Although radioactive iodinated derivatives have been studied for single photon emission computed tomography (SPECT) and are still of particular interest today [2], compounds labelled with positron emitters, suitable for the positron emission tomography (PET), were the main subjects of study in human functional imaging for the last twenty years. This relies on the high sensitivity and the quantitative capability of PET [3] and on the characteristics of carbon-11, an isotope of a naturally existing element in biological molecule, or fluorine-18, a bioisostere of naturally existing chemical groups, for example the OH group [4,5].

L-[S-methyl-¹¹C]-methionine has been used to generate interesting clinical results for the diagnosis of brain tumours [6]. Unfortunately, due to the very short half-life of carbon-11 (20.4 minutes), the use of tracers labelled with this radionuclide is logistically only possible on sites equipped with their own cyclotron. Even in this case, the completion of a high number of patient examinations with the same radiopharmaceutical batch is technically difficult except if several PET scanners are simultaneously accessible. This physical limitation therefore restricts the usefulness of carbon-11 labelled tracers in the study of biochemical processes to a time scale of maximum one hour.

With a half-life of 109.8 minutes, fluorine-18 presents better characteristics for the geographic expansion of radiopharmaceuticals supply. This capability has been practically proven worldwide with the use of the PET tracer 2-[¹⁸F]fluoro-deoxyglucose ([¹⁸F]FDG). Moreover, the longer half-life of fluorine-18 is more suitable for the study of processes with a time scale of more than one hour, such as brain protein synthesis by example.

The fluorinated aromatic amino acids 2-, 3- and 4- fluorophenylalanine, 2- and 3- fluorotyrosine (figure ​(figure1)1) have been described by geneticists as potential surrogates for canonical phenylalanine and tyrosine in an expanded repertoire of amino acids corresponding to a less restricted genetic code [7]. All of them have been previously been evaluated as [¹⁸F] labelled tracers for the quantitative assessment of cerebral protein synthesis in vivo [8]. 2-[¹⁸F]fluorophenylalanine and 2-[¹⁸F]fluorotyrosine have also been considered for application in tumour imaging [9].

Among the few fluorinated amino acids incorporated into proteins, 2-[¹⁸F]fluoro-L-tyrosine (2-[¹⁸F]Tyr) is the most promising. Its incorporation has been studied in mice at a low specific activity (10–20 GBq/mmol). This revealed a very simple metabolism as well as rapid incorporation into proteins [10].

All animal experiments were approved by the Ethical Committee at our institution. Chronic cannulations of rats (male, Sprague Dawley, 180–250 g) were performed in the abdominal aorta and in the vena cava under general anaesthesia (ketamine 35 mg/kg and medetomidin 0.35 mg/kg) following the procedure described by Lestage [14]. The animals were allowed to recover for a minimum of 4 days before being used for metabolism studies. Most of them were injected with the tracer within the first week after surgery. Some of them were kept up to 4 weeks before injection. Twenty hours before the injection of the tracer, the food was removed from the cage. The arterial line was used for blood sampling and the other for i.v. injection of the tracer. The two lines were flushed with saline a few minutes before use in order to check potency. 14 rats (230–410 g) were injected with 20–325 MBq of n-c-a 2-[¹⁸F]Tyr. The tracer was delivered as a bolus of maximum volume 0.75 ml in 15 s. Samples of blood were obtained via the arterial cannula during the first minutes after the injection and at different times (15, 30, 45, 60 and/or 90 minutes) thereafter. The line was flushed with saline between each sample to avoid any loss of potency and to ensure the homogeneity of the next sample. The animals (n = number of rats) were sacrificed after 30 (n = 2), 60 (n = 5), 90 (n = 3) or 120 (n = 4) minutes. After the final time point, samples of blood were taken and the major organs were dissected. The brain was dissected to obtain striatum, cortex and cerebellum.

Trichloroacetic acid (TCA) was used to precipitate the proteins from biological samples in order to assay the radioactive content associated with proteins and to perform a chromatographic analysis of the radioactive compounds in solution in the supernatant after centrifugation. The solution used for the precipitation was Na₂S₂O₅ 100 mg, NaEDTA.2H₂O 10 mg, TCA 20 g, dissolved in water up to 100 ml.

Radioactivity was measured in weighed fractions of plasma. For intermediate and final plasma samples, proteins were precipitated with TCA solution and separated by centrifugation (5°C, 5 minutes, 8000 RPM) before counting. The supernatant was analysed by HPLC. Radioactivity was measured in the organ samples and in one half of the three dissected parts of the brain. The second halves were sonicated with water (Sonics & Materials inc., Vibracell VXC-400). The proteins were precipitated with TCA solution and separated by centrifugation (5°C, 5 minutes, 8000 RPM) before counting. The supernatant was analysed by HPLC.

The metabolites analysis was performed by HPLC [11] on fractions of the supernatants obtained after precipitation and centrifugation. Column: Phenomenex Bondclone 10 μm C18 300 mm × 3.9 mm. Injection volume: 100 μl. Mobile phase: 80% of 0.1 M NaH₂PO₄, 0.1 mM EDTA, 2.6 mM octane sulfonic acid sodium salt and 20% methanol, pH 3.1. The flow rate was 1 ml/min from start to 8 minutes and was then increased in 30 seconds interval at 1.5 ml/min. Fractions of 30 seconds were collected for 16 minutes. Capacity factors: FDOPA: 0.8; 3-OMFD: 1.5; FTYR: 1.9; FDOPAC: 2.6; FDA: 3.3; FHVA: 5.0 (k' = t_r-t₀ /t₀, t_r: retention time of the solute, t₀: column dead time).

Different parameters, including amino acid transport and PSR, are needed to fully characterise a tumour and monitor the response to treatments [9]. An incorporated labelled amino acid has the advantage of potentially imaging, with a single injection, either early transport of amino acids into cells and the delayed incorporation into the proteins afterwards. Nevertheless, this may only be a virtual advantage for tumour imaging as some studies indicate that the major phenomenon leading to an increased uptake of amino acids in tumors is an enhancement of amino acid transport rather than an increased PSR. So, the uptake of L-[S-methyl-¹¹C]methionine in mice brain was not influenced by the use of a inhibitor for protein synthesis [16]. In the same conditions, the uptake of 2-[¹⁸F]Tyr was partially reduced. In another study evaluating 2-[¹⁸F]Tyr in patients with gliomas [17], pharmacokinetic modelling of the results indicated that the main difference between normal and tumorous tissues was a significantly increased transport rate constant from plasma to tissue. Thus, these two studies suggested that 2-[¹⁸F]Tyr could be used to assess the amino acid transport system. In this context, more recent developments have led to the evaluation of O-(2-[¹⁸F]Fluoroethyl)-L-tyrosine. This tracer, not incorporated into proteins, has the important advantage of being available with a relatively high radiochemical yield (40%) allowing the distribution of the product and its use to a larger extent. As a result of this availability, it has already been evaluated for different applications in clinical oncology [18]. In comparison, 2-[¹⁸F]Tyr suffers from a less efficient radiosynthesis and must thus still be better evaluated. In a small study for whole-body tumour imaging, the tracer was less sensitive than [¹⁸F]FDG for staging NSCLCs and lymphomas [19]. In another study focusing on primary brain tumors, 2-[¹⁸F]Tyr allowed a better tumor to background contrast in low-grade gliomas [20].

In addition to neuroendocrine activity, nucleic acid synthesis and gene transcription, protein synthesis has been described as a potential basic mechanism for brain synaptic plasticity [21]. Thus, incorporated amino acids may have further advantages for the study of various non tumourous processes. For example, the evaluation of PSR during brain development, sleep or memory consolidation is only possible with an incorporated tracer. Thus, beside uses in oncology, imaging PSR with amino acids presents a real interest for functional research, as well as potential applications in pathologies related to neurological physiological processes. The use of a single pharmaceutical agent adapted for several applications is a good approach from an economical point of view. As markedly demonstrated by [¹⁸F]FDG, this is especially true for PET pharmaceuticals, whose constraints concerning logistics, radioprotection and quality assurance, are considerable. All these arguments are in favour of the development of a multi-purpose [¹⁸F] fluorinated incorporated amino acid.

2-[¹⁸F]Tyr was identified early by Coenen and colleagues [10] as a promising tracer for PSR assessment in an extensive study in mice. At a low specific activity, its acceptance by the amino acid-tRNA synthetase has been proven by phenolic extraction of the tRNA bound fraction. The rapid and general incorporation into proteins has been confirmed by SDS gel electrophoresis. A very small proportion of metabolites was detected in striatum tissue and the authors raised the possibility of a very small entrance of the tracer to the catecholamine metabolic pathway [22].

The present study was performed using n-c-a 2-[¹⁸F]fluoro-L-tyrosine allowing the injection of a high amount of radioactivity (up to 325 MBq) supported by a low amount of non radioactive equivalent (less than 10^-2 μmole). The higher specific activity used in this study may not be of major physiological consequence since the concentration of competing amino acids in the plasma is high under fasting conditions. On the other hand, one can not exclude the existence of a biochemical mechanism (e.g. transport or metabolism) truly specific for the fluorinated analogue, in which the non fluorinated natural amino acid could not be involved at all. Using a higher specific activity increases the sensitivity for the detection of these potential mechanisms. The results of the present study can thus be interestingly compared to the results published for mice [10].

From the biodistribution data presented in table 1 [see Additional file 1], we can observe that the fluorinated analogue rapidly passes the blood-brain-barrier as the uptake in brain already reached a high value at only 30 minutes post injection. The absolute uptake into rat brain tissue (~0.15%IA/g) was lower than the value reported for mice (~1.3%IA/g at 40 min.; ~2%IA/g at 60 min.), as may be expected due to species differences (including size, weight and distribution volume). On the other hand, the total radioactivity of the plasma increased with time after the initial peak of injection, indicating the elimination of radioactive compounds in the blood. The uptake of radioactivity in bone increased with time and to a greater degree than in other organs such as muscle, heart or lung, indicating the probable liberation of [¹⁸F]F^-during the metabolism of the tracer. The comparison of uptake in the different brain regions shows a homogeneous distribution of activity (table 2 [see Additional file 2]).

In all studied tissues, the radioactivity associated with proteins increased from 30 minutes to 120 minutes (table 2 [see Additional file 2]). Expressed as the percentage of the total activity in the tissue, this fraction at 60 minutes in rat cerebellum and cortex (respectively 84% and 85%) matches the reported value for mouse cortex (84%). In rats, the incorporation in non dopaminergic rich regions reaches almost 90% of the total activity at 120 minutes. On the other hand, the observed incorporation of tracer in the striatum may be less than in other brain regions (66% at 60 min.). This tendency should be confirmed by a more extensive study including a higher number of subjects and allowing accurate statistical evaluation of the observations.

The percentage of unmetabolised 2-[¹⁸F]Tyr in the supernatant after protein precipitation decreased with time in all tissues, indicating that the tracer enters another pathway that simply the incorporation into proteins. This percentage of unchanged 2-[¹⁸F]Tyr was smaller in the striatum at all time points, indicating that a fraction probably enters the catecholamine metabolic chain. These observations are compatible with the affirmation that 2-[¹⁸F]Tyr is a substrate for the enzyme tyrosine hydroxylase. A strict identification and a precise quantification of the metabolites require an improved methodology to concentrate the solutes before HPLC analysis. Nevertheless, radioactive compounds corresponding to products eluted before 2-[¹⁸F]Tyr (as FDOPA does) and after (as FDA and FHVA do), were observed for the striatum.