n-Alkane ¹³C/¹²C indicates differential metabolic controls of fatty lipid chain extension in C₃ and C₄ grasses Free

Dear Editor,

n-Alkanes, the unfunctionalized component of leaf waxes, are indispensable in protecting land plants from desiccation, UV radiation, pathogens, and insect attack (Jetter and Kunst 2008). Although the fatty acids (FAs) biosynthetic pathway was elucidated only in bacteria (by Konrad Bloch, Feodor Lynen and colleagues), the pathway knowledge was “extrapolated” to land plants and algae. The current consensus on land plants is that (i) the immediate precursor for FA biosynthesis is pyruvate-derived acetyl-CoA, (ii) the chain growth is effected by the continual (stepwise) addition of a C₂ unit (acetyl-CoA, derived from malonyl-CoA), and (iii) n-alkanes are formed by direct decarboxylation from FAs or indirect decarbonylation of fatty aldehydes formed by FA reduction (Jetter et al. 2006). However, the metabolic controls on FA chain elongation remain poorly understood, in addition to the metabolic branching into aldehydes, alkanols, and alkanes (Kunst et al. 2006), and if such metabolic controls differ between C₃ and C₄ land plants, given the sharp differences in their CO₂ fixation biochemistries. Indeed, pathway-specific lipid contents differ between the two distinct photosynthetic modes (Zhou et al. 2015).

Natural abundance carbon isotope composition [δ¹³C, in units of mUr equivalent to ‰ = (R_sample−R_V-PDB)/R_V-PDB, where R_sample and R_V-PDB are the ¹³C/¹²C ratios for the sample and international standard V-PDB (Vienna Pee Dee Belemnite), respectively] is particularly useful for elucidating/confirming biosynthetic pathways and gauging fluxes at metabolic branch points (Monson and Hayes 1980; Schmidt and Kexel 1997; Gleixner et al. 1998; Zhou et al. 2016; Zhu et al. 2020) as the biochemical reactions involved in lipid biosynthesis are accompanied by isotope effects (IEs). The split of biochemical labor between mesophyll and bundle sheath cells in C₄ photosynthesis distinguishes itself isotopically from C₃ photosynthesis. For example, there are clear demarcations between bulk organic matter δ¹³C of C₃ (−20 to −35 mUr, averaging −26 mUr) and C₄ (−10 to −14 mUr, averaging −12 mUr) (Farquhar et al. 1989; Brugnoli and Farquhar 2000). Such distinction is also seen in lipids (Chikaraishi and Naraoka 2007; Zhou et al. 2010), and is tied to the physiological roles of lipids (Zhou et al. 2015).

To identify the possible differences in metabolic controls on lipid biosynthesis in C₃ and C₄ plants, we obtained the δ¹³C-versus-chain length profiles of leaf n-alkanes for field-grown C₃ (14 species) and C₄ (24 species) grasses (see Supplemental Information S1). In all C₃ species surveyed, n-alkanes δ¹³C showed an overall decreasing trend with chain growth (C₁₉ to C₃₇), while an overall opposite trend was observed across all C₄ species surveyed (Fig. 1; see Supplemental Figs. S2 and S3 for a complete list of individual profiles). The contrast is clear, regardless of whether the leaves are healthy or unhealthy, or whether the leaves are grown at high or low latitudinal zones (see also Chikaraishi and Naraoka 2007).

Figure 1.

The observed contrasting leaf n-alkanes δ¹³C-versus-chain length profiles between the “averaged” C₃ and C₄ grasses. The y-axes in A and B) represent the arithmetic mean of δ¹³C for all individual species investigated, and the (stepwise) δ¹³C difference between two neighboring (C_2n−1 and C_2n+1) n-alkanes, $δ2n+1ak−δ2n−1ak$⁠, respectively. The error bars represent mean ± Se (n = 14 for C₃ grasses, n = 24 for C₄ grasses). Juxtaposed are the profiles of two representative individual species (Lophatherum gracile Brongniart, a C₃, C), and Ischaemum ciliare Retzius, a C₄, D)). The even-carbon-numbered (ECN) n-alkanes, which are much lower in abundance than, and track the δ¹³C trend of their odd-carbon-numbered (OCN) homologs, are excluded here. The error bars represent mean ± Sd (n = 2).

Open in new tabDownload slide

To facilitate a general (physiological and biochemical) explanation for the contrasting trends of n-alkane δ¹³C-versus-chain length, we related the δ¹³C of n-alkanes to the IEs accompanying the FA chain growth and branching into other lipid derivatives following the established biosynthetic pathways (see Supplemental Information S2 for the histories of carbons in the lipids) using the following equation (see Supplemental Information S2 for the derivation of the below equation):

where $δ2nfa$⁠, $δ2n+2fa$⁠, and $δ2n−1ak$ are the isotope compositions of C_2n FA, C_2n+2 FA, and C_2n−1 alkane, respectively; $εext,$$εrsd$⁠, and $εdcb$ are the isotope fractionations (all negative) associated with the formation of a C–C bond between a C_2n fatty acyl-ACP chain and an acetyl-CoA during chain elongation, between the residual and the original FAs, and decarboxylation or decarbonylation, respectively; $δact$ and $δ2nCO2$ are the isotope compositions of the C₂ extension unit (acetyl-CoA) and the carboxyl carbon of the C_2n FA, respectively; f is the respective flux (F) ratio between decarboxylation/decarbonylation and the elongated, and the residual FAs (lumped with alkanols and aldehydes); n (>10 but <19 in this work) is the number of chain extensions with C₂ units needed for the formation of C_20–38 FAs.

The stepwise δ¹³C change from C_2n−1 alkane to C_2n+1 alkane (⁠$δ2n+1ak−δ2n−1ak$⁠) can be approximated by using the below equation (see Supplemental Information S2 for further details of the derivation),

that is to say, (⁠$δ2n+1ak−δ2n−1ak$⁠) depends largely on $εext,δact$⁠, $εrsd$ and the differences (⁠$Δfrsd$⁠) between two neighboring FAs in their respective f (flux ratio) values. Because only $εext/n+1$ is always negative and chain length dependent among the three terms on the right of Eq. 2, it is therefore the best candidate to account for the observed decreasing trend in the “averaged” C₃ plants (Fig. 1A), which is also observed by Abraham and Hesse (2003), and confirmed by our most recent position-specific carbon isotope analysis (PSIA-¹³C) for a C_16:0 FA of C₃ plant origin, which showed that the C₂ unit ending up in the fatty chain is progressively depleted (Zhao et al. 2023). However, the magnitude of the stepwise decrease (⁠$δ2n+1ak−δ2n−1ak$⁠) increased from C₁₉ to C₂₇ alkane, rather than the predicted decrease, before leveling off slightly thereafter. As $δn+1act$ can either stay unchanged in a metabolic steady state (i.e. acetyl-CoA pool size is stable) or become progressively enriched as more acetyl-CoA is drawn down, reducing the acetyl-CoA pool for chain elongation, and therefore, cannot contribute to the observed magnitude increase of the stepwise decrease (⁠$δ2n+1ak−δ2n−1ak$⁠), substantial flux ratio change $Δfrsd$ during the chain extension of C₃ leaf lipids is therefore inferred.

Because the chain-length-dependent decrease, $εext/n+1$⁠, should apply to both C₃ and C₄ plants, the actual magnitude of the increase with chain length in C₄ plants, caused by an increase in $δn+1act$⁠, alone, or an increase in both $δn+1act$ and $Δfrsd$⁠, should be even greater than that observed across the entire chain length range and at every single cycle of chain extension. Because, as stated above, the sign of $Δfrsd$ is uncertain, the chain-length-dependent isotopic change in the precursor acetyl-CoA, $δn+1act$⁠, must account for the observed increasing trend in the “averaged” C₄ plants. The intramolecular δ¹³C profiles of hexoses (the ultimate precursor, Gilbert et al. 2012) and acetyl-CoA (the immediate precursor) for lipid biosynthesis (see Supplemental Fig. S5) differ between C₃ and C₄ plants (Henn and Chapela 2000). However, such difference can explain only the possible C₃-versus-C₄ difference in the intramolecular ¹³C distribution of the fatty lipid chains, but not the here deduced chain-length dependence of $δn+1act$ in C₄ plants, as chain growth involves only C–C bond formation between fatty acyl-ACP and acetyl-CoA (Julien et al. 2022).

Here, we invoke a Rayleigh-type isotope fractionation mechanism associated with the export of sugars from leaves to explain the deduced chain-length dependence of $δn+1act$ in C₄ plants (Fig. 2). In C₄ monocots, ¹⁴CO₂-pulse-chase experiments showed that >77% of the hexoses (sucrose) are exported, while only about 30% are exported in C₃ monocots (Grodzinski et al. 1998). The IE associated with the export (⁠$εxpt$⁠) of sugars enriches the residual sugars in an export ratio (p)–dependent fashion. A higher p (as in C₄) leads to greater ¹³C-enrichment in the residual sucrose (δ_R) than a lower p (as in C₃). When the residual sucrose is converted to acetyl-CoA, such differential enrichment is passed first onto the FAs and then to other lipids. Considering that only a fraction of the pyruvate generated from the “residual” sugars is converted to acetyl-CoA, the actual p values for C₄ and C₃ are likely even greater. For a realistic $εxpt$ = −3 mUr, the isotopic enrichment difference between C₄ (with a realistic p of 0.9) and C₃ (with a realistic p of 0.4) is estimated to be 5.2 mUr. This difference is sufficient to account for the bulk of the observed stepwise increase (⁠$δ2n+1ak−δ2n−1ak$⁠) (Fig. 1).

Figure 2.

A Rayleigh-type fractionation mechanism explaining the isotopic dependence of the residual sugars (δ_R), the ultimate precursor for the synthesis of the acetyl-CoA used in fatty chain extension, on the portion (p) of exported sugars. The curves are represented by $(δR+1)=(δR,0+1)(1−p)εxpt$⁠, where $εxpt$ and δ_R,0 are the isotopic fractionation associated with sugars export and the initial isotope composition of sugars (the dashed line), respectively. The $εxpt$ = −3 mUr used for the calculation of δ_R is a realistic estimation based on the known −3 mUr fractionation associated with the export of triose phosphates from chloroplast (Gleixner et al. 1998; note that even though there is evidence (Bögelein et al. 2019) for a relative ¹³C-enrichment of sucrose in the phloem sap over leaf sucrose, we believe that is a consequence of respiration in the sink tissues, rather than sugar export from leaf). Because, in general, C₄ (not C₃) acetyl-CoA for the synthesis of FAs is limited, the isotopic consequence of the across-the-board improvement in acetyl-CoA supply (p) increase following infection (upper right for C₃, lower right for C₄) is predicted to be more pronounced in the C₃ longer chain and C₄ short chain during lipid biosynthesis. The error bars represent mean ± Sd (n = 2).

Open in new tabDownload slide

The validity of the proposed Rayleigh-type fractionation mechanism can be tested with growth experiments where paired δ¹³C values and pool sizes of acetyl-CoA, δ¹³C values and abundances of FAs of different classes and chain lengths, and leaf sugar export ratios at various stages of FA biosynthesis can be obtained and compared between C₃ and C₄ grasses. However, an easier, albeit indirect, test is to compare the n-alkane δ¹³C-versus-chain length profiles between healthy and unhealthy (infected with phyllospheric microorganisms such as fungi, bacteria, and viruses) leaves. Here, the p sensitivity to infection is predicted to differ between C₄ and C₃. Indeed, such profiles for both Arundinella anomala Steud. (a C₄) and Leymus chinensis Trin. (a C₃) showed, in addition to a clear δ¹³C decrease in the infected grasses (compared with healthy ones), which can be attributed to respiration and blockage of stomata by microorganisms, and forced recycling of internal CO₂ for photosynthesis (Suryanarayanan et al. 2022), a diminishing decrease toward the longer chains in C₄ plants, and a growing decrease toward the longer chains in C₃ plants. This is likely because C₄ photosynthesis is less affected by infection, as it is equipped with a carbon concentration mechanism, and the priority to allocate acetyl-CoA resources for shorter FA biosynthesis.

In conclusion, we observed an overall decreasing leaf n-alkanes δ¹³C trend with chain growth in C₃ grasses and an overall opposite trend in C₄ grasses. The δ¹³C-versus-chain length profile in C₃ grasses can be explained by a kinetic isotope effect associated with C–C bond formation between a fatty acyl-ACP chain and an acetyl-CoA during chain elongation and flux changes from FA to their longer homologs, fatty alcohols/aldehydes, and n-alkanes, while that for C₄ can be rationalized mainly by a (much) higher export ratio (relative to C₃) of leaf sucrose that enriches the ¹³C of the residual sucrose, the precursor for FA synthesis. The inferred fundamental difference in the metabolic controls of fatty lipid chain extension between C₃ and C₄ grasses can be validated by directly targeting the C₂ unit pool size and isotopic composition in the context of fatty chain elongation.

Acknowledgments

Dr S.T. Li, and Dr W. Sun and Ms S.X. Zhou are thanked for their help with isotopic instrumentation and sample collection, respectively. Discussions with Drs X.Y. Gong, Y.Y. He, and X. Song and Profs B.L. Zhang and C.H. Hocart improved the quality of the manuscript. This is contribution #33 from the Isotopomics in Chemical Biology (ICB) group.

Author contributions

Y.P.Z. conceived and designed the study. Y.Z., Y.P.Z., and H.B.Y. collected samples, conducted experiments, acquired data, and prepared the manuscript; Z.Y.Z., B.W., Q.L.Y., S.W.H., and Z.G.S. contributed to the discussion.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. The GC/MS and GC/C/IRMS chromatograms for Ottochloa nodosa.

Supplemental Figure S2. The δ¹³C and relative abundance-versus-chain length profiles of C₄ grass leaf n-alkanes.

Supplemental Figure S3. The δ¹³C and relative abundance-versus-chain length profiles of C₃ grass leaf n-alkanes.

Supplemental Figure S4. The δ¹³C of individual n-alkanes-versus-chain length profiles for the “averaged” C₃ and C₄ grasses.

Supplemental Figure S5. The pathways (enclosed in the blue box with dashed lines) for n-alkane (and other lipids) biosynthesis in plant leaves and the differences in the ultimate precursors (sucrose 27 in C₃ and 27′ in C₄) and immediate precursors (acetyl-CoA 6a and 6b in C₃, and 6c and 6d in C₄), metabolic controls between C₃ and C₄ cells.

Supplemental Table S1. The δ¹³C differences of individual n-alkanes between the original (certified) and the recovered n-alkane standard mixture from the whole procedure blank experiment.

Supplemental Table S2. The photosynthetic modes of the investigated grass species, and the δ¹³C and chemical abundances of their leaf n-alkanes, together with the published sugar export ratios of C₃ and C₄ grasses.

Supplemental Information S1. The experimental details.

Supplemental Information S2. Deriving the quantitative relationship of δ¹³C of individual n-alkanes with chain length and metabolic branching into other lipids in plant leaves.

Funding

Chinese Natural Science Foundation (NSF) grants (42250203 and 41973072 to Y.P.Z. and 32071507 to Z.Y.Z.) are gratefully acknowledged.

Data availability

All data were included in the supplemental files.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• acetyl-CoA CHEBI: CHEBI:15351

References

Author notes