol ( i e . 7.5 mM) in the absence of Na2S03 in the extraction buffer led to a phytochrome product which contained significant covalent modification possibly from reactions with oxidized phenolic species (18). Phytochrome purified in this way exhibited a new absorption maximum at 320 nm. The species responsible for this absorption could not be separated from the purified chromoprotein by gel filtration on A-1.5m-agarose in 6 M urea. The presence of PMSF and Na2S03 had little influence on the efficacy of the initial PEI and ammonium sulfate precipitations as described previously (13,15). The addition of PEI resulted in no loss of phytochrome in the crude extract after centrifugation, but effected a significant loss of UVabsorbing material, especially nucleic acids (13, 15). After subsequent ammonium sulfate fractionation and resuspension in TASEG buffer, a 70 to 90% recovery of phytochrome was obtained, with a 40-to 50-fold increase in purity as determined spectrophotometrically.
Following ammonium sulfate fractionation, the phytochrome solution was adjusted to l M in (NH&S04 and applied to the pentyl agarose column. The results of a representative pentyl agarose chromatographic separation are shown in Fig.  1. This column and the subsequent (NHJ2S04 concentration of the pooled phytochrome-containing fractions typically resulted in a 4-to 5-fold purification of phytochrome with recoveries of 60 to 80%. The use of propyl and butyl agarose for the purification of phytochrome at this stage was also investigated. These columns proved to be inferior to pentyl agarose because of the reduced interaction of phytochrome with these resins in TASEG buffers. A TASEG wash buffer containing 1 M (NH&SO4 eluted phytochrome from butyl agarose whereas under identical conditions phytochrome remained bound to pentyl agarose. Adsorption of phytochrome to butyl agarose could only be accomplished by reducing the ethylene glycol concentration to 10% by volume. However, when chromatographed in this way, phytochrome was significantly proteolytically degraded. In contrast, phytochrome purified on pentyl agarose as described here exhibited little evidence of proteolytic breakdown. Moreover, pentyl agarose chromatography afforded significant proteolytic stability to the pooled phytochrome fractions when left at 4°C for 48 h even in the absence of PMSF. A 3-fold purification of phytochrome was obtained by HA chromatography. In TASEG buffers, phytochrome remained bound to the HA column when washed with 15 mM phosphate. Elution was accomplished with an exponential gradient to 100 mM phosphate, consistent with the results of Litts (15). A 70-80% recovery of phytochrome with a 3to 4-fold purification was typically obtained from HA chromatography. Ammonium sulfate was then used to concentrate the pooled phytochrome fractions to a level where the PEG fractionation was effective. Little purification was effected by this third (NHJ2S04 precipitation step which was utilized only if the phytochrome concentration was less than 8 x M (ie. A& < 0.1 for 1-cm path length). The pelleted phytochrome was resolubilized in TASEG to a concentration between 1.6 X and 8 X M, which was necessary for the quantitative precipitation of phytochrome with PEG at 150 g/liter. PEG fractionation effected a 2-fold purification of phytochrome and 80-90% yields were typically obtained. PEGpurified phytochrome exhibited SAR values between 0.6 and 0.8 and could be stored for several months at -20 "C in TASEG buffer without measurable loss of spectral activity.
The final step of this purification method utilized HA chromatography. An HA column served two functions: 1) the removal of excess PEG and 2) the final purification and concentration of phytochrome affording a product with SAR values between 0.80 and 0.95. Two pools of phytochrome were obtained from this column. Pool  The resolubilized (NH4),S04 precipitate containingphytochrome was applied to a pentyl agarose column as described under "Materials and Methods." Elution of phytochrome with an exponential gradient (W) increasing in ethylene glycol (25 to 35% by volume) and decreasing in (NH,),SO, (1000 to 10 mM) was monitored at 280 nm (0) and at 668 nm (A). The phytochrome-containing fractions (117 to 190 ml) were pooled.

Phytochrome from Avena sativa 11027
and represented 60430% of the applied phytochrome sample. Pool B, eluted with 100 mM phosphate, represented 5 1 0 % of the applied phytochrome sample and was typically less pure Phytochrome Characterization-Discontinuous NaDodS0,-PAGE utilizing 7% gels indicated that a representative phytochrome sample prepared by this method migrates as a single species with an estimated molecular mass of 124 kDa (Fig.   2a, lane I). Coomassie blue-stained gels run with 1 to 10 pg of purified phytochrome were scanned and indicated a protein purity exceeding 95% (results not shown). Phytochrome purified with a protocol similar to that of Litts (15) in the presence of PMSF was also examined by electrophoresis (Fig.  2a, lane 2). This purification method, which did not employ a pentyl agarose chromatographic step and utilized reduced ionic strength buffers for HA chromatography, yielded a more degraded product.
A more highly resolving 6-9% acrylamide gradient gel system was used to analyze the same phytochrome preparation (Fig. 2b). Our phytochrome preparation contains predominantly one species with an estimated M, = 124,000 (Fig. 2b, lane 1); however, a second species with an estimated M, = 118,000 is resolved with this gel system. Gel scans of AgN03stained gels indicated that the 124-and 118-kDa species were present in 70 and 30% relative amounts. Similar analysis of other phytochrome samples prepared by the present protocol (SAR < 0.5).

-45
FIG. 2. Gels of purified phytochrome. a, Coomassie bluestained NaDodS0.-PAGE (7.0% acrylamide) of purified phytochrome prepared by the present protocol ( l a n e I ) and by the modified method of Litts (Ref. 20,lane 2). Lane 3 shows the protein molecular weight standards: myosin (200,000); &galactosidase (116,300); phosphorylaae b (92,500); bovine serum albumin (66,200); and ovalbumin (45,000). One pg of the phytochrome sample and each of the protein standards were applied to the gel. b, AgN03-stained NaDodSOI-PAGE ( 6 9 % acrylamide gradient) of purified phytochrome prepared by the present protocol ( l a n e I ) , prepared by immunoaffinity purification (Ref. 6, lane 2), mixture of both phytochrome samples ( l a n e 3). Lane 4 shows the protein molecular weight standards (see a) including RNA polymerase 88' (145,000). Loads of 0.08 pg of each protein were applied to the gel. has shown that the enrichment in the 124-kDa species can be as great as 85%, although values between 60 and 80% enrichment are typically obtained. The purity of these phytochrome preparations was estimated to be 95 & 5% on overloaded gels stained with AgN03 (1-2 pg of phytochrome; results not shown). Immunoaffinity-purified phytochrome with a SAR of 0.62 (6) exhibited two species on this gradient gel system with M, corresponding to 118,000 and 114,000 (Fig. 2b, lane 2). Coelectrophoresis of our phytochrome sample with the immunoaffinity-purified sample (Fig. 2b, lane 3) demonstrates resolution of the three phytochrome species with M, = 124,000, 118,000, and 114,000.
Sequence analysis of 9.6 nmol of the purified phytochrome sample (shown in gels of Fig. 2, a and b) indicated the presence of less than 0.4 nmol of sequenceable material. In addition to suggesting that the NH, terminus is blocked, these results also indicate that the phytochrome sample contained a maximum of 5% impurities. The amount of protein analyzed was determined by spectrophotometric assay and amino acid analysis.
The absorption spectra of the 124-kDa phytochrome product in TEG buffer (4 'C) as P, and Pf, are shown in Fig. 3a. P, exhibits absorption maxima at 279,378, and 668 nm and Prr shows maxima at 279,402, and 730 nm in this buffer. The first derivative of the Pf, spectrum intersects zero at 731 nm whereas the second derivative shows a minimum at 738 nm (results not shown). Saturating red irradiation of P, gives an increase in absorbance at 730 nm (AA;;) and a decrease in absorbance of 668 nm (AAL) such that the photoequilibrium ratio AAF&/ AAk equals 0.92. This saturating red light treatment produces a Pfr species with absorption maxima at 730 and 674 nm with a relative ratio of intensities equal to 1.43. The phytochrome sample exhibits 100% photoreversibility and undergoes no observable bleaching, loss of photoreversibility, or precipitation after repeated photocycling in TEG buffer at 4 "C. Furthermore, dialysis into either 25 mM KPB buffer (pH 7.8 at 4 "C), 15 mM NMM-HOAc buffer (pH 7.8 at 4 "C) or TASEG buffer (pH 7.8 at 4 "C) has little measurable effect on the absorbance characteristics of these phytochrome preparations. The absorption spectra of phytochrome preparations which contain roughly 50% 124-kDa and 50% 118-kDa species are indistinguishable from that of the phytochrome preparation described in this report. A significant blue shift of the Prr absorption maximum accompanies the more extensive proteolysis of phytochrome to the 114-Phytochrome from Avena sativa kDa species.
Quantitative amino acid analysis showed that the amino acid composition of phytochrome prepared by this method is similar to those of two other phytochrome preparations from oats (5, 17). These data are summarized in Table 11 and are based on a monomer M, = 124,000. The calculation of the extinction coefficient of 124-kDa phytochrome is based on quantitative amino acid analysis of two samples, one in a native buffer system (15 mM NMM-HOAc, pH 7.8) and another in a strongly denaturing solvent (95% (v/v) HCOOH). These results (Table 111) are in close agreement with recent values determined for conventionally purified oat phytochromes (17,19). Native phytochrome and acid-denatured phytochrome exhibit nearly identical molar extinction coefficients at 279 nm (~~7 9 , 1.3 X lo5 liters mol" cm"). The red absorption maximum of native P, (E&, 1.2 x lo5 liters mol" cm" in 15 mM NMM-HOAc) underwent a 3-fold decrease in absorptivity upon denaturation in acid (ti& 4.0 X lo4 liters mol-' cm" in 95% HCOOH). A slight increase in molar absorptivity of the blue absorption band at 378 nm was observed upon acid denaturation of phytochrome. The molar extinction coefficient of the far red absorption maximum of Pf, in 15 mM NMM-HOAc under saturating red light illumination (t?gq', 6.9 x lo4 liters mol" cm") represented 57% the value observed for the red absorption maximum of P,.
Since this value of t;gq) was calculated by dividing the absorbance of 730 nm by the molar concentration of phytochrome (1-cm path length), it does not take into account the presence of residual P, at red light photoequilibrium. The true molar extinction coefficient for Pfr requires the determination of the percentage of P, and Pf, at photoequilibrium (work in progress).
The CD spectra of P, and Pf, in TEG buffer between 250 and 800 nm are illustrated in Fig. 3b. The CD spectrum of P, shows three major bands, two negative bands at 284 and 668 nm and one positive band at 370 nm similar to those previously described (20-26). The positive band at 370 nm also appears to exhibit a doublet fine structure with peaks at 365 and 375 nm which has been observed recently (26). In contrast, the CD spectrum of Pfr shows considerably reduced ellipticities. This spectrum consists of five major bands, two negative bands at 302 and 670 nm and three positive bands at 358, 430, and 720 nm. The similarities and differences of the Pfr CD spectra of the present phytochrome samples and those of others will be discussed subsequently (see "Discussion"). The calculation of the molar ellipticities shown in Table IV utilized the extinction coefficient values determined from quantitative amino acid analysis (Table 111). Comparative CD studies of P, and Pfr in the far UV region (200-250 nm) were also accomplished. In this spectral region, no measurable differences were observed (results not shown).
Dark reversion studies of phytochrome prepared by this method (in TEG and in 25 mM KPB, both pH 7.8 at 4 "c) are shown in Fig. 4. This figure illustrates the decrease in the amount of Pfr, expressed as the log of the percentage of Pf, remaining at time t in darkness versus time. The dark reversion rate is similar in the two buffers examined after 12 h, only 10% of the Pfr had undergone dark reversion to P,. These results also show that the reversion kinetics is not log linear, but appears to be composed of at least two first order rate constants. Peeling of the curves according to the method of van Liew (27)

DISCUSSION
Previously reported methods for the purification of "large" phytochrome from oat seedlings have yielded heterogeneous protein products which exhibit molecular and spectroscopic properties different from those observed in vivo (1,2). In contrast, we have succeeded in isolating phytochrome from etiolated oat seedlings in a highly purified form with molecular properties more similar to those of the endogenous chromoprotein. This phytochrome species exhibits a larger molecular weight on discontinuous NaDodS0,-PAGE, shows an increased A%/A% equilibrium ratio under red light illumination, and manifests considerably reduced dark reversion relative to those of the previously reported phytochrome species purified from oats. Our phytochrome preparation also exhibits a blocked NH2 terminus, not observed previously, and differs from other preparations in the blue spectral region of the CD spectrum.
Recent studies have shown that phytochrome exists predominantly as a single species with monomer molecular mass of 124 kDa in etiolated grass seedlings; however, during its extraction, limited proteolysis occurs to produce lower molecular mass species of 118 and 114 kDa (3,4). Immunoaffinitypurified oat phytochrome (3) and conventionally purified rye phytochrome (4) appear to be mixtures of predominantly 118and 114-kDa species. In contrast, the present phytochrome preparation is significantly enriched in the 124-kDa species, indicating that a substantial reduction in proteolysis has been accomplished with this purification method.
Two important factors contributed to the success of the present protocol: the choice of buffer systems and the use of hydrophobic chromatography early in the purification protocol. PMSF was included in all of the buffers because of the recent demonstration of its inhibitory effect on the degradation of 124-kDa phytochrome species in crude extracts (3,4). The presence of phenolics in the plant extracts necessitates the addition of a reducing agent to minimize side reactions with phytochrome. However, we observed that elevated concentrations of 2-mercaptoethanol are incompatible with PMSF for stabilizing phytochrome against breakdown. The mixtures of PMSF and 2-mercaptoethanol used in our buffers represented compromise solutions which significantly reduce the breakdown of phytochrome while preventing its modification by oxidized phenolic compounds. We are currently exploring other experimental methods which address the en-by guest on March 23, 2020 http://www.jbc.org/ Downloaded from Phytochrome from Avena sativa dogenous phenolic problem without enhancing protease activity in crude extracts from Avena. In addition to the use of PMSF/B-mercaptoethanol, the choice of TASEG buffer was also found to inhibit proteolysis during the isolation procedure. Concentrations of ammonium sulfate below 75 mM and ethylene glycol less than 25% by volumes led to an increase in the degradation of phytochrome. Similar inhibitory effects of ammonium sulfate and glycerol on the activity of endogenous yeast proteases have been observed (28,29). Elevated salt concentrations have also been shown to inhibit the major neutral protease found in etiolated oat shoots which may be responsible for the limited proteolysis of phytochrome (30).
The second major factor which contributed to the isolation of the larger species of phytochrome was the use of hydrophobic chromatography early in the isolation procedure. Alkyl and w-aminoalkyl-substituted agaroses have been used to probe the hydrophobicity of phytochrome (31), and the use of w-aminooctyl agarose has been recently reported for the final purification of phytochrome from pea seedlings (32). For the present work, the choice of the pentyl agarose matrix was made for three reasons: 1) phytochrome binds tightly in TASEG buffer which is adjusted to 1 M in ammonium sulfate; 2) phytochrome can be eluted in high yields (>70%) by simultaneously lowering the ammonium sulfate concentration, increasing the ethylene glycol concentration, and raising the pH; and 3) phytochrome eluted from this column is relatively free of contaminating proteases judging from the stability of the 730-nm P, absorption maximum of the eluted fractions (see later discussion). The binding and elution conditions which are effective for chromatography of phytochrome on pentyl agarose clearly indicate that the interaction of phytochrome with the column matrix is predominantly hydrophobic in nature (33). The use of HA chromatography prior to hydrophobic chromatography on pentyl agarose resulted in a significantly more degraded product based on NaDodS0,-PAGE analysis. For these reasons, it is evident that the early use of pentyl agarose chromatography is a major factor in minimizing the degradation of phytochrome.
In addition to the larger M, of our phytochrome preparation, repetitive Edman degradation revealed that its NH, terminus is blocked. In contrast to these results, NHn-terminal analysis of immunoaffinity-purified oat phytochrome has revealed the presence of both alanine and lysine which appear to be part of a single NHz-terminal sequence, H,N-Lys-Ala- . In that study, two major species were observed, recently determined to be 118 and 114 kDa (3). These results have been confirmed by us. Hence, based on the present evidence for a blocked NH, terminus, a facile proteolytic cleavage site exists within the NHP-terminal domain of phytochrome. Furthermore our NH,-terminal sequence studies suggest that both 124-and 118-kDa species contain blocked NH, termini. These results support the hypothesis that limited proteolysis of the native 124-kDa species occurs via an initial cleavage of a COOH-terminal polypeptide to give the 118-kDa species followed by the cleavage of a small NH,terminal polypeptide to afford the 114-kDa species. Proof for this hypothesis must await the results of more rigorous experimentation which must first address the nature of the NHZ-terminal blocking group of phytochrome species. It is interesting that rye phytochrome, purified by a blue agarose chromatographic method (35) and enriched in the 124-kDa species, also exhibits a blocked NH, t e r m i n~s .~ Several spectrophotometric properties of phytochrome have been shown to change during its extraction and purification from oats (2,11,12). These dissimilarities between the ab-W. 0. Smith, Jr., personal communication.
sorption spectra of phytochrome in vivo and in vitro are especially notable in the wavelength position of the Prr absorption maximum and in the apparent PfJP, photoequilibrium under saturating red light illumination determined by the ratio A:&/AE&. The two visible absorption maxima of Pfr in etiolated oat seedlings in vivo are red shifted by 5 to 10 nm (9)(10)(11)(12) from those observed for purified phytochrome samples from oats (5-7). Furthermore, oat phytochrome in vivo shows a significantly greater A : $ / A~ ratio under red light-induced photoequilibrium conditions when compared with purified oat phytochrome. With respect to both of these spectral characteristics, oat phytochrome purified with the present protocol is more similar to phytochrome measured in vivo than in preparations of phytochrome described to date.
Pfr prepared by the present method exhibits absorption maxima at 400 and 730 nm whereas, under similar buffer conditions, Pfr prepared by other methods (5-7) exhibits absorption maxima which are considerably blue shifted at 390 and 724 nm. These spectral differences are not a result of solvent perturbation effects, since the absorption spectra of phytochrome prepared by the present method is independent of all pH 7.8 buffers tested. These include 25 mM KPB, TEG, TASEG (with various concentrations of (NH4)zS0~,10-1000 mM and EG, 10-30% by volume), and 15 mM NMM-HOAc, at 4 "C. Furthermore, during the development of the present purification method, we have consistently observed that more degraded phytochrome preparations containing 114-kDa chromoprotein species, show blue-shifted absorption maxima for P,. These data with highly purified phytochrome preparations provide independent corroboration of the conclusions of Vierstra and Quail (12)  These data are in agreement with recent observations that the A:&/Ar& ratio decreases with the limited proteolysis of phytochrome (12). We offer three possible interpretations for these results: 1) a greater proportion of Pf, is formed under red light irradiation for these phytochrome samples compared with those described previously (5-7); 2) the extinction coefficient of P, a t 730 nm under red light-induced photoequilibrium conditions is larger for phytochrome prepared with the present protocol; or 3) there is a small proportion of P, in earlier phytochrome preparations (5-7) which is not phototransformable. The molecular basis for these spectral effects is not well understood at present.
The differences observed by circular dichroism spectroscopy between the phytochrome species obtained by the present method and those previously reported (13,(20)(21)(22)(23)(24)(25)(26) are also more significant for the Pfr form than for the P, form. The present phytochrome preparation as Pfr exhibits a major positive ellipticity at 430 nm. Only two previous CD studies on phytochrome have shown any extrema in this blue spectral region for Pfr. One of these studies, employing 60-kDa oat phytochrome, exhibited a small negative ellipticity in this region of the spectrum (22). A recent study on a more degraded oat phytochrome sample in a buffer similar to ours has revealed the presence of a CD extrema near 430 nm, although the intensity was considerably more attenuated than that of the present Pf, species (13). These results suggest that the 430-nm CD band is due to the influence of the proteolytically labile domain on the Pfr chromophoric region. Experiments are in progress which address the dependence of these new CD spectral characteristics on solvent and/or phytochrome size.
Although extensive dark reversion of Pfr to P, has been measured in purified phytochrome species from etiolated grass seedlings, in monocotyledonous plants dark reversion in uiuo has not been observed (reviewed in Refs. 1 and 36). The present analysis shows that our phytochrome preparation undergoes considerably less dark reversion than has been previously observed for purified phytochrome preparations.
In a 12-h period, 10% of the Pfr of our phytochrome preparation had reverted to P, in darkness. By comparison, dark reversion analysis of rye phytochrome (37) and immunoaffinity-purified oat phytochrome4 revealed nearly 80% reversion within this period of time. Since all experiments were conducted in phosphate buffers, the differences in the amount of dark reversion appear to be due to differences between phytochrome samples. The lesser degree of degradation of phytochrome purified with the present methodology supports the hypothesis that these differences in dark reversion characteristics may reflect the degree of degradation of the native chromoprotein. It is interesting to note that dark reversion is well documented in dicotyledonous plants (38). It is therefore intriguing to consider whether phytochrome in dicots is inherently different in structure from that of monocots or whether it is similar but has undergone limited proteolytic processing in uiuo. The stability of Pfr toward dark reversion therefore may depend on the presence or absence of this small protease-sensitive domain of the native chromoprotein.
In summary, we have obtained a purified phytochrome preparation which is enriched in the 124-kDa species. The present results show that this phytochrome preparation displays spectroscopic properties which are similar to those observed for phytochrome in uiuo, and considerably different from those previously reported. These differences are consistent with a lesser degree of degradation of the native chromoprotein during its extraction. These regions of phytochrome which are highly susceptible to proteolysis appear to be important structural and functional domains of the native holoprotein.