RubisCO的研究进展

1,5-二磷酸核酮糖羧化酶/加氧酶(RubisCO)是调节光合和光呼吸,决定净光合作用的一个关键酶;也是植物可溶性蛋白质中含量最高的蛋白质.该酶广泛存在于植物及一些微生物体内.综述了近年来有关RubisCO的一些研究进展. 包括RubisCO的基本性质、结构与功能、酶基因工程、酶活性调节及其活化酶等.     

Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Content and Activity in Wheat, Rye and Triticale

Photosynthetic parameters were measured in triticale and its parents wheat and rye. Soluble protein content in leaves, ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) content per fresh mass, total chlorophyll content, biomass yield, leaf area, leaf mass and specific leaf mass were higher but Rubisco content expressed as percentage of soluble protein, carboxylase activity, photosynthetic rate and stomatal conductance were significantly lower in rye than in wheat. Native-PAGE of Rubisco revealed that rye carboxylase was different from that of wheat. The difference was not related to either the small or large subunit of Rubisco but, may be, to the ionic and/or other properties of the Rubisco protein moiety. Triticale Rubisco was similar to wheat. For most of the studied physiological parameters, triticale showed much more similarity with wheat than with rye.     

Activity of Membrane-Associated Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase during Development of the Inner Membrane System in Potato Chloroplasts

Membranes of etioplasts and chloroplasts from potato plants (Solanum tuberosumL., cv. Zhu-kovskii) were fractionated by centrifugation in a sucrose gradient, and the fractions of primary membranes, fretlike structures, and frets and granal thylakoids were isolated. The carboxylase and oxygenase activities of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) were measured in these fractions. The membrane-bound Rubisco (mRubisco) was reconstructed in the artificial system from the primary membranes and the soluble form of Rubisco (sRubisco), both isolated from the chloroplasts. The higher carboxylase and lower oxygenase activities were found to be characteristic of this reconstructed mRubisco as compared to sRubisco. The degree of sRubisco association with primary membranes depended on the Mg²⁺concentration. The data suggested that the association of sRubisco with primary membranes occurred at the early stages of membrane formation. The biological role of mRubisco is assumed to consist in controlling the ratio between photosynthesis and photorespiration at the membrane level in plants of various physiological states.     

Low Activation State of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase in Carboxysome-Defective Synechococcus Mutants

The high-CO2-requiring mutant of Synechococcus sp. PCC 7942, EK6, was obtained after extension of the C terminus of the small subunit of ribulose-1,5-bisphosphate (RuBP) carboxylase/oxygenase (Rubisco). The carboxysomes in EK6 were much larger than in the wild type, but the cellular distribution of the large and small sub-units of Rubisco was not affected. The kinetic parameters of in vitro-activated Rubisco were similar in EK6 and in the wild type. On the other hand, Rubisco appeared to be in a low state of activation in situ in EK6 cells pretreated with an air level of CO2. This was deduced from the appearance of a lag phase when carboxylation was followed with time in cells permeabilized by detergent and subsequently supplied with saturating CO2 and RuBP. Pretreatment of the cells with high CO2 virtually abolished the lag. After low-CO2 treatment, the internal RuBP pool was much higher in mutant cells than in the wild-type cells; pretreatment with high CO2 reduced the pool in mutant cells. We suggest that the high-CO2-requiring phenotype in mutants that possess aberrant carboxysomes arises from the inactivated state of Rubisco when the cells are exposed to low CO2.     

Vacuole Cysteine Proteases and Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Degradation during Monocarpic Senescence in Cowpea Leaves

Characterisation of proteases degrading ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBPCO, EC: 4.1.1.39) was studied in the cowpea leaf during monocarpic senescence 3 and 9 d after flowering (DAF), representing early and mid pod fill. The stage at 3 DAF coincided with decrease in the metabolic parameters characterising senescence, i.e., contents of total soluble proteins, RuBPCO, and leaf nitrogen. At 9 DAF, there was a decline in total soluble proteins and an appearance of a 48 kDa cysteine protease. Characterisation of the proteases was done using specific inhibitors. Subcellular localisation at 3 DAF was studied by following the degradation of RuBPCO large subunit (LSU) in the vacuole lysates using immunoblot analyses. Cysteine proteases played a predominant role in the degradation of RuBPCO LSU at the crude extract level. At 9 DAF, expression of cysteine protease isoforms was monitored using polyclonal antibodies against papain and two polypeptides of molecular masses 48 and 35 kDa were observed in the vacuole lysates. We confirmed thus the predominance of cysteine proteases in the vacuoles during different stages of pod development in cowpea leaf.     

草酸对氧化胁迫-水稻叶片中核酮糖-1，5-二磷酸羧化酶/加氧酶降解的保护作用

以杂交稻(汕优63)为试验材料,在木村B营养液中培养至三叶期,用草酸5mmol/L预处理水稻2d,再处以氧化胁迫(用0．1mmol／L浓度的活性氧诱发剂甲基紫精处理)。结果表明MV诱发的氧化胁迫下,Rubisco及其它可溶性蛋白快速降解。草酸预处理可明显缓解Rubisco及其它可溶性蛋白的降解,降解速率分别降低1／3和1／2左右。植株经草酸处理后其叶片中几种抗氧化酶如AsA-POD、SOD、CAT活性大大提高,这可能是草酸预处理可缓解氧化胁迫下Rubisco和其它可溶性蛋白降解的重要原因。既然草酸能有效地诱导植物的抗氧化防卫反应,它可能作为一种诱抗剂来提高植物的抗逆性。     

Phylogenetic Diversity of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Large-Subunit Genes from Deep-Sea Microorganisms

The phylogenetic diversity of the ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO, E.C. 4.1.1.39) large-subunit genes of deep-sea microorganisms was analyzed. Bulk genomic DNA was isolated from seven samples, including samples from the Mid-Atlantic Ridge and various deep-sea habitats around Japan. The kinds of samples were hydrothermal vent water and chimney fragment; reducing sediments from a bathyal seep, a hadal seep, and a presumed seep; and symbiont-bearing tissues of the vent mussel, Bathymodiolus sp., and the seep vestimentiferan tubeworm, Lamellibrachia sp. The RuBisCO genes that encode both form I and form II large subunits (cbbL and cbbM) were amplified by PCR from the seven deep-sea sample DNA populations, cloned, and sequenced. From each sample, 50 cbbL clones and 50 cbbM clones, if amplified, were recovered and sequenced to group them into operational taxonomic units (OTUs). A total of 29 OTUs were recorded from the 300 total cbbL clones, and a total of 24 OTUs were recorded from the 250 total cbbM clones. All the current OTUs have the characteristic RuBisCO amino acid motif sequences that exist in other RuBisCOs. The recorded OTUs were related to different RuBisCO groups of proteobacteria, cyanobacteria, and eukarya. The diversity of the RuBisCO genes may be correlated with certain characteristics of the microbial habitats. The RuBisCO sequences from the symbiont-bearing tissues showed a phylogenetic relationship with those from the ambient bacteria. Also, the RuBisCO sequences of known species of thiobacilli and those from widely distributed marine habitats were closely related to each other. This suggests that the Thiobacillus-related RuBisCO may be distributed globally and contribute to the primary production in the deep sea.     

Diversity of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Large-Subunit Genes from Groundwater and Aquifer Microorganisms

To test our hypothesis that microbial autotrophic CO2 fixation plays an important role in subsurface systems of two large groundwater remediation projects, several anaerobic/microaerobic aquifer and groundwater samples were taken and used to investigate the distribution and phylogenetic diversity of ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) large-subunit genes. Two primer sets were designed for amplifying partial-subunit genes of RubisCO forms I and II from the DNA, directly extracted from the samples. PCR products were used to construct five clone libraries with putative RubisCO form I sequences, and two libraries of DNA amplified by form II primers. Selected clones were screened for variation by restriction fragment length polymorphism analysis, and a total of 28 clone inserts were sequenced and further analyzed. The phylogenies constructed from amino acid sequences derived from the partial RubisCO large-subunit sequences showed a distinct pattern. Diverse sequences affiliated to the cluster of green-like type IA RubisCO sequences were found, representing various obligate and facultative chemolithoautotrophic Proteobacteria, whereas type II RubisCO sequences detected were most closely related to those of thiobacilli species. An isolate obtained from aquifer enrichment culture, which has been provisionally named Halothiobacillus sp. RA13 on the basis of its 16S rDNA sequence, was found to contain both types of RubisCO genes, i.e., forms I and II. Physiological and ecological considerations are discussed in the context of additional microbial data and physicochemical properties.     

Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Specific Proteolysis in Barley Chloroplasts During Dark Induced Senescence

Intact chloroplasts were isolated from dark-senescing primary barley (Hordeum vulgare L.) leaves in order to study selective ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBPCO) degradation by the stromal and membrane fractions. RuBPCO specific degradation was estimated and characterised applying sensitive avidin-biotin ELISA method with non-modified or oxidatively modified biotinylated RuBPCO (BR) as substrates. Distinct proteolytic activities were detected. They differed in ATP and divalent metal ion dependence, protease inhibitory profile, and dynamics in the time-course of dark-induced senescence. The results supported involvement of ATP- and metal ion-dependent serine type proteolytic activity against non-modified BR early in induced senescence and appearance of ATP-independent activity at later stage. Active oxygen-modified BR was degraded by ATP-independent serine-type protease probably containing essential SH-groups and requiring divalent metal ions.     

Activities of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase and Phosphoenolpyruvate Carboxylase, and Oxygen Evolution in Transgenic Tobacco Plants

Three clones of tobacco transformed with the T-DNA of Agrobacterium rhizogenes Ri plasmid A4b cultivated in vitro on a solid agar medium neither showed pronounced morphological diversity nor significantly differed in chlorophyll (Chl) contents from control plants. However, the transformation caused a 27 to 83 % decay in leaf oxygen evolution and in both ribulose-1,5-bisphosphate carboxylase (RuBPC) and phosphoenolpyruvate carboxylase (PEPC) activities. Therefore, the transformation brought about a reduction of active PEPC as well as activable RuBPC amounts in plant tissues. Individual clones did not mutually differ. In tobacco transformed with A. rhizogenes strain TR101 and grown in soil only the mean leaf area tended to reduce. Chl contents, Chl a/b ratio, oxygen evolution, and activities of both RuBPC and PEPC were insignificantly affected by the transformation.     

Regulation of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Activity in Response to Reduced Light Intensity in C4 Plants

The light-dependent regulation of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activity was studied in 16 species of C4 plants representing all three biochemical subtypes and a variety of taxonomic groups. Rubisco regulation was assessed by measuring (a) the ratio of initial to total Rubisco activity, which reflects primarily the carbamylation state of the enzyme, and (b) total Rubisco activity per mol of Rubisco catalytic sites, which declines when 2-carboxyarabinitol 1-phosphate (CA1P) binds to carbamylated Rubisco. In all species examined, the activity ratio of Rubisco declined with a reduction in light intensity, although substantial variation was apparent between species in the degree of Rubisco deactivation. No relationship existed between the degree of Rubisco deactivation and C4 subtype. Dicots generally deactivated Rubisco to a greater degree than monocots. The total activity of Rubisco per catalytic site was generally independent of light intensity, indicating that CA1P and other inhibitors are not major contributors to the light-dependent regulation of Rubisco activity in C4 plants. The light response of the activity ratio of Rubisco was measured in detail in Amaranthus retroflexus, Brachiaria texana, and Zea mays. In A. retroflexus and B. texana, the activity ratio declined dramatically below a light intensity of 400 to 500 [mu]mol of photons m-2 s-1. In Z. mays, the activity ratio of Rubisco was relatively insensitive to light intensity compared with the other species. In A. retroflexus, the pool size of ribulose bisphosphate (RuBP) declined with reduced light intensity except between 50 and 500 [mu]mol m-2 s-1, when the activity ratio of Rubisco was light dependent. In Z. mays, by contrast, the pool size of RuBP was light dependent only below 350 [mu]mol m-2 s-1. These results indicate that, in response to changes in light intensity, most C4 species regulate Rubisco by reversible carbamylation of catalytic sites, as commonly observed in C3 plants. In a few species, notably Z. mays, Rubisco is not extensively regulated in response to changes in light intensity, possibly because the activity of the CO2 pump may become limiting for photosynthesis at subsaturating light intensity.     

Effect of Betaine on Enzyme Activity and Subunit Interaction of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase from Aphanothece halophytica

The presence of betaine, a quaternary ammonium compound, at a concentration (0.5 molar) reported to accumulate inside Aphanothece halophytica in response to increasing external salinity, slightly promoted ribulose-1,5-bisphosphate (RuBP) carboxylase activity. KCl at 0.25 molar inhibited RuBP carboxylase about 55%. Betaine relieved the inhibition by 0.25 m KCl and the original uninhibited activity was restored at 1 m betaine. Other osmoregulatory solutes such as sucrose and glycerol also reduced KCl inhibition, though to a lesser extent than betaine. Proline had no effect. The protective effect of betaine against KCl inhibition of RuBP carboxylase activity was also observed in other cyanobacteria, i.e. Synechococcus ACMM 323, Plectonema boryanum, and Anabaena variabilis, and in the photosynthetic bacterium Rhodospirillum rubrum but not in Chromatium vinosum. Apart from betaine, other quaternary ammonium compounds, i.e. sarcosine and trimethylamine-N-oxide (TMAO), but not glycine, also protected the enzyme against KCl inhibition and the effectiveness of such compounds appeared to correlate with the extent of N-methylation. Heat and cold inactivation of the enzyme could be protected by either betaine or KCl. However, best protection occurred when both betaine and KCl were present together. The K_m (CO₂) was not altered by either betaine or KCl, nor when they were present together. However, the K_m (RuBP) was increased about 5-fold by KCl, but was unaffected by betaine. The presence of betaine together with KCl lowered the KCl-raised K_m (RuBP) by about half. The extent of the dissociation of the enzyme molecule under the condition of low ionic strength was reduced by either betaine or KCl alone and more so when they were present together. Glycine, sarcosine, and TMAO were more effective than betaine or KCl in lowering the extent of the dissociation of the enzyme molecule.     

Acclimation of Two Tomato Species to High Atmospheric CO(2): II. Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase and Phosphoenolpyruvate Carboxylase

Lycopersicon esculentum Mill. cv Vedettos and Lycopersicon chmielewskii Rick, LA 1028, were exposed to two CO₂ concentrations (330 or 900 microliters per liter) for 10 weeks. The elevated CO₂ concentrations increased the initial ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activity of both species for the first 5 weeks of treatment but the difference did not persist during the last 5 weeks. The activity of Mg²⁺-CO₂-activated Rubisco was higher in 900 microliters per liter for the first 2 weeks but declined sharply thereafter. After 10 weeks, leaves grown at 330 microliters per liter CO₂ had about twice the Rubisco activity compared with those grown at 900 microliters per liter CO₂. The two species showed the same trend to Rubisco declines under high CO₂ concentrations. The percent activation of Rubisco was always higher under high CO₂. The phosphoenolpyruvate carboxylase (PEPCase) activity measured in tomato leaves averaged 7.9% of the total Rubisco. PEPCase showed a similar trend with time as the initial Rubisco but with no significant difference between nonenriched and CO₂-enriched plants. Long-term exposure of tomato plants to high CO₂ was previously shown to induce a decline of photosynthetic efficiency. Based on the current study and on previous results, we propose that the decline of activated Rubisco is the main cause of the acclimation of tomato plants to high CO₂ concentrations.     

Activities of Phosphoenolpyruvate Carboxylase and Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase in Leaves and Fruit Pericarp Tissue of Different Coffee (Coffea sp.) Genotypes

In order to study photosynthetic characteristics, phosphoenolpyruvate carboxylase (PEPC) and ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBPCO) activities as well as soluble protein and chlorophyll contents were determined in leaf and fruit pericarp samples from diverse coffee genotypes (Coffea arabica cv. Colombia, Caturra, Caturra Erecta, San Pacho, Tipica, C. stenophylla, C. eugenioides, C. congensis, C. canephora, C. canephora cv. Arabusta, C. arabica cv. Caturra×C. canephora and Hibrido de Timor. We found a slightly higher PEPC activity in fruit pericarp than in leaves, while RuBPCO activity was much lower in pericarp than leaf tissue. Partial purification of PEPC and RuBPCO was carried out from leaves of C. arabica cv. Caturra and Michaelis-Menten kinetics for RuBPCO (K_m CO₂ = 5.34 µM), (K_m RuBP = 9.09 µM) and PEPC (K_m PEP = 19.5 µM) were determined. Leaf tissues of Colombia, Hibrido de Timor, and Caturra consistently showed higher content of protein [55.4–64.4 g kg^–1 (f.m.)] than San Pacho, C. stenophylla, Tipica, Caturra Erecta, and Caturra×C. canephora [25.6–36.9 g kg^–1 (f.m.)] and C. canephora cv. Arabusta, Borbon, C. congensis, C. eugenioides, and C. canephora [16.1–21.1 g kg^–1 (f.m.)].     

Characterization of Thylakoid-Derived Lipid-Protein Particles Bearing the Large Subunit of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase

Lipid-protein particles bearing the 55-kD ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) (EC 4.1.1.39) large subunit (RLSU) and no detectable corresponding Rubisco small subunit (RSSU) were isolated from the stroma of intact chloroplasts by flotation centrifugation. Stromal RLSU-bearing particles appear to originate from thylakoids because they can also be generated in vitro by illumination of isolated thylakoids. Their formation in vitro is largely heat denaturable and is facilitated by light or ATP. RLSU-containing lipid-protein particles range from 0.05 to 0.10 [mu]m in radius, contain the same fatty acids as thylakoids, but have a 10- to 15-fold higher free-to-esterified fatty acid ratio than thylakoids. RLSU-bearing lipid-protein particles with no detectable RSSU were also immunopurified from the populations of both stromal lipid-protein particles and those generated in vitro from illuminated thylakoids. Protease shaving indicated that the RLSU is embedded in the lipid-protein particles and that there is also a protease-protected RLSU in thylakoids. These observations collectively indicate that the RLSU associated with thylakoids is released into the stroma by light-facilitated blebbing of lipid-protein particles. The release of RLSU-containing particles may in turn be coordinated with the assembly of Rubisco holoenzyme because chaperonin 60 is also associated with lipid-protein particles isolated from stroma.     

Variability in Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Small Subunits and Carboxylation Activity in Fern Gametophytes Grown under Different Light Spectra

Two distinct ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) small subunit (SSU) populations were observed in Pteris vittata gametophytes grown under different illumination conditions. Exposure of the fern gametophytes to continuous red light (R) resulted in Rubisco SSUs that were not recognized by polyclonal antibodies raised against SSUs from spinach. Unlike the R-induced SSUs, blue light (B) induced SSUs were well recognized. This difference in SSU composition also reflected in Rubisco activity. In vitro, B-induced Rubisco exhibits a significantly higher carboxylation activity as compared to the R-induced Rubisco. Approximately a two- to threefold increase in the V_max value of the B-induced carboxylase as compared to the R-induced one was measured. It thus seems very likely that certain domains in the SSU molecule affect enzyme activity.     

Effect of Mg2+ on the Structure and Function of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase

Mg²⁺ in various concentrations was added to purified Rubisco in vitro to gain insight into the mechanism of molecular interactions between Mg²⁺ and Rubisco. The enzyme activity assays showed that the reaction between Rubisco and Mg²⁺ was two order, which means that the enhancement of Rubisco activity was accelerated by low concentration of Mg²⁺ and slowed by high concentration of Mg²⁺. The kinetics constant (K _m) and V _max was 1.91 μM and 1.13 μmol CO₂ mg⁻¹ protein∙min⁻¹, respectively, at a low concentration of Mg²⁺, and 3.45 μM and 0.32 μmol CO₂∙mg⁻¹ protein∙min⁻¹, respectively, at a high concentration of Mg²⁺. By UV absorption and fluorescence spectroscopy assays, the Mg²⁺ was determined to be directly bound to Rubisco; the binding site of Mg²⁺ to Rubisco was 0.275, the binding constants (K _A) of the binding site were 6.33 × 10⁴ and 5.5 × 10⁴ l·mol⁻¹. Based on the analysis of the circular dichroism (CD) spectra, it was concluded that the binding of Mg²⁺ did not alter the secondary structure of Rubisco, suggesting that the observed enhancement of Rubisco carboxylase activity was caused by a subtle structural change in the active site through the formation of the complex with Mg²⁺.     

Xylulose 1,5-Bisphosphate Synthesized by Ribulose 1,5-Bisphosphate Carboxylase/Oxygenase during Catalysis Binds to Decarbamylated Enzyme

Xylulose 1,5-bisphosphate (XuBP) is synthesized from ribulose 1,5-bisphosphate (RuBP) at carbamylated catalytic sites on ribulose 1,5-bisphosphate carboxylase (Rubisco) with significant amounts of XuBP being formed at pH less than 8.0. XuBP has been separated by high performance liquid chromatography and identified by pulsed amperometry from compounds bound to Rubisco during catalysis with the purified enzyme and from celery (Apium graveolens var Utah) leaf extracts. XuBP does not bind tightly to carbamylated sites, but does bind tightly to decarbamylated sites. Upon incubation of fully activated Rubisco with 5 micromolar XuBP, loss of activator CO₂ occurred before XuBP bound to the enzyme catalytic sites, even in the presence of excess CO₂ and Mg²⁺. Binding of XuBP to decarbamylated Rubisco sites was highly pH dependent. At pH 7.0 and 7.5 with 10 millimolar MgCl₂ and 10 millimolar KHCO₃, the apparent dissociation constant for XuBP, K_d, was 0.03 micromolar, whereas at pH 8.0 and 8.5, the apparent K_d was 0.35 and 2.0 micromolar, respectively. This increase in K_d with pH was a result of a decrease in the association rate constant and an increase in the dissociation rate constant of XuBP bound to decarbamylated sites on Rubisco. The K_d of 2-carboxyarabinitol 1-phosphate binding to carbamylated sites was only slightly pH dependent.     

The Two Forms of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Activase Differ in Sensitivity to Elevated Temperature

Ribulose-1,5-bisphosphate carboxylase/oxygenase activase often consists of two polypeptides that arise from alternative splicing of pre-mRNA. In this study recombinant versions of the spinach (Spinacea oleracea L.) 45- and 41-kD forms of activase were analyzed for their response to temperature. The temperature optimum for ATP hydrolysis by the 45-kD form was 45[deg]C, approximately 13[deg]C higher than the 41-kD form. When the two forms were mixed, the temperature response of the hybrid enzyme was similar to the 45-kD form. In the absence of adenine nucleotide, preincubation of either activase form at temperatures above 25[deg}C inactivated ATPase activity. Adenosine 5[prime]-([gamma]-thio)triphosphate, but not ADP, significantly enhanced the thermostability of the 45-kD form but was much less effective for the 41-kD form. Intrinsic fluorescence showed that the adenosine 5[prime]-([gamma]-thio)triphosphate-induced subunit aggregation was lost at a much lower temperature for the 41-kD than for the 45-kD form. However, the two activase forms were equally susceptible to limited proteolysis after heat treatment. The results indicate that (a) the 45-kD form is more thermostable than, and confers increased thermal stability to, the 41-kD form, and (b) a loss of subunit interactions, rather than enzyme denaturation, appears to be the initial cause of temperature inactivation of activase.