LlPOPHlLlC CATIONIC PROBE FOR PET - IMAGING FIELD OF THE INVENTION 5 The present invention relates to imaging probes for use in techniques such as positron-emission tomography (PET) and single photon emission computed tomography (SPECT) for the visualisation of mitochondria1 energisation in vivo. lo BACKGROUND TO THE INVENTION Mitochondnal dyshnction contributes to a wide range of pathologies, including. cancer, diabetes, heart failure, cardiovascular and liver diseases, AIDS, autoimmune disorders, degenerative diseases and the pathophysiology of aging. 15 Diseases to which mitochondnal dysfunction contribute are often associated with significant changes in mitochondrial membrane potential (AYm). Alteration in mitochondnal membrane potential is an important characteristic in pathologies that either involve suppressed apoptosis, such as cancer; or enhanced apoptosis, such as 20 AIDS and degenerative diseases. It is also associated with the many diseases caused directly by mitochondnal dyshction such as DNA mutations and oxidative stress. There is thus a need to monitor mitochondria1 hnction in patients. 25 Positron-emission tomography (PET) is a widely used technique to image biological tissues and metabolism withn patients. A short-lived positron-emitting nucleus, such as "F, is incorporated into a probe molecule and injected into a patient. The probe then accumulates in certain tissues. The location of the probe 30 may be visualised from the gamma ray emission using a PET scanner, and the local concentration of the probes deduced from tomography. Lipophilic cations such as tetra- or ti-phenylphosphonium cations penetrate the plasma and mitochondria1 membranes and selectively accumulate in mitochondria because of the negative membrane potential across the inner membrane. 5 The use of tracers based on simple TPP cations for PET has been suggested for imaging mitochondria1 dysfunction (Cheng et a1 (2005) J. Label. Compd. Radiopharm. 48: 13 1-1 37), turnour imaging (Madar et a1 (1 999) J. Nucl. Med. 40:1180-1185; Wang et a1 (2007) 50:5057-5069) and imaging for diagnosis of coronary artery disease (Madar et a1 (2006) J. Nucl. Med. 47: 1359-1 366). 10 The present invention relates to improvements to the known mitochondria-targeted PET probes. DESCRIPTION OF THE FIGURES 15 Figure 1 - Structures of the TPP cations used in the Examples. Figure 2 - Time course of uptake of [ 3 ~ ] ~ iinttoo m~ou se tissues following iv injection. Mice were injected with a bolus of 100 nmol [ 3 ~ ] ~ ibyt iov ~tai l vein 20 injection. At the indicated times the mice were killed and the [ 3 ~ ] ~ icotnoten~t i n the tissues were determined. Data are in nmol MitoQIg wet weight tissue and are means * range for two separate mice per time point. A, liver and kidney, B, heart, muscle, brain and white adipose tissue (fat). C and D, view of the first 1 hour after injection of MitoQ for liver and kidney (C) and for heart, muscle, brain and whte 25 adipose tissue (fat) (D) respectively. Figure 3 - Time course of clearance of [ 3 ~com]pou~nds~ from~ the circulation following iv injection. Mice were injected with a bolus of 100 nmol [ 3 ~ ] ~ ~ ~ compound by iv tail vein injection. At the indicated times the mice were killed and 30 the [ 3 ~con]ten~t in ~the~ blo od were determined. Data are in nmol TPP compound/ml blood and are means * range for two separate mice per time point. A, [)H]M~~oQB,; [ 3 ~ ] ~ e canyd [l3~~~] ~~l u o r o ~ n d e Cc ,y [l ~3~ ~~; ] ~ ~ ~ ~ . Figure 4 - Time~couneo f uptake of []H]D~C~ITaPnPd [ 3 ~ ] ~ l u o r o ~ n d einctoy l ~ ~ ~ tissues. Mice were injected with a bolus of 100 nmol of [ 3 ~ ] ~ e coyr l ~ ~ ~ [ 3 ~ ] ~ l u o r o ~ n d ebcy yinlje~ct~io~n into the tail vein. At the indicated times the 5 mice were killed and the content of ['H]D~C~~ToPr P[ 3 ~ ] ~ l u o r o ~ n d einc thyel ~ ~ ~ tissues determined. Data are means * range for two separate mice per time point. A, liver and ladney for [ 3 ~ ] ~ e c yB,l h~ea~rt, ~mu, s cle, brain and whte adipose tissue (fat) for [ 3 ~ ] ~ e c, Cy, alll~ tis~sue~s for [ 3 ~ ] ~ l u o r o ~ n d e c y l ~ ~ ~ . 10 Figure 5 - Time course for uptake of [ 3 ~into ]tissue~s. M~ice w~ere in~jected with a bolus of 100 nmol of [ 3 ~by in]jecti~on int~o the ~tail v~ein. A t indicated times the mice were killed and the content of [ 3 ~in th]e tiss~ues de~termi~ned. ~ Data are means * range for two separate mice per time point. A, liver and kidney, B, heart, muscle, brain and white adipose tissue (fat). 15 Figure 6 - Comparison of uptake of TPP compounds into tissues at different times. Mice were injected with an iv bolus of 100 nmol of [ 3 ~ ] ~ i t[o3~~, ] ~ e c y l ~ ~ ~ , [ 3 ~ ] ~ l u o r o ~ n d eorc [y l ~3~ ~ ~and a]t 15 ~min (~A, B)~, 1 h (~C, D) or 5 h (E, F) the mice were killed and the tissue content of [ 3 ~com]pou~nds~ dete~rmi ned. 20 Data are means f range for two separate mice per time point and in A, C & E are nmol TPP compound/g wet weight tissue and in By D & F are in nmol TPP compound/ml blood. Figure 7 - Mitochondnal uptake of TPP compounds in vivo. A, Mice were injected 25 with a bolus of 100 nmol [ 3 ~ ] ~ l u o r o ~ n d ebyc yinlje~ct~io~n into the tail vein. After 15 min the mice were injected ip with DNP (200 or 300 pgkg) or saline carrier and after a hrther 15 min later they were killed and the content of [ 3 ~ ] ~ l u o r o ~ n d einc tyhel ~tis~su~es determined. Data are means range for two mice per condition. B, IAM-TPP is shown being taken up in to mitochondria withn 30 a cell where it reacts with tho1 proteins to form a thioether adduct that can then be detected by irnmunoblotting. C, Confocal image of IAM-TPP binding to mitochondria in cells. C2C12 cells were incubated with 1 pM IAM-TPP for 3 h * 10 pM FCCP. The cells were then fixed and the location of the TPP moiety within the cells determined by labelling with antiserum against the TPP moiety, visualised by irnrnunofluorescence confocal microscopy. Control experiments confirmed that the IAM-TPP binding colocalised with the mitochondria-specific dye Mitotracker 5 Orange (data not shown). D, Mice were injected with a bolus of 500 nmol of IAMTPP by injection into the tail vein. After 1 h the mice were killed and liver and heart mitochondria were prepared. The mitochondria (40 pg protein) were separated by SDS-PAGE and proteins that had been labelled with IAM-TPP were detected by irnrnunoblotting using antiserum against the TPP moiety. Mitochondria 10 from mice that had not been exposed to IAM-TPP were used as controls. The experiment was repeated on three separate mice with similar results. Figure 8 - Synthesis of the mitochondria-targeted PET probe 18FFluoroUndecylTPP 15 SUMMARY OF ASPECTS OF THE INVENTION The present inventors have surprisingly found that if the hydrophobicity of the imaging probe is increased, for example by incorporating a hydrophobic moiety, 20 this greatly increases the extent of accumulation in mitochondria and increases clearance of the probe from circulation, leading to a greater tissue:circulation ratio. Together these factors mean that the hydrophobic mitochondna-targeted imaging probes of the present invention are 20-100 fold more sensitive and have better 25 tissue loading and contrast properties than currently used imaging probes for the visualisation of mitochondria1 energisation in vivo. Thus, in a first aspect, the present invention provides an imaging probe whch comprises a lipophilic cation, a hydrophobic moiety and a PET nucleus. 30 The imaging probe may be for use, for example, in positron-emission tomography (PET) andlor single photon emission computed tomography (SPECT) The lipophilic cation may be or comprise triphenylphosphonium (TPP). The hydrophobic moiety may be or comprise an aliphatic chain, for example an aliphatic chain comprising at least 5 carbon atoms. The hydrophobic moiety may comprise a linear alkane chain, for example a linear decyl or undecyl chain. The PET nucleus may, for example, be "F. 10 The hydrophobic moiety may act as a linker between the lipophilic cation and the PET nucleus. In a second aspect, the present invention provides a method for analysing 15 mitochondnal membrane potential in a subject which comprises the following steps: (i) administration of an imaging probe according to the first aspect of the invention to the subject; (ii) visualisation of the probe; and (iii) deduction of the absolute or relative magnitude of the mitochondrial membrane potential. Mitochondria1 membrane potential may be analysed, for example, to visualise tumours, investigate mitochondrial damage, diagnose/monitor a pathology which 25 involves a change in mitochondrial energisation in a subject, or to investigate the effect of a test compound on mitochondrial potential. In a third aspect, the present invention provides an imaging probe according to the first aspect of the invention for use in (i) a method for analysing mitochondrial membrane potential; (ii) a method for visualising tumours in a subject; (iii) a method for investigating mitochondrial damage in a subject; (iv) a method for diagnosing and/or monitoring a pathology in a subject; or (v) a method for investigating the effect of a test compound on mitochondria1 potential. 5 In a fourth aspect, the present invention provides a precursor molecule comprising a lipophlic cation and a hydrophobic moiety which can be reacted with an anionic form of the PET nucleus to produce an imaging probe according to the first aspect of the invention. 10 The precursor molecule may comprise a mesylate group which reacts with the anionic form of the PET nucleus. For example, the precursor molecule may be a rnesylated alkyl triphenylphosphonium compound, reactable with 1 8 ~t-o form "FFluoroalkylTPP. 15 In a fifth aspect, the present invention provides a method for producing an imaging probe according to the first aspect of the invention, which comprises the step of reacting an anionic form of the PET nucleus with a precursor molecule according to the fourth aspect of the invention. 20 The present invention also provides method for producing and administering an imaging probe according to the first aspect of the invention to a subject which comprises the following steps: (i) synthesis of an anionic form of a PET nucleus, such as "F- 25 (ii) incorporation of the anionic form of the PET nucleus into a precursor molecule according to any of the fourth aspect of the invention to produce an imaging probe; (ii) administration of the Imaging probe to the subject. 30 The present invention also provides: (i) a method for increasing the uptake of an imaging probe which comprises a triphenylphosphonium (TPP) cation which comprises the step of increasing the hydrophobicity of the imaging probe; and (ii) a method for increasing the tissue:circulation ratio of an imaging probe 5 which comprises a triphenylphosphoniurn (TPP) cation, whlch method comprises the step of increasing the hydrophobicity of the Imaging probe. Hydrophobicity of the imaging probe may, for example, be increased by incorporation of an alkyl chain having at least 5 carbon atoms. 10 DETAILED DESCRIPTION POSITRON-EMISSION TOMOGRAPHY (PET) 15 Positron emission tomography (PET) is a nuclear medicine imaging technique which produces a three-dimensional image or picture of functional processes in the body. The system detects pairs of gamma rays emitted indirectly by a positronemitting radionuclide (tracer), which is introduced into the body on a biologically active molecule. Images of tracer concentration in 3-dimensional space within the 20 body are then reconstructed by computer analysis. In modem scanners, this reconstruction is often accomplished with the aid of a CT X-ray scan performed on the patient during the same session, in the same machine. The glucose analog fluorine-1 8 (F- 18) fluorodeoxyglucose (FDG) is a biologically 25 active molecule for PET whch is widely used in clinical oncology. This tracer is taken up by glucose-using cells and phosphorylated by hexokinase, the concentrations of tracer imaged then give tissue metabolic activity, in terms of regional glucose uptake. 30 For imaging mitochondria andlor changes in mitochondria1 membrane potential, it is possible to use lipophilic cations which selectively accumulate in mitochondria due to the negative inner membrane potential (-1 20 to -1 70 mV). Such lipophilic cations include rhodamine- 123 (Rh123) and tetraphenyphosphonium salts. The term "positron-emission tomography (PET) probe" or "imaging probe" used 5 herein means a molecule suitable for use in positron-emission tomography, SPECT or any other imaging technique, which can be administered to a patient, for example, by injection, and which accumulates in a tissue of interest. The location and local concentration of the probe can then be deduced using PET scanning and tomography, SPECT or another type of imaging technque. 10 A rnitochondna-targeted imaging probe selectively accumulates in mitochondna, possibly due to the high mitochondria1 membrane potential. The uptake of the probe may be ATm-dependent so that the probe can give information on the energisation status of mitochondna. 15 The probe of the present invention may have a distribution profile in the body which is a fimction of mitochondrial integrity. The probes of the present invention may be useful for imaging variations in 20 mitochondrial surface potential (AT,) imaging cells or tissues having dysfunctional mitochondna and imaging or monitoring diseases or conditions associated with dysfunctional mitochondria. The imaging probe of the present invention comprises a lipophilic cation, a 25 hydrophobic moiety and a PET nucleus. SINGLE PHOTON EMISSION COMPUTED TOMOGRAPHY (SPECT) Single photon emission computed tomography (SPECT) is a nuclear medicine 30 tomographic imaging technique using gamma rays. It is very similar to conventional nuclear medicine planar imaging using a gamma camera, but is able to provide true 3D information. This information is typically presented as crosssectional slices through the patient, but can be freely reformatted or manipulated as required. The basic technique requires injection of a gamma-emitting radioisotope (also s called radionuclide) into the bloodsteam of the patient. This may involve the attachment of a marker radioisotope to a ligand which is of interest for its chemical bindmg properties to certain types of tissues. This marriage allows the combination of ligand and radioisotope (the radiopharmaceutical) to be carried and bound to a place of interest in the body, which then (due to the gamma-emission of the 10 isotope) allows the ligand concentration to be seen by a gamma-camera. LIPOPHILIC CATION The lipophilic cation moiety of the probe of the present invention may be any 15 cation which accumulates in mitochondria due to the hlgh mitochondnal membrane potential. The cation may have a delocalised positive charge which promotes its AY,-dependent accumulation into mitochondria and its passage through phospholipid bilayers. 20 Examples of such cations include Rhodamine-123 and phosphonium cations, triphenyl and tetraphenyl phosphonium derivatives, arsonium derivatives, quaternary amines with hydrophobic groups e.g. tetrabenzyl ammonium, and hydrophobic aromatic systems with delocalised positive charges akin to rhodamine. 25 Several labelled phosphonium cations, such as [' '~]meth~ltri~hen~l~hos~honiurn, 223-['8~]~luoropropaynd1 4-[18~]~luorobenzyltriarylphosphonhiauvme been used as mitochondnal targeting agents. The lipophilic cation may be triphenylphosphonium which when linked to the 30 hydrophobic moiety (see below) produces a lipophilic alkyl tiphenylphosphonium cation. HYDROPHOBIC MOIETY The hydrophobic moiety increases the overall hydrophobicity of the cation when associated with it, for example by covalent linkage. The the octanol1PBS partition 5 coefficient of the imaging probe including the hydrophobic moiety may be at least 50, 100,250, 500, 750 or 1000. The coefficients for MethylTPP and Decyl TPP are 0.35 and 5000 respectively where the larger the number reflects the higher hydrophobicity. FluoroUndecyl has a value of 740. 10 The hydrophobic moiety may, for example be an aliphatic chain. The aliphatic chain may comprise at least 2 carbon atoms, for example between 5 to 20, 8 to 15, or 10 to 12 carbon atoms. The aliphatic chain may have 10 or 1 I carbon atoms. The hydrophobic moiety may comprise an alkyl chain, whch may be a 15 substantially or completely linear alkyl chain, or include some branching. The chain may comprise one or more hetero atoms (e.g 0 , S, N, P) internally and or at the terminus. The hydrophobic moiety may, in addition, contain unsaturated (alkenyl, alkynyl, aryl, heteroaryl) components andlor may comprise one or more aromatic insertions. 20 The hydrophobic moiety may be covalently linked to the lipophilic cation. Where the lipophilic cation is triphenylphosphonium (TPP) the hydrophobic moiety may be linked to the central phosphorus ion as shown in Figure 1. 25 Without wishing to be bound by theory, the present inventors believe that the increased uptake associated with imaging probes comprising a hydrophobic moiety is due to more rapid permeation of the plasma membrane and the increased adsorption of the hydrophobic moiety to the matrix-facing surface of the mitochondria1 inner membrane. 3 0 PET NUCLEUS Non-radioactive elements and their counterparts that can be used in the probes of the present invention include: F- 19 (F-18); C-12 (C-1 1); 1-1 27 (I- 125, I- 124, 1-1 3 1 and 1-123); CI-36 (CI-32, CI-33, CI-34); Br-80 (Br-74, Br-75, Br-76, Br-77, Br- 78); Re-1851187 (Re-186, Re-188); Y-89 (Y-90, Y-86); Lu-177 and Sm-153. 5 Alternatively the probes of the present invention may be labeled with one or more radio-isotopes, such as "c, "F, 7 6 ~ r1,2 31~2, 4~13, '1, 1 3 ~or, 150. Radionuclides used in PET scanning are typically isotopes with short half lives lo such as carbon- 1 1 (-20 min), nitrogen-1 3 (-1 0 min), oxygen- 15 (-2 min), and fluorine- 1 8 (-1 10 min). The PET nucleus may comprise 1 8 ~ . 1s The term "PET nucleus" refers to a non-radioactive element or radionuclide which may be used in PET, SPECT or other imaging processes. The PET nucleus may be attached, for example covalently linked to the lipophilic cation and/or the hydrophobic moiety. For example the hydrophobic moeity may 20 act as a linker between the lipophilic cation and the PET nucleus. The probe may be 1 8 ~ - ~ l u o r o ~ n d e c y l ~ ~ ~ . METHOD FOR USE 25 The present invention also provides a method for analysing mitochondnal membrane potential in a subject which comprises the following steps: (i) administration of an imaging probe according to the present invention to the subject; 30 (ii) visualisation of the probe; and (iii) deduction of the mitochondria1 membrane potential. The imaging apparatus used to detect ,and monitor the imaging agent include imaging technologies such as gamma camera, PET apparatus and SPECT apparatus. 5 Analysis of the mitochondnal membrane potential may be used in, for example, diagnosing andlor monitoring a pathology which involves a change in mitochondrial energisation in a subject. Analysis of the mitochondrial membrane potential may be used in, for example, a 10 method for visualising turnours or a method for investigating mitochondrial damage in a subject. The probe may be administered by any suitable techque known in the art, such as direct injection. Injection may be intravenous (IV). Administration may be general 15 or local to the site of interest, such as to a tumour. The probe may be used in conjunction with another probe, for example a probe capable of visualising a particular tissue or a tumour. The two (or more) probes may be administered together, separately or sequentially. 20 The imaging probe of the present invention may be used to diagnose, assess or monitor the progression or treatment of a disease or condition. The imaging probe of the present invention may be used to investigate the effects 25 of a test compound on mitochondnal energisation. For example, the imaging probe may be administered together with a test compound, to and the effect of the test compound on mitochondrial energisation be assayed in real time in vivo using a method in accordance with the present invention. 30 DISEASE The disease or condition may be characterized by a change in mitochondnal energisation. For example, a change in mitochondrial energisation (either a higher or lower mitochondrial membrane potential) may be a symptom of the disease or may be the, or one of the, causative factors of the disease, 5 Full or partial reversal of the pathogenic mitochondrial energisation state following treatment may be indicative of therapeutic efficacy. Mitochondria1 oxidative damage contributes to many pathologies because 10 mitochondna are a source of reactive oxygen species and are also susceptible to oxidative damage. Various diseases and conditions are associated with dysfunctional mitochondria, including various cancers, diabetes, heart failure, cardiovascular and liver diseases, 15 AIDS, degenerative diseases, immune disorders, aging and other myopathies. The present invention provides probes that are taken up by mitochondria, the uptake being proportional to AY,. This allows detection and imaging of dysfbnctional mitochondria, for example mitochondria with suppressed or 20 enhanced activity. Tumours commonly have a higher mitochondnal membrane potential, whereas areas of tissue damage may have a lower AY,. The condition andor its treatment may be characterised by increased or decreased apoptosis, which may be monitored using an imagng probe according to the 25 present invention. Loss of mitochondrial membrane potential is an early event in cell death caused by pro-apoptotic agents. Mitochondria-controlled apoptosis is thought to underlie cell loss in heart failure The imaging methods of the invention may also be used to assess the efficacy of 30 chemotherapy or radiation treatment protocols used to retard or destroy cancer and other malignant tumours. The imaging methods of the present invention may be used to diagnose or assess cancer, for example lung, breast or prostate cancer. It has been demonstrated that the mitochondrial transmembrane potential in s carcinoma cells is significantly higher than in normal epithelia1 cells. For example, the difference in AYm between the CX-1 colon carcinoma cell line and the control green monkey epithelial cell line CV-1 was approximately 60mV (163mV in tumour cells versus 104 mV in normal cells). 10 SUBJECT - The subject may be human or animal'subject. The subject may be a healthy subject or a subject having or at risk from contracting a disease. 15 In particular the subject may have or be at risk from contracting one of the diseases or conditions mentioned in the previous section. The subject may be undergoing treatment for the disease. The imaging probe of the invention may be used to investigate changes in mitochondria1 energisation which 20 are associated with progression of or amelioration of the disease or condition. The subject may be an experimental animal, in particular and animal model of one of the diseases or conditions mentioned in the previous section. 2s PROBE SYNTHESIS The PET nucleus, for example "F, may be incorporated into a precursor form of the imaging probe which is capable of receiving or adapted to receive the PET nucleus. 3 0 For example, ' 8 m~a y be synthesised in a cyclotron by methods known in the art. After synthesis, the "F is commonly in the F- form and, in view of its 110 minute half-life, needs to be rapidly incorporated into the imaging probe, purified and administered to the subject. The present invention also provides a precursor molecule adapted to receive a PET 5 nucleus. For example, the present invention provides a precursor molecule comprising a lipophilic cation and a hydrophobic moiety which can be reacted with an anionic form of a PET nucleus to produce an imaging probe according to the present invention. 10 The precursor molecule may, for example, comprise a leaving group which is susceptible to nucleophilic substitution. For example the precursor molecule may comprise a mesylate, tosylate, nosylate, triflate or iodo group, which reacts with the anionic form of the PET nucleus. 15 The precursor molecule may be a mesylated alkyl triphenylphosphonium compound, reactable with "F- to form 1 8 ~ - ~ l u o r o a l k yFlig~u~re~ 8. shows the reaction of a mesylated undecylTPP precursor with "F- to form 1 8 ~ - u n d e c y l ~ ~ ~ . The present invention also provides a method for producing an imaging probe 20 which comprises the step of reacting an anionic form of the PET nucleus with such a precursor molecule. This provides a convenient :one-step procedure for production of the imaging probe. The present invention also provides a method for producing and administering an 25 imaging probe of the invention to a subject which comprises the following steps: (i) synthesis of "F-, for example in a cyclotron (ii) incorporation of into a precursor molecule to produce an imaging probe; (iii) optionally purifying the imaging probe; and 3 0 (ii) administration of the optionally purified imaging probe to the subject. The probe should be administered to the subject as soon as possible after its synthesis. INCREASE IN UPTAKE AND CLEARANCE FROM CIRCULATION 5 The present inventors have found that inclusion of a hydrophobic moiety into a mitochondria-targeted imaging probe increases its uptake into tissues and the extent to which it is cleared fiom the circulation. The uptake relative to background circulation is greater, leading to a greater tissue/circulation ratio. This greatly 10 enhances the sensitivity of these probes for detecting and visualising changes in mitochondnal energisation in vivo. The present invention thus provides a method for increasing the uptake of an imaging probe which comprises a tiphenylphosphoniurn (TPP) cation whlch 15 comprises the step of increasing the hydrophobicity of the imaging probe. The rate andor extent of uptake may be increased 5 to 50, 10 to 40, or 20 to 30 fold when compared to the uptake of the corresponding compound which lacks the hydrophobic moiety. 20 Uptake may be increased, in particular into certain tissues, such as kidney, muscle, heart, liver and fat. Differences in uptake may be measured, for example between lhr and Shrs after 25 administration. The present invention also provides a method for increasing the tissue:circulation ratio of an imaging probe which comprises a tiphenylphosphonium (TPP) cation, which method comprises the step of increasing the hydrophobicity of the imaging 30 probe. The tissue:circulation ratio may be obtained by comparing the concentration of the compound in the tissue (e.g. ludney, liver, muscle or heart) with the concentration of the compound in the circulation (e.g. blood). 5 Clearance of the imaging probe from the circulation may be increased 5 to 50, 10 to 40, or 20 to 30 fold when compared to clearance of the corresponding imaging probe whch lacks the hydrophobic moiety. The tissue/circulation ratio may be at least 50-, 80-, or 100-fold greater than that of the corresponding compound which lacks the hydrophobic moiety. 10 An example of an imaging probe having a hydrophobic moiety is "FFluoroUndecylTPP, and the corresponhng imaging probe lacking the hydrophobic moiety is "F-TPP or "F-TPMP. As the PET nucleus is unlikely to affect uptake or clearance of the probe, the PET may be compared with the corresponding molecule 15 lacking the hydrophobic moiety and the PET nucleus (e.g. TPP or TPMP). Hydrophobicity may therefore be increased by introducing or increasing the length of an alkyl side chain of a triphenylphosphonium lipophilic cation, such that is has at least 5, for example between 8 and 15 carbon atoms. Hydrophobicity may also 20 be increased by adding side chains to the alkyl groups, putting aromatic groups in the chain and adding side group to the phenyl rings on the triphenylphosphonium moiety. The invention will now be further described by way of Examples, whch are meant 25 to serve to assist one of ordinary skill in the art in carrying out the invention and are not intended in any way to limit the scope of the invention. EXAMPLES 30 Example 1 - Investigating; the kinetics of or~and istribution of a mitochondriatargeted alkvlTPP compound follow in^ iv injection To determine how rapidly alkylTPP compounds distribute within the body, the uptake of the mitochondria-targeted antioxidant MitoQ was first assessed following intravenous (iv) injection (Fig. 2). To do ths a bolus of 100 nrnol [ 3 ~ ] ~ iwtaso ~ administered to mice via tail vein injection and the amount of [ 3 ~ ] ~ int voari~ou s 5 organs was measured over the subsequent 48 h (Fig. 2). There was extensive uptake of MitoQ by the liver and kidneys when measured 1 h after injection (Fig. 2A). At this time there was less extensive uptake by the heart, lower uptake by muscle and adipose tissue and very little uptake by the brain (Fig. 2B). This rapid uptake of MitoQ into organs was mirrored by its early loss from the blood (Fig. 10 3A). The [)H]Mito~ac cumulated by the tissues was then gradually lost with half lives of -2 h for the liver, - 4 h for the hdneys and -15 h for the heart. There was almost complete clearance of the administered radioactivity from these tissues within 48 h, consistent with other studies where -80% of MitoQ was eliminated 24 h after iv injection. To assess the initial uptake of [ 3 ~ ] ~ iinttoo ti~ssu es focus was 15 made on its tissue distribution over the first hour following injection (Figs. 2C & D). This showed that MitoQ uptake into tissues is very rapid, reaching its maximum at the earliest time measurable, 5 min after injection. Together these data show that MitoQ is taken up rapidly from the blood into most organs, but with significant variations in the extent of uptake by different organs. 20 To see if the rapid uptake of MitoQ into organs from the circulation was a general property of all alkylTPP cations, or was specific to MitoQ, the organ distribution following iv injection of three alkylTPP cations, [ 3 ~ ] d e c y l ~ ~ ~ , [ 3 ~ ] ~ l u o r o ~ n d eacnyd l[~ ~ ~3 ~was] asse~ssed ~next.~ The~se co mpounds 25 ranged in hydrophobicity from the relatively hydrophilic [ 3 ~(Oct]anolI~PBS ~ ~ ~ partition coefficient 0.35) to the hydrophobic [ 3 ~ ] ~ e c yanld~ ~ ~ [ 3 ~ ] ~ l u o r o ~ n d e(Occytaln~oU~P~BS partition coefficients of 5000 and 740 * 100, respectively). The tissue distribution of [ 3 ~ ] ~ e covyerl 4~8 h~ is~ sh own in Figs. 4A & B, and that of [ 3 ~ ] ~ l u o r o ~ n d eocveyr l5~ h~ is~ s hown in Fig. 4C. Their 30 uptake profiles were similar to that of MitoQ, with rapid uptake into the liver, kidney and heart, followed by loss with half lives of -3 h for both compounds from the liver and kidneys. As was the case with MitoQ, loss from the heart was slower, with half lives of -1 5 h and - 2 1 h for ['H]D~C~~TanPdP [ ' ~ ] ~ l u o r o ~ n d e c ~ l ~ ~ ~ , respectively. The clearance of ['H]D~C~~TaPndP [ 3 ~ ] ~ l u o r o ~ n d efic-oyml t~he~ ~ blood (Fig. 3B) was qualitatively similar to that of ['HlMito~ (Fig. 3A), but their blood concentrations were lower. The organ distribution of ['HITPMP was 5 measured for 5 h following iv injection (Figs. 5 A,B). The [ 3 ~distr]ibutio~n ~ ~ ~ was qualitatively similar to those of MitoQ, DecylTPP and FluoroUndecylTPP, with rapid uptake into the liver, heart, kidneys and fat. However the extent of TPMP uptake was generally less than that of MitoQ, DecylTPP and FluoroUndecylTPP, with the exception of the brain. The half lives for loss of 10 ['HITPMP From the liver and kidneys were -3 h and - 2 h, respectively, similar to the other TPP compounds. However, the half life for loss of TPMP fi-om the heart was - 2 h, significantly shorter than for the other alkylTPP compounds. The clearance of [ 3 ~froln] the b~lood ~was q~ualit~ativel y similar to that of the more hydrophobic alkylTPP compounds, although the concentration of ['HITPMP in the 15 blood remained higher (Fig. 3C). Comparing the half lives for clearance of [ 3 ~and ]the o~ther ~alkyl~TPP ~comp ounds horn the remaining tissues was technically difficult due to the low amounts of compound taken up, however for both fat and muscle the half life for clearance of TPMP was estimated at -1-2 h, while those for DecylTPP and FluoroUndecylTPP were significantly longer at -7- 20 12 h. Together these experiments demonstrated that there was extensive and rapid uptake of all alkylTPP compounds into organs fi-om the blood. Example 2 - Comparison of the extent of uptake of alkylTPP compounds into organs over time 2 5 The data in Figs. 2, 4 & 5 show that the organ distributions of the four TPP compounds assessed are qualitatively similar. However, there were significant quantitative differences between the extents of uptake of the compounds into different organs. To assess these differences we have plotted the tissue contents of 30 the four TPP compounds at 15 min, 1 h and 5 h after iv injection (Figs. 6 A, C, E). As the amounts taken up into muscle, fat and brain were lower than other organs, these data are also presented as expanded inserts within each panel. This analysis showed that for the ludneys, heart and muscle the extent of uptake was in order: DecylTPP, FluoroUndecylTPP > MitoQ > TPMP, and these differences were most evident 1 to 5 h after injection. The magnitudes of these differences were large, for example the uptake of DecylTPP, MitoQ and TPMP into the heart 1 h after 5 injection being in the ratio of 17:6: 1 and by 5 h this ratio was 32: 10: 1. The organ distribution for MitoQ, DecylTPP and FluoroUndecylTPP were similar with uptake in the order: kidney > liver > heart > muscle, fat > brain. The organ distribution of TPMP is broadly similar to that of the three more hydrophobic compounds, with the exception that the uptake into the liver was greater than in the kidney for this 10 compound and there was greater uptake into the brain. AlkylTPP compounds are accumulated into tissues in response to the plasma and mitochondrial membrane potentials, as described by the Nernst equation. Consequently the concentration of the compound w i t h mitochondria in the tissue 15 will be determined by the mitochondrial and plasma membrane potentials, and by the concentration of compound in the circulation. As these membrane potentials are similar for all experiments reported here, a major determinant of the extent of compound uptake into tissues is its concentration in the blood. To correct for alterations in organ uptake due to this we also determined the ratio of the 20 concentration of compound in the organ to its concentration in the circulation, and these data are shown at 15 min, 1 h and 5 h after iv injection (Figs 6B, D and F). These data corroborate the findings of Figs. 6A, C and E and confirm that the extent of uptake was in order: DecylTPP, FluoroUndecylTPP > MitoQ > TPMP. For example, the heart/blood ratios for DecylTPP, MitoQ and TPMP 1 h after 25 injection are in the ratio of 109: 11 :1 and by 5 h this ratio was 248:26: 1. One exception is that while the liver/blood ratio of MitoQ and TPMP were similar, those for DecylTPP and FluoroUndecylTPP were significantly greater. In addition, there was little difference between the brainhlood ratio for all four compounds, suggesting that the apparent greater uptake of TPMP seen in Figs. 6 A, C and E 30 may arise due to its presence in entrapped blood, rather than in the brain itself. Together, the data in Fig. 6 indicate that the organ distribution for all four compounds is qualitatively similar. DecylTPP and FluoroUndecylTPP are taken up to a significantly greater extent than MitoQ, and in turn MitoQ is taken up and retained to a far greater extent than TPMP. Example 3 - Uptake of alkvlTPP compounds bv mitochondria in vivo 5 The results of Examples 1 and 2 show that alkylTPP compounds are taken up rapidly into organs in vivo. The uptake of TPP compounds into mitochondna and cells has been shown to be due to the mitochondnal and plasma membrane potentials with most of the accumulated compound being within mitochondria. lo Therefore it is likely that once the alkylTPP compounds are accumulated withn tissues they are predominantly localised withn rnitochondna, driven by the membrane potential. To see if this was the case in these experiments, it was next determined if the uptake of the alkylTPP compounds in vivo was decreased by lowering the mitochondria1 membrane potential. To do this, 15 [ 3 ~ ] ~ l u o r o ~ n d ewcasy iln~jec~te~d into mice iv and after 15 min the mice were hrther injected with various doses of the mitochondrial uncoupler 2,4- dinitrophenol (DNP) or saline carrier and after a further 15 min the extent of [ 3 ~ ] ~ l u o r o ~ n d eacccyuml ~ul~ati~on by the organs was measured (Fig. 7A). The amounts of DNP used have been shown to cause partial uncoupling of 20 mitochondna in vivo without toxicity. Our experiments showed that DNP decreased the uptake of [ 3 ~ ] ~ l u o r o ~ n d einctoy tlh~e o~rg~a ns by up to 40%, in a dose dependent manner (Fig. 7A). These findings are consistent with the extent of uptake of alkylTPP compounds within organs in vivo being determined by the mitochondnal membrane potential. 25 To demonstrate directly the localization of alkylTPP compounds within mitochondna in vivo is technically challenging due to their rapid redistribution upon the tissue homogenisation required for mitochondnal isolation. To overcome this problem in the past,the modified alkylTPP compounds 4-iodobutylTPP (IBTP) 30 and 10-iododecylTPP (1DTP)have been used. These compounds are accumulated by mitochondna in the same way as other alkylTPP compounds, but once within the mitochondrial matrix the iodo moiety is displaced by mitochondrial thiol proteins to form stable thioether adducts whlch can be visualised using antibodies against the TPP moiety. As the rates of reaction of IBTP and IDTP with protein thiols are relatively slow, this approach was extended to make a TPP compound attached to the more thiol-reactive iodoacetamide (IAM) moiety (IAM-TPP; Fig 5 7B). The intention was that this molecule should be accumulated by mitochondria and there label mitochondnal thiol proteins, with the enhanced thiol reactivity of the IAM moiety facilitating the reaction of IAM-TPP with mitochondrial protein thiols in vivo before it was cleared from the organ (Fig. 7B). On incubation with isolated mitochondria IAM-TPP rapidly reacted with protein thiols and could be 10 detected using antiTPP antibodies as was shown previously for IBTP (data not shown). Incubation of IAM-TPP with cells followed by analysis by imrnunocytochemistry showed uncoupler-sensitive localisation of IAM-TPP within mitochondria (Fig 7C), confirming that IAM-TPP labelled mitochondnal proteins. A bolus of IAM-TPP was then adrmnistered to mice by iv injection and isolated 15 mitochondria from the heart and liver 1 h later. Western blots of these mitochondrial fractions using antiserum selective for the TPP moiety showed labelling of heart and liver mitochondrial proteins by IAM-TPP (Fig. 7D). Together these data indicate that following iv injection alkylTPP compounds are rapidly taken up by mitochondria within tissues driven by the mitochondrial 20 membrane potential. It has thus been shown that alkylTPP compounds are accumulated by many organs in vivo within 5 min of iv administration giving therapeutically effective amounts of compound in the tissues. This uptake into tissues was due to the membrane 25 potential-dependent accumulation of the compounds into mitochondna. An interesting finding in this study was that DecylTPP and FluoroUndecylTPP compounds were far more extensively taken up into tissues than TPMP. This is due to the greater adsorption of the longer chain akylTPP cations to the matrix-facing 30 surface of the mitochondrial inner membrane. Importantly, the enhancement of the extent of uptake of TPP cations by altering the side chain was large, 10 to 30-fold. In addition to elevating the absolute levels of alkylTPP compounds within organs, modifying the alkyl side chain to a FluoroUndecyl or Decyl moiety greatly enhanced the organlblood concentration ratio of the alkylTPP compounds by over a hundred-fold relative to TPMP. This property will be useful for the design of more 5 effective PET probes to assess mitochondnal function in vivo in order to assess changes in mitochondria1 polarisation associated with cancer and with cell death. The data presented here suggest that PET probes based on long chain, hydrophobic alkylTPP cations may have significantly better tissue loading and contrast properties than the alkylTPP cations that have been developed as PET probes to 10 date, whch are similar to TPMP. In this regard, the data obtained with FluoroUndecylTPP are particularly interesting as theIg~a tom can be readily replaced by the PET-active "F, and the uptake of FluoroUndecylTPP responded to the extent of mitochondnal polarisation in vivo. 15 Exam~le 4 - Use of 18F-UndecvlTPP to visualise tumours and mitochondnal damage in animal models An irnmunocompromised mouse model is used into which tumours of various sizes have been implanted tumours. 18F-UndecylTPP is then administered to the mice 20 and the turnours visualised using PET. A heart or kidney ischaemia reperfusion model is also used show that 18FUndecylTPP may be used to visualise damaged mitochondria within tissues. 25 Finally, models of damage to the blood-brain barrier are used to investigate whether the probes of the invention are taken up in to the brain following this damage, thusindicating whether the blood brain barrier has been compromised. Materials and Methods Chemical syntheses [ 3 ~iodid]e (6~0 Ci~/mmo~l) wa~s fro m American Radiolabeled Chemicals. [ " ] ~ i t o ~a nd ['HID~C~~TprPePpa rations were synthesised and HPLC-purified to >97% radiopurity. To synthesise 1 1 -fluoroundecyltriphenylphosphonium mesylate (FluoroUndecylTPP) a mixture of 11-bromoundecanol (752.1 mg 2.99 mrnol), 5 tetrabutylarnmonium fluoride (2.34 g 7.21 mmol) and H20 (162 pL) in a Kimax tube (15 mm x 150 rnrn), flushed with argon and sealed with a screw cap was stirred at 80°C for 1 h. The reaction was allowed to cool slightly and extracted into pentane (25 mL). The organic layer was washed with H20 (30 mL) three times, dried over MgS04, filtered and concentrated in vacuo to give 1 1-fluoro-1- 10 undecanol as a slightly brown oil (41 6 mg 2.19 mmol 73%) containing -7% 1 1 - fluoroundec-1-ene from 'H NMR integration of the -CH2F (4.5) vs =CH2 (4.9-5) resonances. I H NMR 4.46 (2H, d,t J=47.3, 6.2 Hz, -CH2F), 3.66 (2H, t, J=6.2 Hz, - CHzOH), 1.2-1.8 (18H, m) ppm. 19F NMR -21 8.5 (t,t J=47.5, 24.6 Hz) ppm. A solution of crude 1 1 -fluoro- 1-undecanol (333 mg, 1.75 mmol), triethylamine (35 1 15 mg, 3.47 mmol, 484 pL, 2 equiv.) in dry CH2C12 (10 mL) was stirred at -lO°C for 10 min. A solution of methane sulfonyl chloride (230 mg, 2.02 mrnol, 156 pL) in dry CH2C12 (1 mL) was then added while maintaining a temperature