Introduction

Carbonic anhydrase (CA; EC: 4.2.1.1) is a zinc-containing enzyme that catalyzes the reversible hydration of CO2 to HCO3 − and has been found in the three domains of life: archaea, bacteria and eukarya. So far, five distinct, evolutionary unrelated CA classes, have been reported namely α-, β-, γ-, δ- and ζ-CA, which independently evolved similar catalytic mechanisms [1, 2]. The β-CA can be further subdivided into two main subclasses, the plant type and the cab-type class, named after the β-CA CAB from the archeon Methanobacterium thermoautotropicum [3]. A third sub-class of β-CAs, designated as ε-class was found in several marine cyanobacteria and chemolithoautotrophic bacteria [4, 5]. The δ- and ζ-CA classes are so far restricted to marine diatoms [6].

CA has been found in a broad range of organisms. It participates in photosynthesis and respiration through the carboxylation and decarboxylation reactions. Several other roles have been attributed to this enzyme such as pH regulation [7], ion transport, and water and electrolyte balance [8].

In plants, all known CAs belong to the α, β or γ classes with the β class being the most predominant. Ιn the chloroplast stromal compartment, HCO3 – is reported to be present at almost 100 times the concentration of CO2 [9]. Hence, in photosynthetic tissues, plastidial β-CA has the major role of converting bicarbonate into CO2 which is the substrate for Rubisco [10]. On the other hand, cytosolic β-CA could provide bicarbonate for phosphoenolpyruvate carboxylase (PEPC) either for C4 photosynthesis [11, 12] or for anaplerotic CO2 fixation in C3 plants. CAs in anaplerotic assimilation catalysed by phosphoenolpyruvate carboxylase have also been proposed [13]. Other biochemical roles, in relation to β-CA, are lipogenesis, glyconeogenesis and pyrimidine biosynthesis [14].

cDNAs encoding β-CA isoforms have been isolated and characterized so far from a number of species including Pisum sativum, [15, 16], soybean [17], alfalfa [18], Lotus japonicus [19], tobacco [20], Arabidopsis [21–23], cotton [24], Flaveria bidentis [25], Hordeum vulgare [26], Oryza sativa [27], Zea mays [28]. CA activity has also been reported in Vicia faba and Lupinus angustifolius [29]. Furthermore, the expression of several β-CA isoforms has been reported in Arabidopsis [23], F. bidentis [30] and F. pringlei [31]. Biochemical evidence has also suggested the multiplicity of β-CA isoforms in pea chloroplast [32]. Subsequent analysis of Arabidopsis thaliana β-CA1 indicated that in addition to its localization to the chloroplast it is also found close to the plasma membrane [33].

During development, tomato fruit passes through distinct stages which range from immature green to ripe red. The tomato fruit epidermis has no stomata [34] and bears a relatively thick cuticle. Carbon dioxide produced by fruit respiration could be re-fixed inside the fruit. At the mature green stage, the CO2 concentration inside the fruit was measured at about 0.29 % (i.e. a ninefold increase compared to the external ambient CO2 concentration) [35]. If any CO2 re-fixation in tomato fruits occurs, carbonic anhydrase could be the key enzyme in this process.

In this study, the presence of CA and PEPC activity and the distribution of CA and PEPC proteins were examined in tomato fruit. Furthermore, the isolation and characterization of three cDNAs encoding CA isoforms in tomato is reported. The SlCA1 and SlCA3 genes represent two full-length cDNAs encoding cytosolic β-CAs. The SlCA2 gene corresponds to a full-length cDNA encoding a plastidic β-CA. The expression pattern of the three β-CA genes was investigated, in parallel with that of the already characterized PEPC gene expression pattern, (SlPEPC1; SlPEPC2) in the fruit and other tissues of tomato plant [36].

Materials and methods

Plants of S. lycopersicum (var. Dombito) were grown in the Agricultural University of Athens greenhouse according to standard agricultural practices. Samples were taken from immature green fruit (10 mm) and mature green fruit (55 mm) which was separated into locular parenchyma with seeds, pericarp and placenta. Samples were also taken from leaves, petioles, flowers (mixed preanthesis and anthesis tissues) and roots. Tissues were cut, frozen immediately with liquid nitrogen and stored at −80 °C.

Measurement of CA and PEPC activity in tomato tissues

CA activity was assayed by measuring changes in pH during the reaction using a dye indicator method [37]. Tomato tissues were ground in liquid nitrogen and homogenized in 2.5 volumes of extraction buffer containing 50 mM MES pH 6.3, 1 mM EDTA, 0.1 mM phenylmethane-sulfonylfluoride, and 0.1 % (w/v) polyvinylpolypyrrolidone (PVPP). Supernatants were dialyzed overnight at 4 °C against 50 mM MES pH 6.3, 1 mM ZnCl2. Assays were performed at 4 °C. The buffer/indicator reaction solution used was 50 mM MES containing 0.005 % (w/v) bromothymol blue (pH 6.3, A615). The reaction buffer contained 50 mM sodium sulfate to maintain a relatively constant ionic strength in the reaction medium. The reaction was initiated by the addition of NaHCO3 at final concentrations of 25–100 mM. Measurements were taken at 615 nm using a Hitachi U-2800 Spectrophotometer (Tokyo, Japan). Uncatalyzed rates, measured in the absence of the enzyme, were subtracted from the catalyzed rates. The catalyzed rate was converted to equivalent μmol H+ consumed, after a calibration established by titrating the reaction buffer with NaOH. For the PEPC activity assay, 2 g of powdered tomato tissues were extracted with 5 ml extraction medium according to Guillet et al. [36]. The protein concentration was determined according to Bradford [38].

Protein extraction, SDS-PAGE and immunoblot analysis

Approximately 0.5 g of fruit tissues were ground in a mortar containing liquid nitrogen and the powder was mixed with 800 μl of ice-cold 50 mM TRIS–HCl (pH 6.8), 5 % (w/v) SDS, 10 % (v/v) glycerol, 5 % (v/v) 2-mercaptoethanol, and 0.002 % (w/v) bromophenol blue. The extract was centrifuged at 12,000×g for 5 min and the supernatant was used for SDS-PAGE. Approximately 20 μg of protein was loaded onto each track of the gel for detecting CA and PEPC proteins. These were analyzed in 15 % SDS-PAGE for CA and in 10 % for PEPC and were then blotted onto nitrocellulose filter (Nitrocell, Hoefer). Polyclonal antibodies were raised in rabbits against the overexpressed recombinant GmCA and GmPEPC proteins [17]. The blots were probed with the polyclonal antibodies and bound antibodies were visualized with the ProtoBlot Western Blot AP System (Promega).

Immunohistolocalization

Polyclonal antibodies raised against soybean β-CA and PEPC polypeptides [17] were used for immunolocalization of tomato CA and PEPC. Fruit tissues at the mature green stage were fixed in 4 % (v/v) paraformaldehyde and 0.25 % (v/v) glutaraldehyde in 10 mM phosphate buffer (pH 7) at 4 °C, overnight. Sections of 8-10 μm, were blocked for 1 h in TBS (10 mM Tris–HCl, 100 mM NaCl, pH 8) containing 3 % (w/v) BSA, and 0.05 % (w/v) Tween-20 at room temperature and incubated overnight with an antiserum raised against either the soybean β-CA or PEPC polypeptides. Pre-immune serum and immune serum were used at a dilution of 1:1,500. Alkaline phosphatase-conjugated anti-rabbit IgGs (Promega) [diluted to 1:5,000 in TBS containing 1 % (w/v) BSA, and 0.05 % (v/v) Tween-20] were used as secondary antibodies. Signal was detected using the alkaline phosphatase substrate, BCIP/NBT (Promega). No signal was observed when pre-immune serum was used as a negative control.

Identification and characterization of cDNAs encoding tomato β-CA gene family

Searches within public tomato EST databases revealed the presence of a number of ESTs in S. lycopersicum coding for β-CAs. The corresponding ESTs were aligned and the complete nucleotide sequence of the larger cDNA clone was determined. The deduced amino acid sequences revealed the presence of SlCA cDNAs representing three full-length open reading frames (GenBank accession nos. CAH60890, CAH60891 and CAQ30514 corresponding to SlCA1, SlCA2 and SlCA3 polypeptides, respectively).

Determination of transcripts levels using real-time qRT-PCR

Organs from S. lycopersicum plants, including locular parenchyma, pericarp and placenta, leaves, young stems, mixed flowers (anthesis and preanthesis), roots and small fruit (10 mm in diameter), were harvested and ground in liquid nitrogen. Total RNA was extracted according the method described by Brusslan and Tobin [39]. RNA samples were then treated with DNase I (Promega, Madison, WI) at 37 °C for 45 min to eliminate contaminating genomic DNA.

First-strand cDNA was reverse transcribed from 2 μg of DNase-treated total RNA. All DNase-treated total RNA samples were denatured at 65 °C for 5 min followed by quick chill on ice in a 12 μl reaction mixture containing 500 ng oligo(dT)12–18 mer, and 0.833 mM dNTPs. After the addition of 4 μl of 5× First-Strand buffer (Invitrogen, Paisley, UK), 1 μl (40 U) RNaseOUT (Invitrogen) ribonuclease inhibitor and 0.016 M DTT at final concentrations, the reaction mixture was preheated to 42 °C for 2 min before the addition of 1 μl (200 U) of SuperScript II reverse transcriptase (Invitrogen). The reaction mixture was incubated at 42 °C for 50 min, followed by heat inactivation at 70 °C for 15 min. The resulting first-strand cDNA was diluted to a final volume of 200 μl, and target cDNAs were amplified using gene-specific primers (Table 1) designed from the transcribed region of each gene using Primer Express 1.5 software (Applied Biosystems, Darmstadt, DE). Quantitative RT-PCRs were performed on the ABI PRISM 7900HT Sequence Detection System using SYBR Green master mix (Applied Biosystems), gene specific primers at a final concentration of 0.2 μM each and 1 μl of the cDNA as template. PCR cycling started with the initial polymerase activation at 95 °C for 15 s and 60 °C for 1 min [40]. The primer specificity and the formation of primer-s were monitored by dissociation curve analysis and agarose gel electrophoresis on a 4 % (w/v) gel. The expression levels of a S. lycopersicum actin gene, detected using SlACTIN-F 5′-CCACAACCACCTCAGCAGAAT-3′ and SlACTIN-R 5′-GACAACTGCTGTGGAGGCATT-3′ primers were used as internal standards to normalize small differences in cDNA template concentrations. Real-time RT-PCR was performed using an MxPro MX3005P PCR system and analyzed with MxPro v3.0 software (Stratagene). For the relative quantification of gene expression, a modification of the comparative threshold cycle method was used. In the case of different tomato tissues, relative transcript levels of the gene of interest (X) were calculated as a ratio to the expressed actin gene transcripts (U), as (1 + E)−ΔCt, where ΔCt was calculated as (C xt  − C ut ) [41]. PCR efficiency (E) for each amplicon was calculated using the linear regression method, calculated with the LinRegPCR software [42]. All real-time qPCRs were performed on three biological repeats and all measurements were taken in duplicate.

Table 1 Primers used for real time PCR
Full size table

Results

CA and PEPC activity in tomato fruit

The CA activity in tomato fruit tissues (pericarp, locular parenchyma and placenta) was determined spectrophotometrically using a modified version of the pH-indicator method [37]. CA activity was found to be predominant in the locular parenchyma measuring at about 0.69 mmol H+ consumed per min and per mg of protein and at similar levels in the pericarp (0.57 mmol H+ min−1 mg−1 protein) (Fig. 1a). In contrast, CA activity reached at 0.14 mmol H+ min−1 mg−1 protein in the placenta which is four times lower than the locular parenchyma and pericarp, respectively. PEPC activity was found to be 0.58 μΜ NADH/min mg protein in the pericarp. This value was twice as high as the activity in the locular parenchyma (0.26 μM NADH min−1 mg−1 protein) (Fig. 1b). In the placenta, however, an average activity of 0.38 μM NADH/min mg protein was observed.

Fig. 1
figure 1

Carbonic anhydrase activity (a) in locular parenchyma, pericarp and placenta. The activity was determined spectrophotometrically using a pH-indicator method at pH 6.3. The activity is expressed in mmole of H+ consumed per min and per mg of protein. The PEPC activity (b) was assayed according to [36]. Bars show mean activity +SD (n = 3)

Full size image

Immunohistological localization and tissue distribution of CA and PEPC proteins in tomato mature green fruit

To gain further insight into the localization of CA and PEPC proteins in the fruit tissues an immunohistolocalization approach was employed. Paraffin-embedded sections of mature green fruit including pericarp, locular parenchyma with seeds and placenta were incubated with polyclonal antibodies raised against soybean CA or PEPC recombinant proteins. The specificity of the soybean antibodies was tested against S. lycopersicum mature green fruit proteins. In each western blot polyclonal antibodies raised against soybean PEPC or β-CA recombinant proteins detected one band with a molecular mass of 110 kDa and one with a molecular mass of 24 kDa (Suppl. Fig. 1).

As shown in Fig. 2, β-CA and PEPC proteins were present in all fruit tissues examined with the signal (blue precipitate) being more intense in the seeds. In the developing seed, the accumulation of both proteins is observed in all seed parts, endosperm, integuments and embryo proper (Fig. 2c, g). Moreover, high signal was also observed in vascular bundles (Fig. 2h). Pre-immune serum was used as a negative control (Fig. 2d, insert). In this case no signal above background was detected.

Fig. 2
figure 2

Immunohistolocalization of CA (a–d) and PEPC (e–h) in mature green fruit tissues. Sections (8–10 μm) of S. lycopersicum probed with the antiserum raised against either the GmCA1 or the GmPEPC proteins. The proteins were visualized as a dark blue staining. a, b, e, f Fruit sections containing pericarp (Pe) locular parenchyma (Lp), placenta (Pl) and developing seed with integuments (In) cotyledons (Co) and embryonic root (Ro). c, g Detail of developing seed. Integuments (In) and endosperm (En). d, h Vascular bundle (Vb) from pericarp. Insert control treated with preimmune serum

Full size image

Identification and characterization of cDNAs encoding tomato β-CA gene family

cDNA clones coding for tomato CA were isolated and characterized in our laboratory. The corresponding ESTs were aligned and the complete nucleotide sequence of the larger cDNA clone was determined. The deduced amino acid sequences revealed the presence of SlCA cDNAs representing three full-length open reading frames One codes for a polypeptide designated as SlCA1, which contains an open reading frame (ORF) of 268 amino acids. The second designated as SlCA2, contains an ORF of 321 amino acids. The third designated SlCA3 contains an ORF of 255 aminoacids. Topology prediction indicated localization in the cytoplasm for both SlCA1 and SlCA3 and localization in the chloroplast for SlCA2. The putative chloroplast targeting peptide of the SlCA2 exhibited high homology to chloroplast targeting peptides present in chloroplastic β-CAs from different plant species.

Comparative analysis of the deduced SlCA1, SlCA2 and SlCA3 amino acid sequences with previously characterized CA polypeptides (Suppl. Fig. 2) revealed that the three isoforms exhibit high similarity with other β-CAs. The phylogenetic relationship of SlCA1, SlCA2 and SlCA3 to other β-CAs was investigated by the construction of a dendrogram based on the respective amino acid sequences (Fig. 3).

Fig. 3
figure 3

Phylogenetic relationship of SlCA1, SlCA2 and SlCA3 aminoacid sequences (underlined) to other β-type CAs. The dendrogram is a graphical representation of a multiple sequence alignment constructed with the Clustal method using the PAM 250 residue weight table. N. tabacum (P27141); A. thaliana (CA1 AT3G01500, CA2 AT5G14740, CA3 AT1G23730, CA4 AT1G70410, CA5 AT4G33580, CA6 AT1G58180); F. bidentis (CA1 AAA86939, CA2 AAO17573, CA3 AY167113); F. pringlei (CA1 AAA86992, CA2 ABC41657, CA3 ABC41658), Z. mays (CA1 B4G0V7, CA2 B6T9W3, CA3 Q41729)

Full size image

The updated genome sequencing of tomato revealed that SlCA1 gene is located on chromosome 5 and its position is 5 cM while SlCA2 gene is located on chromosome 2 and its position is 93.00 cM in the Tomato-Expen 2000 Map. On the other hand, SlCA3 gene is located on chromosome 3 at 10.37 cM as annotated to Arabidopsis COSII Map (http://solgenomics.net/).

Expression of S. lycopersicum SlCA1, SlCA2, SlCA3 and SlPEPC1, SlPEPC2 in various tissues

Having established the presence of CA and PEPC proteins in the different fruit tissues of tomato, we then turned our attention towards assessing the CA and PEPC transcript levels in these tissues by performing real time quantitative PCR. Total RNA was isolated from immature green fruit (10 mm), locular parenchyma with seeds, pericarp and placenta of mature green fruit as well as from the leaves, flowers, stems and roots. Figure 4a shows the relative abundance of SlCA1, SlCA2 and SlCA3 transcripts in the various tissues and organs. Accumulation of SlCA1 gene transcript was detected in all the fruit tissues (especially in locular parenchyma), flowers and stems. On the contrary, SlCA3 and SlCA2 gene transcripts were not detected in any fruit tissues (or they were detected at very low levels). SlCA2 gene transcript was present in the leaf, stem and flower while being most abundant in the leaf tissue. SlCA3 transcript was present at significant levels in the flowers and roots. Interestingly, SlPEPC1 and SlPEPC2 gene transcripts were detected in all tissues and organs (Fig. 4b). The SlPEPC1 mRNA transcripts were 10 times higher in the fruit (10 mm) than in the leaf. SlPEPC2 mRNA transcripts were highly abundant in the fruit tissues and particularly in the locular parenchyma with seeds, the pericarp, the placenta and the fruit (10 mm) (Fig. 4b). Finally, SlPEPC2 mRNA transcript abundance reached up to 80 times that of SlPEPC1 mRNAs in the placenta (Fig. 4b).

Fig. 4
figure 4

Accumulation of SlCA1, SlCA2, SlCA3 (a) and SlPEPC1, SlPEPC2 (b) transcripts in various tissues of tomato plant. Total RNA was isolated from various organs was subjected to reverse-transcription and followed real-time quantitative polymerase chain reaction analysis. Transcript levels in the different samples were normalized to those of the constitutive gene, actin. Relative mRNA levels were calculated with respect to the levels of actin transcripts. Bars show mean +SD (n = 3)

Full size image

Discussion

As tomato fruit has a relatively thick cuticle and no stomata, gas exchange should be extremely low. Carbon dioxide concentration inside mature green fruit was measured to be, on average, ninefold higher than that of the ambient air [35]. However CO2 concentration in specific tissues, like the seed tissues, should be even higher, given the high rates of respiration in developing embryos [43]. The high concentration of CO2 inside the fruit raises the question of its fate. It could be transported out of fruit, reused in the fruit [44] or both. A mechanism that could participate in CO2 re-fixation would involve the activity of a CA enzyme. A possible role for cytoplasmic β-CAs could be the conversion of CO2 to bicarbonate, which serves as a substrate for PEPC. The produced oxaloacetate through the action of PEPC could be used for amino acid synthesis [45] or converted to malate and/or citrate which are known to accumulate in tomato fruits during its development [36].

Our studies revealed CA and PEPC enzyme activity in all the fruit tissues examined. Furthermore, the immunohistolocalization studies indicated that both the β-CA and PEPC proteins were co-localized in all the fruit tissues examined with the signal being more intense in the vascular bundles and developing seeds (Fig. 2). Similar results have been observed in Medicago seeds [46]. High local CO2 concentration is expected to occur in the vascular bundles due to the high metabolic activity [47, 48]. The seed coat surrounding the developing embryo is a major barrier for gas diffusion [49]. Thus, the presence of high levels of enzymes involved in CO2 fixation in the seeds and the vascular bundles may contribute to the efficient removal of CO2 and its conversion to oxaloacetate, a precursor for amino acid synthesis.

The observed high levels of SlCA1 transcript accumulation in the tomato pericarp are in close agreement with the reported accumulation of the protein encoded by the SlCA1 gene in cherry tomato pericarp [50]. These data taken together suggested that SlCA1 may provide the bicarbonate as a substrate for SlPEPC2, which is highly expressed, thus increasing the malic acid content, reaching up to 35 μmol g−1 FW [51]. The detection of SlCA3 transcripts exclusively in the roots and flowers, suggests that this isoform may serve a special function in these tissues. Recent observations reported the presence of CA activity in tomato roots [52]. The identification of root CA isoforms is not surprising given that recent observations also reported the expression of three CA isoforms in roots of F. bidentis [31] and Arabidopsis [24]. However, these isoforms are also found in other tissues besides the roots. The accumulation of SlPEPC1 and SlPEPC2 along with SlCA1 and SlCA3 transcripts in tomato roots suggest that some of the respired CO2 may be re-fixed through a mechanism similar to that proposed in tomato fruits.

In the plastids, CA could participate in CO2 re-fixation through photosynthesis or lipogenesis [14]. Plastidial CA has been reported to facilitate the diffusion of CO2 across the chloroplast envelope [10, 53]. In this respect, the very low or undetectable levels of SlCA2 transcript in the fruit tissues, is surprising. However, the possibility of a second, not yet identified, plastidial isoform present in the fruit tissues can not be excluded. The expression of two plastidial isoforms has been detected in the leaves, siliques, stems and flowers of Arabidopsis [23].

The presence of two PEPC isoforms expressed in tomato fruit has also been reported [36]. PEPC also appears to play a role in the extension of cotton fibres and it is proposed that PEPC activity allows malate production and therefore increased turgor that is required for fibre elongation [54]. The relative levels of SlPEPC1 transcript was similar in the different fruit tissues examined, whereas the relative levels of SlPEPC2 transcripts were very high compared to those found for SlPEPC1 (Fig. 4b). It has been suggested that the SlPEPC1 isoform could play a housekeeping role [36] in tomato fruit, and SlPEPC2 could serve in the expansion of the fruit. The high PEPC activity ratio (20–40 fold) between the fruit and leaf [51] is in accordance with our qRT-PCR data (Fig. 4b) rendering PEPC a key enzyme in the CO2 metabolism of the tomato fruit.

In conclusion, we have provided data which point to the combined action of CA and PEPC in re-fixing the respired CO2 thus making a net contribution to fruit carbon economy.