Photosynthesis Research

, Volume 91, Issue 1, pp 81–89

Multiple sources of carbonic anhydrase activity in pea thylakoids: soluble and membrane-bound forms


  • Natalia N. Rudenko
    • Photosynthetic Electron Transport Laboratory, Institute of Basic Biological ProblemsRussian Academy of Sciences
  • Lyudmila K. Ignatova
    • Photosynthetic Electron Transport Laboratory, Institute of Basic Biological ProblemsRussian Academy of Sciences
    • Photosynthetic Electron Transport Laboratory, Institute of Basic Biological ProblemsRussian Academy of Sciences
Research Article

DOI: 10.1007/s11120-007-9148-2

Cite this article as:
Rudenko, N.N., Ignatova, L.K. & Ivanov, B.N. Photosynth Res (2007) 91: 81. doi:10.1007/s11120-007-9148-2


Carbonic anhydrase (CA) activity of pea thylakoids, thylakoid membranes enriched with photosystem I (PSI-membranes), or photosystem II (PSII-membranes) as well as both supernatant and pellet after precipitation of thylakoids treated with detergent Triton X-100 were studied. CA activity of thylakoids in the presence of varying concentrations of Triton X-100 had two maxima, at Triton/chlorophyll (triton/Chl) ratios of 0.3 and 1.0. CA activities of PSI-membranes and PSII-membranes had only one maximum each, at Triton/Chl ratio 0.3 or 1.0, respectively. Two CAs with characteristics of the membrane-bound proteins and one CA with characteristics of the soluble proteins were found in the medium after thylakoids were incubated with Triton. One of the first two CAs had mobility in PAAG after native electrophoresis the same as that of CA residing in PSI-membranes, and the other CA had mobility the same as the mobility of CA residing in PSII-membranes, but the latter was different from CA situated in PSII core-complex (Ignatova et al. 2006 Biochemistry (Moscow) 71:525–532). The properties of the “soluble” CA removed from thylakoids were different from the properties of the known soluble CAs of plant cell: apparent molecular mass was about 262 kD and it was three orders more sensitive to the specific CA inhibitor, ethoxyzolamide, than soluble stromal CA. The data are discussed as indicating the presence of, at least, four CAs in pea thylakoids.


Carbonic anhydrasePhotosystem IPhotosystem IIPisum sativum L.Thylakoids





Carbonic anhydrase








Polyacrylamide gel




Photosystem I


Photosystem II




Thylakoid CA (tCA) was found for the first time in algae (Semenenko et al. 1977). In 1998 one tCA was isolated from Chlamydomonas reinhardtii (Karlsson et al. 1998) and sequenced. It was identified as α-CA (Cah3) and was shown to be situated on the lumenal side of photosystem II (PSII). Information about the presence of CA activity in thylakoids of higher plant cells has accumulated since 1982 (Komarova et al. 1982; Vaklinova et al. 1982). Stemler showed CA activity of thylakoids and PSII particles from maize mesophyll chloroplasts (Stemler 1986). The existence of specific thylakoid CA has been recognized rather recently (Stemler 1997) when doubts about contamination with soluble stromal CA were overcome. The presence of CA activity in a band of membrane-bound protein in a gel after native electrophoresis of thylakoids was shown using bean (Komarova et al. 1982) and pea (Lazova 1994) chloroplasts. In previous studies we have found that the properties (Km (CO2), pH-dependence Km (HCO3¯) of tCA activity distinguished it from CA activity of soluble proteins extract; tCA activity demonstrated unusual stimulation with specific sulfonamide CA inhibitors acetazolamide and azide in small concentrations (Ignatova et al. 1998). The antibodies against soluble β-CA of spinach had no cross-reactivity with thylakoid proteins solubilized with SDS (Moskvin et al. 2004).

In recent years results implying the presence of more than one CA in higher plants thylakoids have appeared. Antibodies against periplasmic α-CA of C. reinhardtii (Cah1) labeled pea chloroplasts (Arancibia-Avila et al. 2001), while antibodies against the above mentioned Cah3 from C. reinhardtii showed cross-reactivity with a component of PSII-membranes from pea (Pronina et al. 2002) and maize (Lu and Stemler 2002) and moreover with the 33 kD extrinsic protein of maize and pea (Lu and Stemler 2002; Lu et al. 2005). PSII-membranes of maize contained two CAs one of which was called by these authors as extrinsic, that passed into the solution after treatment with high concentration of salts especially Ca2+, and the other as intrinsic, that was tightly associated with PSII components (Lu and Stemler 2002). Our recent studies have also revealed that in thylakoid membranes CA activity is present in the PSII core-complex (Khristin et al. 2004) as well as in PSI-membranes (Ignatova et al. 2006). The latter CA differed from CAs of PSII-membranes in sensitivity to specific CA inhibitors (Ignatova et al. 2006) and had a lower apparent molecular mass (Rudenko et al. 2006).

This work presents new evidences that pea thylakoids contain several (possibly four) CAs.

Material and methods

Pisum sativum plants were grown in a greenhouse in soil, at 22/18°C (day/night) and 400 μmol of photons m−2 s−1 illumination provided by tungsten halogen lamps.

Thylakoids were isolated from two upper tiers of 10–16-day-old plants according to the method developed earlier (Moskvin et al. 1995) which provided membranes free of highly active stromal CA. Thylakoids were resuspended in medium that contained 50 mM Na-K-phosphate buffer (pH 7.1), 100 mM sucrose, 2 mM ascorbate, 1 mM KHCO3, 5 mM EDTA-Na, 1 mM benzamidine, 1 mM α-aminocaproic acid, and 1 mM phenylmethylsulfonylfluoride (PMSF). Supernatant-12 and pellet-12 were obtained after thylakoids were incubated for 20 min at 0°C with nonionic detergent Triton X-100 at triton/Chl ratio of 1.0 and then centrifugated at 12000 × g for 30 min.

PSI-membranes (thylakoid membranes enriched with PSI) and PSII-membranes (thylakoid membranes enriched with PSII) were prepared from pellet-12. First, the thylakoids were treated with Triton X-100 (at triton/Chl ratio of 20.0) for 30 min, then the membrane fragments were stepwise precipitated to obtain the preparations of PSI- and PSII-membranes (for details see (Ignatova et al. 2006)). PSI-membranes had no P680, the reaction center of PSII and PSII-membranes contained only a small quantity of PSI (Chl/P700 ratio of 3685 at Chl/P680 ratio of 376).

Electrophoresis was carried out in 7% and 13% PAA cylindrical gels, containing 12.4 mM tris-48.6 mM glycine, pH 8.5 (Peter and Thornber 1991). The upper electrode buffer contained 0.2% of derifate-160. Electrophoresis was carried out at 4°C and dim light at current strength of 0.8–1 mA on tube. The samples of supernatant-12 were treated with 0.1% dodecyl-β-D-maltoside (DM) for 30 s, PSI- and PSII-membranes were treated with DM at DM/chlorophyll ratio of 15 for 20 s. on a Vortex mixer before loading onto the gel.

Visualization of CA activity in PAAG was carried out as in (Edwards and Patton 1966). Gels were placed in 44 mM veronal buffer, pH 8.1, and incubated for 30 min with 0.2% bromthymol blue, then the gels were placed into water saturated with CO2 at 0°C. The blue gel became yellow at the places where CA was situated.

CA activity of samples was measured as CO2 hydration at 2°С in 13.6 mM veronal buffer (pH 8.4) (Wilbur and Andersen 1948). Water, saturated by bubbling with CO2 at 0°C for 1 h, was added to the reaction mixture up to 36% of the final volume, and the time for the pH to decrease from 8.3 to 7.8 was recorded. CA activity was calculated as the difference between the rates in the presence and in the absence of enzyme preparation in the medium. Assay probe and reaction mixture were titrated with HCl to take into account the difference in buffer capacity in both the cases. CA activity was expressed on chlorophyll or protein basis.

The proteins of supernatant-12 were precipitated by incubation for 10 min at 0°С with acetone added gradually to final concentration of 70%. Proteins were precipitated at 2500 × g for 10 min. The pellet was incubated for 1 h at 0°C in 0.05 M tris–H2SO4, pH 8.5, and then centrifuged at 13000 × g for 30 min; the supernatant contained the soluble proteins. Then the pellet was incubated in the same buffer but also containing 8 M urea and detergents, 0.1% derifate, 0.1% Triton X-100, and precipitated as above to obtain the preparation of membrane-bound proteins. Protein content was determined according to (Bradford 1976).

Soluble proteins from pea leaves were extracted after grinding leaves with a pestle in the presence of broken glass in a cooled mortar in the buffer containing 0.1 M Tris–H2SO4 (pH 8.1), 5 mM dithiothreitol, 1 mM EDTA, 1 mM PMSF, and Polyclar AT (2% of leaves weight). The homogenate was centrifuged at 150 × g for 1 min, then the supernatant was centrifuged at 13000 × g for 30 min, and the new supernatant was centrifuged at 100000 × g for 1 h. The supernatant after last centrifugation was used. Protein content was determined after precipitation with TCA according to (Lowry et al. 1951).

Chlorophyll was determined in ethanol extracts according to (Winterman and De Mots 1965).


CA activity of thylakoids and membrane fragments enriched either with PSI or PSII

Figure 1 shows the dependencies of CA activity of thylakoids, PSI-membranes, and PSII-membranes on the triton/Chl ratio. Two peaks of CA activity were observed in thylakoids after 20 min of incubation with Triton at triton/Chl ratios of 0.3 and 1.0 (Fig. 1). The detergent stimulation of enzyme activity is the property of membrane-bound CAs (Utsunomiya and Muto 1993), and the increase of CA activity at the different Triton concentrations indicates the presence of two distinct pools of CA activity carriers. PSI- and PSII-membranes were treated in the same way and maximum of CA activity of PSI-membranes was observed at the triton/Chl ratio of 0.3 and maximum of CA activity of PSII-membranes—at the ratio of 1.0 (Fig. 1). This fact implies that the two peaks of CA activity of thylakoids are presumably conditioned by the presence in thylakoids of at least two CAs, one of that resides close to PSI and the other—close to PSII.
Fig. 1

The effect of Triton X-100 on CA activity of thylakoids (□), PSI-membranes (○) and PSII-membranes (▼). 100% is CA activity in the absence of Triton; of 61.2 μmol H+(mg Chl)−1 min−1 for thylakoids, 1060 μmol H+(mg Chl)−1 min−1 for PSI-membranes and 78 μmol H+(mg Chl)−1 min−1 for PSII-membranes. The bars are the standard deviations

CA activity passing to solution from thylakoids

Further investigations of Triton treated thylakoids showed that the two peaks of CA activity were observed also in the case of the pellet obtained after spinning thylakoids at 12000 × g (Fig. 2, curve 1). However, these peaks, especially at triton/chl ratio of 1.0 were connected not only with stimulation of CA activity associated with thylakoid membranes but also with the passage of some CA activity to the solution (Fig. 2, curve 2). Electrophoresis of the supernatant obtained after centrifugation at 12000 × g thylakoids incubated without any additions showed the presence of CA activity band in the region of low molecular mass proteins (Fig. 3A), however, only after treatment of the preparation with 0.1 % DM before application to the gel. After thylakoids were incubated with Triton X-100 at triton/Chl ratio of 0.3 (Fig. 3B), an additional band of CA activity was found in the upper part of the gel. At triton/Chl ratio of 1.0, one more CA passed into solution, “L” in Fig. 3C. Without DM treatment, membrane-bound CAs remained at the start of the gel (Fig. 3D), while a single CA activity band of the same mobility as “L” was also seen in the gel. Therefore, the mobility of “L“ did not depend on DM treatment. Rather, it was typical for soluble forms of the CA enzymes. «L» remained in solution after centrifugation of supernatant-12 at 144000 × g for 1 h (Figure 3E, “L”). At the same time, the electrophoresis showed that the pellet obtained after this centrifugation did not contain CA of the same mobility (Fig. 3F). It is known that high molecular mass stromal and cytoplasmic β-CAs are present in higher plants cells (Rumeau et al. 1996; Kisiel and Graf 1972; Kimber and Pai 2000). Separation of pea leaf soluble proteins under our conditions of electrophoresis demonstrated that the apparent molecular masses of soluble CAs were about 234 and 312 kD (Fig. 3G). “L” had apparent molecular mass about 262 kD (Fig. 3C, D, E).
Fig. 2

The effect of Triton X-100 on CA activity in the pellet (1) and supernatant (2) obtained after centrifugation at 12000 × g for 30 min of thylakoids treated with Triton X-100. 100% is CA activity of thylakoids in the absence of Triton of 108.8 μmol H+min−1. The bars are the standard deviations
Fig. 3

The distribution of CA activity after native electrophoresis in 7% PAAG. (A) supernatant obtained after centrifugation, at 12000 × g for 30 min of thylakoids incubated without Triton; (B) treated with Triton X-100 at triton/Chl ratio of 0.3; (C, D) at triton/Chl ratio of 1.0 (supernatant-12); (E, F) supernatant and pellet respectively after centrifugation of supernatant-12 at 144000 × g for 1 h; (G) soluble proteins of pea leaves. (A, B, C, F) preparations were treated with 0.1 % dodecylmaltoside before loading the gel. (D, E, G) preparations without treatment with dodecylmaltoside. Gel G was loaded with 140 μg of protein. Arrows show CA activity; molecular mass markers are shown on the right

Fig. 3 (A, B, C, F) also shows the band of CA activity in the region of the low molecular mass proteins. The mobility of this band was the same as the mobilities after electrophoresis in 7% PAAG of CAs one of which resided close to PSI and the other which resided close to PSII (Ignatova et al. 2006). From this past work we surmise that this band contained two CAs with the same mobilities in 7% PAAG and these CAs partly passed into solution from thylakoids during incubation with Triton and even without Triton addition.

The effect of specific CA inhibitors and DTT

We used two sulfonamide inhibitors: ethoxyzolamide (EZ) that was able to penetrate easily into membranes and acetazolamide (AA) that was poorly capable to permeate into membranes. Figure 4 shows the dependencies of CA activity of pea leaves soluble proteins (Fig. 4А), pellet-12 (Fig. 4B), and supernatant-12 (Fig. 4C) on inhibitors concentration. The highest sensitivity of CA activity to EZ was found in supernatant-12. CA activity was completely inhibited at EZ concentration of 10−7 M (Fig. 4C), whereas CA activity of pea leaves extract that contained soluble stromal and cytoplasmic CAs was completely suppressed at EZ concentration of 10−4 M (Fig. 4А). AA in low concentrations (10−8 M) stimulated CA activity of pellet-12 (Fig. 4B). This effect was found on CA activity of thylakoids (Moskvin et al. 1995) and PSII-membranes (Ignatova et al. 2006). Such stimulation was not observed on CA activity of soluble proteins and supernatant-12 (Fig. 4A, C).
Fig. 4

The effect of CA inhibitors acetazolamide (AA) (1) and ethoxyzolamide (EZ) (2) on CA activity of pea leaves soluble proteins (A), pellet-12 (B) and supernatant-12 (C) (see Methods). 100% is the activity without inhibitor: Fig. A, 2578 μmol H+(mg Pr)−1 min−1 for experiments with both inhibitors; Fig. B, 31.7 μmol H+(mg Chl)−1 min−1 for experiments with AA and 189.3 μmol H+(mg Chl)−1 min−1 for experiments with EZ; Fig. C, 71.7 μmol H+(mg Pr)−1 min−1 for experiments with AA and 88.9 μmol H+(mg Pr)−1 min−1 for experiments with EZ. The bars are the standard deviations

One more property distinguishes CA activity of thylakoids from CA activity of the stromal enzyme. DTT, the disulfide groups protector is used for isolation of plant soluble stromal CA that includes numerous sulfur-containing amino acids (Graham et al. 1984). DTT had no effect on CA activity of both supernatant and pellet of thylakoids obtained after incubation with DTT for 20 min at 0°C and centrifugation at 12000 × g for 30 min (Table 1). Probably, thylakoid CAs contain less sulfur-containing amino acids than stromal CA or their activity does not depend on the redox state of these amino acids.
Table 1

The effect of DTT on CA activity of supernatant and thylakoid pellet obtained after centrifugation at 12000 × g for 30 min of thylakoids incubated with DTT (were taken equal volumes of probes)

Additions in incubation medium

CA activity, μmol H+(min)−1




146 ± 47

68 ± 18

104 ± 18

5 mM DTT

141 ± 43

66 ± 17

113 ± 14

Supernatant-12 proteins showing CA activity

The electrophoresis of water-soluble proteins extracted from an acetone pellet of supernatant-12 showed one CA activity band (Fig. 5A) approximately in the same region of high molecular mass proteins that was observed in the gel after electrophoresis of supernatant-12 not treated with DM (“L” on Fig. 3C). After electrophoretic separation of membrane-bound proteins of supernatant-12 (apparently with some contamination by soluble proteins) in 7% PAAG, there was one more CA activity band in the low part of the gel (Fig. 5B), similar to the low band on Fig. 3 (A, B, C, F). We suppose that this band contained two CAs and their mobility was the same because previously we have found in thylakoids two low molecular mass CAs with close mobility after electrophoresis in 7% PAAG (Ignatova et al. 2006). The difference in their mobility became visible after electrophoresis in more dense gels (Rudenko et al. 2006). Actually, after electrophoresis in 13% PAAG of supernatant-12 extract obtained straightway in the presence of urea and detergents there were three CA activity bands in the gel (Fig. 5C). We suppose that the top band was “L”, because it was shown that “L” was a high molecular mass protein (Fig. 3C), the middle was CA residing close to PSII (it’s apparent molecular mass was about 50 kD (Rudenko et al. 2006), and the lowest band was CA residing close to PSI (it’s apparent molecular mass was about 20 kD (Rudenko et al. 2006).
Fig. 5

The native electrophoresis in 7% PAAG (A, B) and in 13% PAAG (C) of soluble (A), membrane-bound (B) proteins (see Methods) and soluble plus membrane-bound proteins (C)


Maximum of CA activity of PSII-membranes (Fig. 1) and one of the maxima of CA activities of both thylakoids (Fig. 1) and pellet-12 (Fig. 2, curve 1) were observed when these preparations were treated with Triton at triton/Chl ratio of 1.0. Apparently this effect was due to the presence of CA in the PSII core-complex. This CA did not pass into solution after Triton or CaCl2 treatments (Khristin et al. 2004). However, the most proteins of PSII core-complexes are well-studied, and CA still has not been revealed among them. Therefore we do not exclude the possibility that the active site of this CA is formed by several subunits of well-known proteins of PSII core-complex, similar to the active site of CA γ-family members (Kisker et al. 1996).

It was found that the addition of Zn2+ and Mg2+ decreased CA activity of isolated PSII core-complexes of higher plants whereas the addition of Ca2+ and Mn2+ increased it (Moskvin et al. 2004). Previously from other data we proposed that a metal in the active center of CA situated in PSII core-complex may not be zinc as in ordinary CAs. It is known that the presence of bicarbonate is necessary for electron transfer from QA to QB (van Rensen et al. 1998). We propose that the function of CA situated in the core-complex is to provide bicarbonate to this site (CA1 on Fig. 6).
Fig. 6

The tentative scheme of CAs arrangement in the thylakoid. CA1—is in PSII core-complex close to the acceptor site of bicarbonate action; CA2—is on the lumenal side of thylakoid membrane, near PSII water-oxidizing complex; CA3—is in stromal thylakoid membranes, close to PSI; CA4—is in thylakoid lumen

After thylakoids were treated with Triton X-100 at triton/Chl of 1.0, three CAs passed into solution (supernatant-12). One of these CAs that was observed in 13% gel after electrophoresis of the proteins of supernatant-12 (Fig. 5, gel C) and previously in the region of low molecular mass proteins in 7% gel after electrophoresis of PSII-membranes (Ignatova et al. 2006). This CA probably resided close to PSII, it’s apparent molecular mass was estimated as 50 kD (Rudenko et al. 2006). An increase of CA activity in presence of AA at low (about 10−8—10−7 M) concentrations was observed in pellet-12 (Fig. 4B), thylakoids (Moskvin et al. 1995), BBY-particles (Moskvin et al. 2004), PSII-membranes, and eluate from the part of the gel with the low molecular mass proteins after PSII-membranes electrophoresis (Ignatova et al. 2006). Moreover, CA activity of eluate of the gel’s part, enriched with PSII core-complexes after PSII-membranes electrophoresis (the above “CA1”), did not increase, but decreased under АА action in concentration of 10−8 M (not shown). The increase of CA activity by AA is characteristic of low molecular mass CA that resides close to PSII. Besides this CA was highly sensitive to EZ with I50 = 10−9 M (Ignatova et al. 2006). It was found that Cah3, α-CA of C. reinhardtii that was situated on the lumenal side of PSII, was inhibited with EZ with I50 of 6 × 10−9 M (Mitra et al. 2005) and proteins of PSII-membranes of pea (Pronina et al. 2002), and maize (Lu and Stemler 2002) had cross reaction with antibodies against Cah3. Thus, it would be possible that Cah3 from algae and CA residing close to PSII from higher plants both have similar amino acid sequences. It was found that OEC33, 33 kD protein of PSII (both native and recombinant), possessed CA activity (Lu et al. 2005), but the OEC33 and Cah3 are proteins of completely different structure.

As has been found over the past several years, bicarbonate stabilizes the Mn-cluster on the PSII donor side, and this effect was especially expressed in Mn-depleted PSII preparations (Allakhverdiev et al.1997; Klimov and Baranov 2001). We suppose that the low molecular mass CA resides close to PSII on the lumenal side of thylakoid membranes and it may be responsible for supplying the water-oxidizing complex with bicarbonate and/or for the removal of excess protons released during oxygen evolution particularly at high light intensity (Villarejo et al. 2002) (CA2 on Fig. 6).

The other CA activity maximum in thylakoids treated with Triton X-100 was observed at triton/Chl ratio of 0.3 and CA activity of PSI-membranes was also the highest at the same triton/Chl ratio (Fig. 1). As it has been shown earlier (Ignatova et al. 2006; Rudenko et al. 2006) CA activity of PSI-membranes was higher than PSII-membranes on both chlorophyll and protein basis. CA activity of PSI-membranes was equally sensitive to both sulfonamide inhibitors AA and EZ with I50 = 10−6 M (Ignatova et al. 2006). It’s apparent molecular mass was about 20 kD (Rudenko et al. 2006). Presumably this low-molecular mass CA that resided in thylakoids close to PSI, partly passed into the supernatant, received after centrifugation at 12000 × g thylakoids incubated with Triton (Fig. 3B, C). It was also observed in the acetone pellet of supernatant-12 proteins (Fig. 5C, the lowest band). The presence of CA activity in PSI-membranes from pea thylakoids has been shown by Pronina et al. (2002) and it was found that proteins of PSI-membranes had no cross-reactivity with antibodies against Cah3, the α-CA of C. reinhardtii. Earlier we presented the data that intrathylakoid protons accumulated in the light facilitated the bicarbonate dehydration reaction with participation of tCA (Moskvin et al. 2000). We speculate that CA that resides close to PSI can form channels in thylakoid membranes as does α-CAIV in animals (Diaz et al. 1982) and can use the protons from thylakoid lumen for bicarbonate dehydration on the stromal side of thylakoid membrane. This mechanism could supply CO2 to Rubisco situated in a multienzyme complex (Jebanathirajah and Coleman 1998) in contact with stromal thylakoid membranes (Anderson et al. 1996; Süss et al. 1995), where PSI is situated (CA3 on Fig. 6).

In this work we have discovered a high molecular mass tCA (262 kD). The presence of detergents, derifate in the top electrode buffer and DM in preparation (see Methods), should prevent multimerization of any thylakoid CA with lower molecular mass («CA2» or «CA3») and, sooner, lead to degradation of complexes. The detection of CA activity in gel with our method was possible only if CA was in native form, while disintegration to subunits, that is the degradation of quaternary structure of a protein, causes the loss of its enzymatycal activity. Besides, «CA4» differed in its properties from «CA2» and «CA3». The last two were apparently membrane-bound proteins since they (1) precipitated after centrifugation at 144000 × g, during 1 h, they presented in electrophoresis of precipitate «144» (Fig. 3F), (2) stayed on start, and did not move into the gel if the preparation were not treated with DM (Fig. 3D). «CA4» possessed the properties of soluble proteins: (1) it was present on gels from membranes both treated and non-treated with DM (Fig. 3C, D); (2) it didn’t precipitate after centrifugation at 144000 × g for 1 h. CA activity remained in the supernatant after such treatment (Fig. 3E) but was not found in the pellet (Fig. 3F); (3) CA activity of supernatant-12 did not depend on Triton concentration after «L» was released from the lumen under perforation of membrane at triton/Chl ratio of 1.0 or more (Fig. 2, curve 2).

The following facts contradict the «L» is contamination of thylakoids with high-molecular mass stromal and cytoplasmic soluble CAs: (a) the apparent molecular mass of «L» (Fig. 3C) differs from the apparent molecular masses of both stromal and cytoplasmic CAs (Fig. 3G); (b) «L» is absent in the supernatant after precipitation of thylakoids, either not treated with Triton X-100, or even after treatment with Triton at triton/Chl ratio of 0.3 (Fig. 3A, B); (c) EZ completely suppressed CA activity of supernatant-12 at the concentration of 10−7 M (Fig. 4C), that is three orders less than for the extract of soluble proteins (Fig. 4A). The speculative hypothesis is that «CA4» is the protein identical structurally to the stromal CA and it is attached to thylakoid membrane. During preparation it became inactive and under triton action at triton/Chl ratio more than 1.0, this CA activated as the result of refolding. However, even at triton/Chl ratio of 0.3 triton concentration exceeded micellization constant (Findlay and Evans 1987) and CA activity and protein band were absent after electrophoresis of supernatant after centrifugation (12000 × g, 30 min) of thylakoids incubated with triton at triton/Chl ratio of 0.3, in the position of «L» (Fig. 3B). Besides, we have found that triton did not increase the activity of soluble CAs of stroma and cytoplasm (Khristin et al. 2004).

These results testify that «L» is a protein differing from soluble CAs of cytoplasm and stroma. All well-studied soluble CAs of higher plants are β-CAs with molecular mass of 42–270 kD (Tripp et al. 2001). The molecular mass of «CA4» was detected approximately in native electrophoresis, but it was comparable with molecular mass of stromal and cytoplasmic soluble CAs (Fig. 3G). Possibly «CA4», like the other soluble CAs of dicotyledons is octamer with molecular mass of subunits of about 30-40 kD.

The function of this CA situated probably in thylakoid lumen (CA4 on Fig. 6) may be to regulate of lumenal pH.

Characteristics of thylakoid CAs found in this work and as well as of those found in the literature are collected in Table 2. We propose that three CAs are membrane-bound proteins and the forth is soluble.
Table 2

The characteristics of pea thylakoid CAs


Membrane-bound CAs

Soluble CA




CA4 (“L”)

Position in thylakoid

PSII core-complex (Khristin et al. 2004)

Near PSII at the lumenal side of thylakoid membrane (Lu and Stemler 2002)

Near PSI in stromal thylakoid membrane (Ignatova et al. 2006; Rudenko et al. 2006)

Thylakoid lumen (Rudenko et al. 2006)

Effect of Triton X-100

Maximum of activity at triton/Chl ratio of 1.0 (Khristin et al. 2004)

No detected effect

Maximum of activity at triton/Chl ratio of 0.3

No detected effect

Sulfonamide inhibitors action

High sensitivity to EZ with I50 = 10−9 M (Ignatova et al. 2006)

Stimulation of activity by AA (at 10−8—10−5 M), high sensitivity to EZ with I50 = 10−9 M (Ignatova et al. 2006)

Similar effects of AA and EZ with I50∼10−6 M (Ignatova et al. 2006)

High sensitivity to EZ, I50 ∼ 10−9 M

Apparent molecular mass


50 kD (Rudenko et al. 2006), 33 kD (Lu and Stemler, 2002; Lu et al. 2005)

20 kD (Rudenko et al. 2006)

262 kD (Rudenko et al. 2006)

Possible functions

Supplying non-haem Fe2+ of D1 protein with bicarbonate (Ignatova et al. 2006; Rudenko et al. 2006)

Supplying WOC with bicarbonate or remove the excess of protons released during WOC functioning (Villarejo et al. 2002)

CO2 supply to Rubisco in contact with thylakoid membrane

The regulation of lumenal pH to protect lumen proteins under rapid changes in light intensity.

The presence of several CA in thylakoids seems not unusual. Eighteen genes coding CAs were found in Arabidopsis thaliana genome ( Five products of these genes were found in mitochodria (Perales et al. 2004; Perales et al. 2005; Sunderhaus et al. 2006). Other two CAs have been isolated from cytoplasma and chloroplast stroma (Kachru and Anderson 1974; Rumeau et al. 1996). Thus the above data testify to the presence of four CAs in thylakoids from, at least, eleven, that are yet to be identified.


The authors express their gratitude to Dr. M.S. Khristin for valuable discussions.

Copyright information

© Springer Science+Business Media B.V. 2007