Life in the biosphere is carbon based. All molecules needed for life are made up of a carbon skeleton which is complemented by organic elements such as O, H, N, and inorganic elements such as K, P, Ca, Fe, etc. Carbon, O and H of organic compounds originate in CO2 and H2O. The carbon skeleton is assembled via the process of photosynthesis which essentially converts solar energy into chemical energy. Nitrogen originates in NH3 and inorganic elements originate in the rocks of the biosphere and are incorporated into the carbon skeleton via enzymatic reactions. Chemical energy consists of the covalent bond energy embedded into the carbon-carbon skeleton as well as the high energy bonds of ATP and NADPH which are formed during the process of photosynthesis.
The carbon cycle essentially describes how photosynthesis supports organic life in the biosphere. The carbon skeleton formed via the process of photosynthesis is converted into the simple and complex food consumed by organic life. The needed energy for enzymatic interconversions and biosynthetic processes is provided by ATP and NADPH. The organic matter of dead biota is converted in turn into CO2, H2O, and inorganic elements by bacterial activity, then the cycle repeats itself all over again.
At issue then, is whether agricultural productivity at today's levels of photosynthetic efficiency is efficient enough to feed a growing world population. Indeed the world population of about 6 billion is expected to increase to 9 billion by the year 2030 and may easily reach 18 billion by the end of the twenty first century.
Photosynthetic efficiency is controlled by intrinsic and extrinsic factors (1). Extrinsic factors include the availability of water, CO2, inorganic nutrients, ambient temperature, and the metabolic and developmental state of the plant. The most important intrinsic factor is the efficiency of the photosynthetic electron transport system (PETS). The latter is driven by two photochemical reactions that take place in membrane-bound photosystem I (PSI), and PSII chlorophyll (Chl)-protein complexes.
The rest of this chapter will therefore be devoted to a discussion of the efficiency of the PETS and possible alterations in the circuitry of the chloroplast that may lead to a higher efficiency of the PETS and higher plant productivity under field conditions.
At the maximal quantum efficiency of one, two photons are required to move one electron across the potential difference of about 1.25 V between Z+ and A0. The maximal efficiency of the photochemical reactions leading to the formation of Z+ and A0 is given by
E = 1.25 eV/ 2 h? (Eq 1)
E = efficiency of PETS
eV = Energy units in electron volts
h? = Energy of the absorbed photon in eV
Since the red 680 nm photons absorbed by PSI and PSII have an energy of 1.83 eV, it ensues from Eq 1 that
E = 1.25 eV/2*1.83 eV = 0.34 eV (Eq 2)
Therefore the absolute maximal efficiency of the PETS under red light is
(0.34 eV/1.25 eV)*100 = 27% (Eq 3)
Under natural white light, although the Chl concentration in photosynthetic membranes is high enough to result in near total absorption of all incident photosynthetically active
photons between 400 and 700 nm, under normal weather conditions, these photons represent only about 44.5 % of the total incident solar radiation. Therefore under these circumstances,
the possible overall maximal energy conversion efficiency amounts to:
(27%*44.5%)/100 = 12% (Eq 4)
Under field conditions however, the average net photosynthetic efficiency results in a net agricultural productivity in the range of 2-8 tons of dry organic matter per acre per year (1). This corresponds to a solar conversion efficiency of 0.1 to 0.4% of the total average incident radiation. Therefore the discrepancy between the 12% maximal theoretical efficiency of the PETS, and the agricultural photosynthetic efficiency observed under field conditions ranges from
(12%/0.4%)*100 = 3000%, (Eq 5) to
(12%/0.1%)*100 = 12000% (Eq 6)
C. Molecular Basis of the Discrepancy Between the Theoretical Maximal Efficiency of the PETS and the Actual Solar Conversion Efficiency of Photosynthesis Under Field Conditions
The discrepancy between the 12% theoretical maximal efficiency of the PETS and the actual 0.1-0.4% solar conversion efficiency of photosynthesis observed under field conditions can be attributed to
(a) factors extrinsic to the PETS, and
(b) to intrinsic rate limitations of the PETS (Lien and San Pietro, 1975).
Using the conventional figure of 200 light harvesting Chl molecules per reaction center (RC) per PS, i.e. for a photosynthetic unit (PSU) size of 200 per PS, under the moderate light intensities of a shady sky (about 1/10 of full sunlight), each RC would receive about 200 photons per second (s) ((Lien and San Pietro, 1975). In other words, each RC would receive about 200 hits or excitons per s. Under these conditions, in order to maintain a quantum efficiency of ONE, the slowest dark reaction of the entire PS must have a turnover rate of 200 per s (Lien and San Pietro, 1975).
Under full sunlight, which is about 10 fold higher than in the shade, the turnover rate of the limiting dark-reaction should be 200*10 = 2000 per s. This turnover rate corresponds to a rate of O2 evolution of about 9000 ?moles of O2 per mg Chl per hour (h). Yet, the maximal rate of O2 evolution observed during a Hill reaction, which results in the oxidation of H2O and the release of O2, under saturating light intensities, and other optimal conditions, rarely exceeds 5-10% of the above value. In other words, it is equal to the optimal rate of O2 evolution observed in the shade (Lien and San Pietro, 1975).
Furthermore extensive kinetic studies have demonstrated that the rate limiting steps of the PETS do not reside in the initial photochemical reactions that take place in the RC, but reside within the redox-carriers, i. e. the electron transport chains connecting PSII to PSI. The discrepancy between the capacity of the photon gathering apparatus, i. e. the antenna Chl-protein complexes and the capacity of the rate-limiting dark reactions has been named the antenna/PS Chl mismatch (Lien and San Pietro, 1975). Correction of the Antenna/PS Chl Mismatch
The first and most important effect of the antenna/PS Chl mismatch is one of reduced quantum conversion efficiencies at light intensities above shade levels. The second effect relates to the photodestructive effects of the excess photons collected by antenna Chl but not used in the initial photochemical acts. The energy of these unused photons leads to serious photodestruction of the PETS which must be repaired at a cost (Lien and San Pietro, 1975).
The single-branched Chl biosynthetic pathway is depicted in Fig. 2. It consists of a linear sequence of biochemical reactions which convert divinyl (DV) protoporphyrin IX (Proto) to monovinyl (MV) Chl a via DV Mg-Proto, DV Mg-Proto monomethyl ester (Mpe), DV protochlorophyllide a (Pchlide a), MV Pchlide a, and MV Chlorophyllide a (Chlide a). The salient features of this pathway are (a) the assumption that DV Pchlide a does not accumulate in higher plants, but is a transient metabolite which is rapidly converted to MV Chl a via MV Pchlide a, and (b) that the formation and accumulation of MV tetrapyrroles between Proto and Mpe and DV tetrapyrroles between Pchlide a and Chl a does not take place (Rebeiz et al, 1994). All in all, experimental evidence gathered over the past 26 years indicates that only a small fraction of the total Chl of green plants is formed via this pathway (Rebeiz et al, 1994, 2002)
Since the 1963 seminal review of Smith and French (Smith and French, 1963), our understanding of the Chl biosynthetic pathway has changed dramatically. Several factors have contributed to this phenomenon, among which: (a) development of systems capable of Chl and thylakoid membrane biosynthesis in organello and in vitro, (Rebeiz and Castelfranco, 1971a, 1971b, Daniell and Rebeiz, 1982a, 1982b, Rebeiz et al, 1984, Kolossov et al, 1999), (b) powerful analytical techniques that allowed the qualitative and quantitative determination of various intermediates of the pathway (Rebeiz, 2002), (c) recognition that the greening process proceeds differently in etiolated and green tissues, in darkness and in the light, ((Carey et al, 1985), and in plants belonging to different greening groups; Ioannides et al, 1994; Abd-El-Magid et al, 1997), and (d) recognition of the probability that the structural and functional complexity of thylakoid membranes is rooted in a multibranched, heterogeneous Chl biosynthetic pathway (Rebeiz et al, 1999).
It was proposed that Chlorophyll biosynthetic heterogeneity referred either (a) to spatial biosynthetic heterogeneity, (b) to chemical biosynthetic heterogeneity, or (c) to a combination of spatial and chemical biosynthetic heterogeneities (Rebeiz et al, 2002). Spatial biosynthetic heterogeneity was defined as the biosynthesis of an anabolic tetrapyrrole or end product by identical sets of enzymes, at several different locations of the thylakoid membrane. On the other hand, chemical biosynthetic heterogeneity was defined as the biosynthesis of an anabolic tetrapyrrole or end product at several different locations of the thylakoid membrane, via different biosynthetic routes, each involving at least one different enzyme. Figure 3 organizes all known carboxylic Chl biosynthetic reactions into a logical scheme made up of 15 different biosynthetic routes.
Fig. 3. Integrated Chl a/b biosynthetic pathway depicting 15 carboxylic routes. To facilitate understanding of the text, various biosynthetic routes are designated by the numbers 1-15. Reproduced from Rebeiz et al, 2002.
Fig. 4 . Schematics of a linear model of a PSU in the unfolded state Reproduced from reference (Bassi et al, 1990).
The SBP-single location model is depicted schematically below, in Fig. 5. Within the PSU, this model accommodates only one Chl-apoprotein biosynthesis center and no Chl-apoprotein biosynthesis subcenters. Within the Chl-apoprotein biosynthesis center, Chl a and b are formed via a single-branched Chl
biosynthetic pathway (Fig. 2) at a location accessible to all Chl-binding apoproteins. The latter will have to access that locationin the unfolded state, pick up a complement of MV Chl a and/or MV Chl b, and undergo appropriate folding. Then the folded Chl-apoprotein complex has to move from the central location to a specific PSI, PSII, or Chl a/b LHC-protein location within the Chl-apoprotein biosynthesis center over distances of up to 225 Å (Kolossov et al, 2003, Kopetz et al, 2004). In this model, it is unlikely to observe resonance energy transfer between metabolic tetrapyrroles and some of the Chl-apoprotein complexes located at distances longer than 100 Å. This is because resonance excitation energy transfer takes place only over distances shorter than 100 Å (Calvert and Pitts, 1967).
The SBP-Multilocation model is depicted schematically below, in Fig. 6. In this model, every biosynthetic location within the PSU is considered to be a Chl-apoprotein thylakoid biosynthesis center. In every Chl-apoprotein biosynthesis location, a complete single-branched Chl a/b biosynthetic pathway (Fig. 2) is active. Association of Chl a and/or Chl b with specific PSI, PSII, or LHC apoproteins at any location is random. In every Chl-apoprotein biosynthesis center, distances separating metabolic tetrapyrroles from the Chl-protein complexes are shorter than in the SBP-single-location model. Because of the shorter distances separating the accumulated tetrapyrroles from Chl-protein complexes, resonance excitation energy transfer between various tetrapyrroles and Chl-apoprotein complexes within each center may be observed. However, formation of MV Mg-Proto (Mp) and its MV methyl ester (Mpe) [i. e. Mp(e)] is not observed in any pigment-protein complex. This is because the single-branched Chl biosynthetic pathway does not account for MV Mp(e) biosynthesis.
Resonance excitation energy transfer from three tetrapyrrole donors to the Chl a of various Chl-protein complexes were monitored, namely: from Proto, Mp(e) and MV and DV Pchlide a. DV Proto is a common precursor of heme and Chl. It is the immediate precursor of DV Mg-Proto. As such, it is an early intermediate along the Chl biosynthetic chain. Biosynthetically, it is several steps removed from the Chl end product (Rebeiz et al, 2003). Mg-Proto is a mixed MV-DV, dicarboxylic tetrapyrrole pool, consisting of DV and MV Mg-Proto (Rebeiz et al, 2003). It is the precursor of DV and MV Pchlide a. The [(Pchl(ide)] of higher plants consists of about 95% Pchlide a and about 5% Pchlide a ester . The latter is esterified with long chain fatty alcohols at position 7 of the macrocycle. While Pchlide a ester consists mainly of MV Pchlide a ester, Pchlide a consists of DV and MV Pchlide a. The latter are the immediate precursors of DV and MV Chlide a(Rebeiz et al, 2003).
Accumulation of the various tetrapyrrole donors was induced by incubation of green tissues with d-aminolevulinic acid (ALA) and/or 2,2’-dipyridyl (Rebeiz et al, 1988). The task of selecting appropriate Chl a-protein acceptors was facilitated by the fluorescence properties of green plastids. At 77 °K, emission spectra of isolated chloroplasts exhibit maxima at 683-686 nm (~F685), 693-696 nm (~F695), and 735-740 nm (~F735). It is believed that the fluorescence emitted at ~F685 nm arises from the Chl a of LHCII, the major thylakoid LHC antenna, and LHCI-680, one of the LHC antennae of PSI (Bassi et al, 1990). That emitted at ~F695 nm is believed to originate mainly from the Chl a of CP47 and CP29, two PSII antennae (Bassi et al, 1990). That emitted at ~F735 nm is believed to originate primarily from the Chl a of LHCI-730, a PSI antenna (Bassi et al, 1990). Since these emission maxima are readily observed in the fluorescence emission spectra of green tissues and are associated with definite thylakoid Chl a-protein complexes, it was conjectured that they would constitute a meaningful resource for monitoring excitation resonance energy transfer between anabolic tetrapyrroles and representative Chl a-protein complexes.
To monitor the possible occurrence of resonance excitation energy transfer between the accumulated anabolic tetrapyrroles and Chl a-protein complexes, excitation spectra were recorded at 77°K at the respective emission maxima of the selected Chl a acceptors, namely at ~685, ~695, and ~735nm. It was conjectured that if resonance excitation energy transfers were to be observed between the tetrapyrrole donors and the selected Chl a acceptors, definite excitation maxima would be observed. These excitation maxima would correspond to absorbance maxima of the various tetrapyrrole donors, and would correspond to the peaks of the resonance excitation energy transfer bands.
The compatibility of the SBP-single location model with the formation of Chl a-thylakoid proteins was tested by monitoring resonance excitation energy transfer between anabolic tetrapyrrole intermediates of the Chl biosynthetic pathway and various thylakoid Chl a-protein complexes. Pronounced resonance excitation energy transfer bands from Proto, Mp(e), and Pchl(ide) a to Chl a ~F685, ~F695, and ~F735 were detected (Table 1).
Assignment of in situ resonance excitation energy transfer maxima to various metabolic tetrapyrroles was unambiguous except for a few cases at the short wavelength and long wavelength extremes of excitation bands. Contrary to previous beliefs, it was surprising to observe a significant diversity in the various intra-membrane environments of Proto, Mp(e), and Pchl(ide) a (Kolossov et al,2003 . This diversity was manifested by a differential donation of resonance excitation energy transfer from multiple Proto, Mp(e) and Pchl(ide) a sites to the different Chl a-apoprotein complexes. This in turn, is strongly compatible with the biosynthetic heterogeneity of the Chl biosynthetic pathway.
Since resonance excitation energy transfer is insignificant at distances larger than 100 Å (Calvert and Pitts, 1967), the detection of pronounced resonance excitation energy transfer from Proto, Mp(e), and Pchl(ide) a to Chl a ~F685, ~F695, and ~F735 (Table 1), indicated that these anabolic tetrapyrroles were within distances of 100 Å or less of the Chl a acceptors. This in turn was incompatible with the functionality of the SBP-single location Chl-thylakoid biogenesis model as detailed below.
The early concept of a PSU consisting of about 500 antenna Chl per reaction center has evolved into two pigment systems each with its own reaction center and antenna Chl (Allen and Forsberg, 2001; Anderson, 2002; Staehelin, 2003). The early visualization of the two photosystems consisted of various pigment-protein complexes arrayed into a linear PSU (the continuous array model), about 450 Å in length and 130 Å in width (Bassi et al, 1990 ). In the PSU, the LHCII was depicted as being shared between the two photosystems (Bassi et al, 1990). More recent models however, favor the concept of a laterally heterogeneous PSU (Allen and Forsberg, 2001; Anderson, 2002; Staehelin, 2003). In this model LHCII shuttles between PSI and PSII upon phosphorylation and dephosphorylation (Allen and Forsberg, 2001). Furthermore while PSII is mainly (but not exclusively) located in appressed thylakoid domains, PSI is located in non-appressed stroma thylakoids, grana margins, and end membranes (Anderson, 2002, Staehelin, 2003).
The SBP-single location model is incompatible with the linear continuous array model, and the laterally heterogeneous PSU model. Indeed the SBP-single location model calls for Chl a and b to be formed via a single-branched Chl biosynthetic pathway at a location accessible to all Chl-binding apoproteins. The latter will have to access that location in the unfolded state, pick up a complement of MV Chl a and/or MV Chl b, and undergo appropriate folding. Then the folded Chl-apoprotein complex has to move from the central location to a specific PSI, PSII, or Chl a/b LHC-protein location within the Chl-apoprotein biosynthesis center over distances of up to 225 Å in the linear continuous array model, or over larger distances, in the laterally heterogeneous model, to become part of PSI, PSII or LHCII. If this were the case, then no resonance excitation energy transfer would be observed between anabolic tetrapyrroles and the various Chl-protein complexes, and the distances separating the anabolic tetrapyrroles from the various Chl-protein complexes would be much larger than the values reported in Table 1.
The shorter distances separating anabolic tetrapyrroles from Chl-protein complexes reported in Table 1 are compatible with the SBP-multilocation and MBP-sublocation models. However, overwhelming experimental evidence argues against the operation of a single-branched Chl biosynthetic pathway in plants (Rebeiz et al, 2003).
The above consideratiopns leaves the MBP-sublocation model (MBPSUBLM) as a viable working hypothesis. In this model (Rebeiz et al, 2003), the unified multibranched Chl a/b biosynthetic pathway, is visualized as the template of a Chl-protein biosynthesis center where the assembly of PSI, PSII and LHC takes place (Rebeiz et al, 1999, 2004). The multiple Chl biosynthetic routes are visualized, individually or in groups of one or several adjacent routes, as Chl-apoprotein biosynthesis subcenters earmarked for the coordinated assembly of individual Chl-apoprotein complexes. Apoproteins destined to some of the subcenters may possess specific polypeptide signals for specific Chl biosynthetic enzymes peculiar to that subcenter, such as 4-vinyl reductases, formyl synthetases or Chl a and Chl b synthetases. Once an apoprotein formed in the cytoplasm or in the plastid reaches its subcenter destination and its signal is split off, it binds nascent Chl formed via one or more biosynthetic routes, as well as carotenoids. During pigment binding, the apoprotein folds properly and acts at that location, while folding or after folding, as a template for the assembly of other pigment-proteins. This model is compatible with the lateral heterogeneity of the PSU and can account for the observed resonance excitation energy transfers (Table 1) and the short distances separating anabolic tetrapyrroles from Chl-protein complexes in the distinct PSI, PSII and shuttling LHCII entities that compose the PSU (Table 2).
In our opinion the ultimate agriculture of the future should consist of bioreactors populated with bioengineered, highly efficient photosynthetic membranes, with a small PSU size and operating at efficiencies that approach the 12% maximal theoretical efficiency of the PETS that may be observed under white light, or the 27% maximal theoretical efficiency that may be achieved under red light. Such conditions may be set up during space travel, in large space stations, or in human colonies established on the moon or on Mars(Rebeiz et al, 1982). The photosynthetic product may well be a short chain carbohydrate such as glycerol that can be converted into food fiber and energy.
In the meanwhile, let us not forget that a journey of 10,000 miles starts with the first step.