Allan G. Rasmusson, Lund University, Sweden; Ian M. Møller, Aarhus University, Denmark

Plant mitochondria must exchange metabolites and information (signal transduction) with the cytosol. To give some central examples, where a large rate of exchange must occur, the end products of glycolysis (pyruvate and malate) need to be taken up from the cytosol, several citric acid cycle intermediates need to be exported to be used as building blocks in biosynthetic processes, and photorespiratory glycine, serine, NH₄⁺ and CO₂ need to be exchanged. Likewise, ADP and P_i must be taken up and ATP exported for the mitochondrion to fulfill its role as the powerhouse of the cell. In addition to the dominant transport processes, plant mitochondria must also exchange several other metabolites and coenzymes at lower rates. To carry out the transport of this large number of molecules, plant mitochondria possess a family of mitochondrial carrier proteins. These are transmembrane proteins with a common general structure, and they typically transport relatively small charged molecules, such as metabolites (Picault et al. 2004). In Arabidopsis there are almost 60 members of the mitochondrial carrier protein family (Palmieri, F. et al. 2011). Most of these are not functionally characterized, yet the number still gives an indication of the complexity of the mitochondrial transport processes. However, mitochondria exchange more than metabolites, and therefore also other types of transport systems are needed.

The outer membrane is permeable to molecules of less than 10 kDa, (i.e., all non-polymeric compounds). The permeability results from the presence of pores, which appear to be open in most in vitro experiments. Studies in animals have characterized pore-forming proteins in the mitochondrial outer membrane, called porins, which appear to regulate transport across the membrane. Plant mitochondria also have porins but their characterization awaits further research. In our discussion here we will focus on the regulation of mitochondrial transport across the inner mitochondrial membrane, this being the metabolic boundary of the mitochondrion. We will first consider the relatively few molecules that can move across the inner membrane without the help of a specific carrier protein. We will then discuss the various metabolite transporters and, finally, briefly look at the transport of macromolecules.

Small, Neutral, and Hydrophobic Molecules Diffuse across the Inner Membrane

Irrespective of size, molecules that either are charged or are very hydrophilic can only cross a lipid bilayer with the help of a specific protein carrier (channel or transporter; see textbook Chapter 6).

Small, neutral molecules can diffuse across the lipid bilayer and do not require protein transporters. Among those are O₂ and CO₂, two important molecules for respiration. O₂ is needed on the inner surface of the inner membrane where the active sites of both cytochrome c oxidase and the alternative oxidase are located. The CO₂, produced as an end product in the citric acid cycle (see textbook Figure 12.6), needs to diffuse out of the mitochondrion, ultimately to be lost to the surroundings or re-fixed in the chloroplasts. In addition, however, it is possible that CO₂ can also be transported out of the mitochondrion as bicarbonate (HCO₃^–). Mitochondria are osmotically active and swell in a hypotonic medium. This means that the inner membrane is also permeable to H₂O, but it is likely that the H₂O moves through water channels/aquaporins. There are also indications that an aquaporin may transport hydrogen peroxide (H₂O₂) across the inner mitochondrial membrane (Calamita et al. 2005, Bienert et al. 2007).

For technical reasons, it is important that several well-known inhibitors of mitochondrial function can diffuse into mitochondria. They do so because they are uncharged and hydrophobic. These inhibitors include rotenone, a specific inhibitor of complex I, antimycin, a specific inhibitor of complex III, HCN, the protonated form of the cyanide (CN^–) inhibitor of complex IV (and other heme-containing enzymes), and oligomycin, a specific inhibitor of the F_oF₁-ATP synthase (complex V) (see textbook Figure 12.8).

Chemical uncouplers are very valuable to study membrane bioenergetics because their structure with conjugated double bonds allow them to diffuse across the membrane in both the protonated form and the unprotonated form, and thus dissipate the electrochemical proton gradient. A case in point is FCCP (trifluoromethoxy carbonyl cyanide phenylhydrazone), a very efficient mitochondrial uncoupler, as illustrated in Web Figure 12.5.A (Nicholls and Ferguson 2013).

Web Figure 12.5.A The mechanism of action of uncouplers (also known as protonophores). (A) The protonation/deprotonation of FCCP (trifluoromethoxy carbonyl cyanide phenylhydrazone), which is a weak acid. (B) In the presence of an electrochemical proton gradient across the membrane, FCCP will become protonated and thus pick up a proton on the positive side of the membrane, move across the membrane in the neutral form, and lose the proton on the negative side. This is driven by the electrochemical proton gradient. In the negative form, FCCP will be driven back out across the membrane by the membrane potential. The net result is the movement of one proton from the positive to the negative side of the membrane, which results in the dissipation of the electrochemical proton gradient, and the movement of one electrical charge from the negative to the positive side of the membrane, which results in the dissipation of the electrical gradient. The shaded area represents the system of conjugated double bonds (π-orbital system). (Modified from Nicholls and Ferguson 1992.)

Transport of Inorganic Ions

Inorganic phosphate (H₂PO₄^–) is taken up into the matrix in exchange for a hydroxyl ion (HO^–). The energy for the uptake is supplied by the proton gradient (ΔpH). We will see below that the phosphate gradient, in turn, drives transport of other metabolites.

Several other inorganic ions—Mg²⁺, Ca²⁺, K⁺, and NH₄⁺—are involved in mitochondrial metabolism, but the mechanism by which they pass the inner membrane of plant mitochondria is not well understood. NH₄⁺ may, however, instead be transported as NH₃, via an aquaporin. In mammalian mitochondria, uptake of K⁺ through a K⁺-specific channel, and extrusion via a K⁺/H⁺ antiporter, is involved in matrix volume regulation and cellular signaling (Garlid and Paucek 2003). A similar system of K⁺ transport is present in plant mitochondria (Pastore et al. 1999; Chiandussi et al. 2002).

Both mammalian and plant mitochondria contain an uncoupling protein (UCP), as discussed briefly in Web Topic 12.3. This protein facilitates the movement of protons across the inner membrane and therefore partially uncouples electron transport and decreases the ATP yield of respiration.

Transport of Adenylates

One of the important functions of mitochondria is to supply the cell with ATP. To do this, mitochondria have an adenine nucleotide transporter that facilitates the exchange of ADP^3– (inward) for ATP^4– (outward) (Web Figure 12.5.B). This exchange is driven by the membrane potential (inside negative), since there is a net outward movement of one negative charge. In combination with the phosphate carrier reaction, this means that the uptake of one ADP and one phosphate and the export of one ATP (i.e., the transport event connected to mitochondrial ATP synthesis) costs one vectorial proton—the electrical component for adenylate exchange and the ΔpH component for phosphate uptake.

Web Figure 12.5.B Transporters in the inner membrane of plant mitochondria. All of these transporters are directly or indirectly driven by the electrical and/or chemical part of the proton gradient. Inorganic phosphate (P_i) uptake is driven by the ∆pH (hydroxyl anion moving in the opposite direction) and the resulting P_i gradient is used to drive the uptake of dicarboxylate anions (malate and succinate) by the dicarboxylate transporter. The dicarboxylates can, in turn, exchange for 2-oxoglutarate (also called α-ketoglutarate) or citrate. In the latter case, citrate is protonated on one of its acid groups to maintain transport electroneutrality. The exchange of glutamate (in) for aspartate (out) is electrogenic in baker’s yeast; this means that it must be driven by the electrochemical proton gradient since electroneutral glutamate (glutamate^2– + 2 H⁺) is transported against negatively charged aspartate (Cavero et al. 2003). The latter transporter is important for the operation of the malate/aspartate shuttle (see Web Figure 12.5.C). Similarly, the electroneutral malate/oxaloacetate exchange is involved in the malate/oxaloacetate shuttle. (Modified from Siedow and Day 2000.)

Transport of Citric Acid Cycle Intermediates and Related Compounds

Pyruvate and malate, the end products of glycolysis, both have transporters in the inner membrane (see Web Figure 12.5.B). Pyruvate^– is taken up in exchange for a hydroxyl ion (OH^–) (Divakaruni and Murphy 2012), and malate^2– in exchange for HPO₄^2–.

The transporter for malate^2– also transports succinate^2– and is therefore called the dicarboxylate transporter. Malate^2– or succinate^2– can, in turn, exchange for citrate^2– on the tricarboxylate transporter. Malate^2– can also exchange for 2-oxoglutarate^2– via a special transporter. An interesting transporter, the oxaloacetate transporter, exchanges oxaloacetate^2– for malate^2– (or citrate^2–), which permits the shuttling of reducing equivalents (see below). The joint activity of these transporters allows plant mitochondria to affect the net import or export of most citric acid cycle intermediates (Laloi 1999, Palmieri, F. et al. 2011). Finally, the glutamate/aspartate transporter facilitates the uptake of glutamate. In yeast mitochondria, glutamate is taken up together with a proton in exchange for an aspartate carrying two negative charges corresponding to a net transport of one H⁺. It is therefore driven by both the electrical and the chemical proton gradient. A similar transporter is present in plant mitochondria, but a proton co-transport has not been shown, so the transport may be passive in plants. Recently, several additional transporters for metabolites have been identified (Picault et al. 2004, Palmieri, F. et al. 2011), and it is likely that certain setups of transporters are only expressed under special developmental conditions, like the mobilization of storage nutrients in developing seedlings.

In several cases the involvement of the mitochondrion in cellular metabolism implies that certain metabolites are taken up across the inner membrane, but the carrier has not yet been described. This is the case for the massive flux of glycine into—and CO₂, serine and NH₄⁺ out of—mitochondria in C₃ leaves during photorespiration (see textbook, Chapter 8). This is also the case for proline, which accumulates to high concentrations in the cytosol during episodes of plant water stress (see textbook, Chapter 24), and it is then degraded during the rehydration phase by a pathway involving two mitochondrial enzymes (Yoshiba et al. 1997).

Transport of Coenzymes

Coenzymes (or vitamins as they are called in human nutrition) are either synthesized inside the mitochondrion or imported (see also Web Essay 12.8). Carriers for NAD⁺ (Palmieri, L. et al. 2008a) and thiamine pyrophosphate (Frelin et al. 2012) have been identified in plant mitochondria, which very likely also contain a carrier for coenzyme A (Palmieri, F. et al. 2011).

The final step in ascorbic acid biosynthesis has recently been shown to be on the outer surface of the inner mitochondrial membrane. In view of the accumulating evidence that ascorbate has an important function in removing ROS inside the mitochondrion (see Web Essay 12.7), we would expect to find an ascorbate transporter in the inner mitochondrial membrane.

The situation with folate (involved in 1C transfers) is the opposite of that with ascorbate: The only site of synthesis of many of the folate metabolites in the plant cell is the mitochondrial matrix. It is therefore logical to expect that there is a carrier in the inner membrane dedicated to transporting folate out of plant mitochondria, although the carrier has not yet been identified (Blancquaert et al. 2010, Palmieri, F. et al. 2011) (see also Web Essay 12.8).

Transport of Reducing Equivalents—Metabolic Shuttles

Mammalian mitochondria cannot oxidize cytosolic NADH or NADPH directly, so they make use of "shuttle mechanisms." An exchange of metabolites by inner membrane transporters takes place through these shuttles so that a reduced compound moves into the matrix in exchange for a more oxidized compound moving out into the cytosol. In this way, reducing equivalents are transported into the matrix, ultimately to be oxidized by the respiratory chain. The shuttles can also be used to transport reducing equivalents in the opposite direction.

Although isolated plant mitochondria can oxidize added NADH and NADPH directly by the two external NAD(P)H dehydrogenases (see Web Topic 12.3), they can also use metabolic shuttles (Web Figure 12.5.C). The malate/oxaloacetate shuttle uses malate dehydrogenase in the cytosol and in the matrix to catalyze the interconversion of malate (reduced) and oxaloacetate (oxidized). These two compounds are exchanged by the oxaloacetate transporter (Oliver and McIntosh 1995, Palmieri, L. et al. 2008b). This exchange, however, is not driven by the electrochemical proton gradient across the inner membrane, so it can only move reducing equivalents from a relatively reduced compartment to a relatively oxidized compartment. Since the mitochondrial matrix is much more reduced than the cytosol with respect to the NADH/NAD ratio, the oxaloacetate transporter is likely to work primarily in the export of reducing equivalents (Krömer and Heldt 1991; Wigge et al. 1993).

Web Figure 12.5.C Metabolite shuttles across the inner membrane of mitochondria. The two shuttles shown have both been demonstrated to work in isolated mitochondria. (Top) The malate/aspartate shuttle uses isoforms of two enzymes, malate dehydrogenase (1), and aspartate aminotransferase (2), to interconvert malate and aspartate, which are then exchanged using two transporters, the glutamate/aspartate (A) and the 2-oxoglutarate (α-ketoglutarate) (B). The electrogenic glutamate/aspartate transporter provides directionality so that reducing equivalents are transferred into the mitochondrial matrix (see Web Figure 12.5.B). (Bottom) The malate/oxaloacetate (OAA) shuttle uses malate dehydrogenase isoforms (1) in the matrix and in the cytosol to interconvert malate and OAA, which are then exchanged by the OAA transporter (C). This shuttle can theoretically transfer reducing equivalents in either direction depending on the redox conditions on the two sides of the membrane. However, in plant cells it may only serve to export reducing equivalents. (Modified from Siedow and Day 2000.)

Under photorespiratory conditions, large amounts of glycine are oxidized in the mitochondria to give NADH while, at the same time, NADH is required in the peroxisomes to reduce hydroxypyruvate to glycerate (see textbook Chapter 8). It is thought that as much as 50% of the matrix NADH is exported, probably via the malate/oxaloacetate shuttle (Krömer 1995).

The more complex malate/aspartate shuttle (see Web Figure 12.5.C) involves two matrix enzymes and two cytosolic enzymes, as well as two inner membrane transporters. Because aspartic acid (two negative charges) is exchanged for glutamate plus 1 H⁺ (one negative charge), this exchange is electrogenic and it powers the malate/aspartate shuttle by using the electrochemical proton gradient to export aspartate and in that way aid in the import of reducing equivalents in yeast mitochondria (Cavero et al. 2003). Whether this also occurs in plant mitochondria is not known.

Transport of Proteins—Protein Import

Plant mitochondria contain their own DNA (mtDNA) and ribosomes and are capable of synthesizing the proteins encoded in the mtDNA (see Web Topic 12.6). However, plant mtDNA encodes only a few of the proteins found in the mitochondrion. The great majority of the mitochondrial proteins are encoded by nuclear genes and have to be imported from the cytosol, where they are synthesized on free ribosomes. These proteins are synthesized with a presequence or targeting peptide at the N-terminus, which is recognized by receptors on the outer mitochondrial membrane (Carrie et al. 2013).

The import of matrix proteins requires elaborate machinery comprising two multisubunit protein complexes—these complexes reside in the outer membrane (TOM, transporter of the outer membrane) and in the inner membrane (TIM, transporter of the inner membrane). Concomitant with, or immediately after, the import of the protein into the correct mitochondrial compartment, the targeting peptide is often cleaved off to yield the mature and active protein (Millar et al. 2006). Most proteins destined for other compartments, such as the intermembrane space, are imported either via just the first part of this route or all the way into the matrix and then exported again across the inner membrane (Carrie et al. 2013).

Transport of tRNA

Mammalian mtDNA encodes the full complement of tRNA required for mitochondrial protein synthesis. In contrast, plant mtDNA encodes only some of the tRNAs required for protein synthesis within the mitochondrion (Dietrich et al. 1996). The missing tRNAs are imported by a mechanism involving the TOM complex and a voltage-dependent anion channel in the outer membrane (Salinas et al. 2008).