Modifying the Siderophore Triacetylfusarinine C for Molecular Imaging of Fungal Infection

Purpose Aspergillus fumigatus produces the siderophore triacetylfusarinine C (TAFC) for iron acquisition which is essential for its virulence. Therefore, TAFC is a specific marker for invasive aspergillosis. We have shown previously that positron emission tomography (PET) imaging with [68Ga]TAFC exhibited excellent targeting properties in an A. fumigatus rat infection model. In this study, we aimed to prepare TAFC analogs modifying fusarinine C (FSC) by acylation with different carbon chain lengths as well as with charged substituents and investigated the influence of introduced substituents on preservation of TAFC characteristics in vitro and in vivo. Procedures Fifteen TAFC derivatives were prepared and labeled with gallium-68. In vitro uptake assays were carried out in A. fumigatus under iron-replete as well as iron-depleted conditions and distribution coefficient was determined. Based on these assays, three compounds, [68Ga]tripropanoyl(FSC) ([68Ga]TPFC), [68Ga]diacetylbutanoyl(FSC) ([68Ga]DABuFC), and [68Ga]trisuccinyl(FSC) ([68Ga]FSC(suc)3), with high, medium, and low in vitro uptake in fungal cultures, were selected for further evaluation. Stability and protein binding were evaluated and in vivo imaging performed in the A. fumigatus rat infection model. Results In vitro uptake studies using A. fumigatus revealed specific uptake of mono- and trisubstituted TAFC derivatives at RT. Lipophilicities as expressed by logD were 0.34 to − 3.80. The selected compounds displayed low protein binding and were stable in PBS and serum. Biodistribution and image contrast in PET/X-ray computed tomography of [68Ga]TPFC and [68Ga]DABuFC were comparable to [68Ga]TAFC, whereas no uptake in the infected region was observed with [68Ga]FSC(suc)3. Conclusions Our studies show the possibility to modify TAFC without losing its properties and specific recognition by A. fumigatus. This opens also new ways for multimodality imaging or theranostics of fungal infection by introducing functionalities such as fluorescent dyes or antifungal moieties.


Introduction
Due to the lack of sensitivity and specificity of current diagnostic methods, invasive fungal diseases are a major cause of high mortality rates (up to 50 % [1,2]) and elevated healthcare costs in Europe [3]. Aspergillus fumigatus is the major cause for pulmonary fungal infections in immunocompromised patients including also transplant recipients and patients undergoing aggressive anti-cancer chemotherapy [4]. Early diagnosis of invasive pulmonary aspergillosis (IPA) is a key to improve survival rate. Various clinical tests and techniques as well as X-ray computed tomography (CT) have shown unsuccessful diagnosis in terms of specificity and sensitivity. Scintigraphic techniques such as single photon emission computer tomography (SPECT) and positron emission tomography (PET) have been applied for imaging fungal infections using non-specific radiotracers including [ 99m Tc]leucocytes, [ 99m Tc]peptides, [ 99m Tc]anti-granulocyte antibody, [ 67 Ga]citrate (inflammation radiotracer), and even 2-deoxy-2-[ 18 F]fluoro-D-glucose ([ 18 F]FDG) [5]. These tracers have revealed only suboptimal characteristics for fungal detection.
Iron is an essential nutrient for growth and virulence of pathogenic microorganisms [6]. Initiation of infection highly depends on the ability of microorganisms to use hostcomplexed iron. A. fumigatus lacks specific uptake systems for host iron sources and uses two high affinity iron uptake systems: reductive iron assimilation (Fe 2+ specific) and siderophore-assisted iron acquisition (Fe 3+ specific), but only the latter system is essential for virulence of A. fumigatus [7]. Siderophores, iron-sequestering compounds, are low molecular weight chelators with high affinity to iron (formation constants of 10 20 -10 50 ) produced by fungi, bacteria, and some plants. A. fumigatus produces two hydroxamate-type siderophores, namely fusarinine C (FSC) and its N-acetylated derivative triacetylfusarinine not interact with human cellular systems. These characteristics promote targeted diagnosis of A. fumigatus infection via the MirB transporter using TAFC as the vector molecule.
PET has been widely used for molecular imaging due to the high-intensity images, limitless depth of penetration, and providing quantitative data. Among positron emitter isotopes, gallium-68  is the most attractive nuclide for radiolabeling of siderophores [15]. Ga 3+ has an equal charge and a comparable radius to ferric ion (Fe 3+ ) allowing displacement of iron by . Moreover, gallium-68 has a half-life of 68 min exhibiting very low radiation burden to patients. Furthermore, it can be obtained from 68 Ge/ 68 Ga generator systems, therefore easy, accessible, simple in use, and relatively inexpensive.
In our previous works, we have demonstrated that different siderophores can be labeled with 68 Ga [16]. We have also shown that [ 68  are able to detect A. fumigatus infection in a rat pulmonary aspergillosis model using PET imaging [17,18] and [ 68 Ga]TAFC is more specific to A. fumigatus in vitro [19].
[ 68 Ga]TAFC exhibits very rapid renal elimination, resulting in a short-term bioavailability. Chemically modifying TAFC would potentially allow adjusting pharmacokinetic properties and introducing functionalities such as fluorescent dyes or even therapeutic moieties. In this paper, we describe chemical modifications of TAFC (see Fig. 1), [ 68 Ga]-radiolabeling, and in vitro characterization of respective gallium complexes. We have also investigated the influence of TAFC modifications on the recognition by the specific siderophore system of A. fumigatus and report the biodistribution as well as PET/CT images of selected compounds in comparison to [ 68 Ga]TAFC.

Chemicals
All commercially available chemicals were obtained as analytical grade and used without further purification. The 68 Ge/ 68 Ga generator (IGG100) was purchased from Eckert & Ziegler Strahlenund Medizintechnik AG (Berlin, Germany) with a nominal activity of 1850 MBq and was eluted with 0.1 M HCl solution (Rotem Industries Ltd., Beer-Sheva, Israel) using the fractionated elution approach.

Fungal Strains and Growth Conditions
For in vitro uptake studies, A. fumigatus wildtype (ATCC46645) was used. Fungal strain was cultured at 37 °C in Aspergillus minimal medium (AMM) containing 1 % (m/v) glucose as carbon source, 20 mM glutamine as nitrogen source, salts, and trace elements [20]. Ironsufficient media were prepared by adding FeSO 4 to a final concentration of 30 μM. For irondeficient conditions, supplementation of iron was omitted. Iron starvation was verified by the detection of extracellular siderophore production which is suppressed by iron supplementation.
For siderophore utilization growth assays, A. fumigatus mutant strain ΔsidA/ΔftrA [7], which can grow only in the presence of externally added iron-siderophores or at high iron concentrations, was used. The fungal strain was inoculated in iron-deficient culture as described above.

Preparation of [Fe]Fusarinine C from A. fumigatus
[Fe]fusarinine C ([Fe]FSC) was extracted from the A. fumigatus mutant strain ΔsidG, which lacks the conversion of FSC to TAFC [9]. Fungi were cultured under iron-depleted condition as described by Schrettl, et al. [9] with slightly changed incubation time and extraction method as previously published [21].

Synthesis
The syntheses of TAFC derivatives are illustrated in Fig. 1 3 ). Iron-free siderophores were finally accomplished by demetalation using disodium-ethylenediaminetetraacetic acid (see ESM).

In Vitro Characterization
Distribution Coefficient (logD)-[ 68 Ga]siderophore (4 μM) in 500 μl phosphatebuffered saline (PBS), pH 7.4, was added to 500 μl octanol. The mixture was vortexed at 1400 rpm for 15 min and subsequently centrifuged at 2000 rcf for 2 min. Aliquots of aqueous and octanol phases were collected and measured in a gamma counter (a 2480 Automatic Gamma Counter Wizard 2 3″; PerkinElmer, Waltham, MA, USA). The logD values were calculated from the obtained data (n = 5).  3 were studied by incubating radiotracers in human serum at 37 °C for 30, 60, and 120 min. At selected time points, 25 μl of serum aliquots was applied on size exclusion spin columns MicroSpin™ G-50 (GE Healthcare, Vienna, Austria). Columns were centrifuged at 2000 rcf for 2 min to separate protein bound radiotracer (eluate) from nonprotein bound radiotracer (column). Protein binding ability was determined by measuring activities of column and eluate in the gamma counter. The results were displayed as % protein bound to total activity applied (n = 3).

In Vitro Uptake Assay in A. fumigatus
Uptake assays were performed in either iron-deficient or iron-sufficient A. fumigatus wildtype cultures. In a 96-well MultiScreen Filter Plates HTS (1 μm glass fiber filter, Merck Millipore, Darmstadt, Germany), 180 μl of A. fumigatus suspended in iron-replete or irondepleted medium was added. [ 68 Ga]siderophores (approx. 1 × 10 5 cpm, 80 nM) were incubated with either PBS pH 7.4 (total series) or blocking solution (10 μM [Fe]TAFC) (non-specific series) at RT for 45 min in quadriplicates. The incubation was interrupted by suction of the medium and the plate was washed twice with ice-cold TRIS-buffered saline (10 mM, pH 7.3). The filters were isolated and the remaining activities were measured in the gamma counter. The uptake was calculated as % uptake of total activity added; each assay was repeated three times. To normalize uptake variations between different fungal cultures, the uptake of each [ 68 Ga]siderophore was expressed as the ratio to [ 68 Ga]TAFC uptake from the same experiment. The mean values of three different assays were used to quantitatively compare the difference in uptake between the compounds.

In Vitro Uptake Assay in A. terreus
To confirm the specificity of uptake of TAFC derivatives by A. fumigatus via the MirB transporter, A. terreus, which lacks the MirB transporter, was used for in vitro uptake assay. A. terreus was grown in either iron-deficient or iron-sufficient cultures. Uptake studies were performed as described for A. fumigatus and expressed as % uptake of total activity added.

Statistical Analysis
The in vitro uptake data of each [ 68 Ga]siderophore and [ 68 Ga]TAFC from the same experiment were compared using unpaired t test (level of significance, P < 0.01). Analysis was performed using Microsoft Office Excel 2010 program. Kaeopookum  . Yields of both iron-containing and iron-free siderophores are summarized in Table 1.

Distribution Coefficient, Protein Binding, and Stability Studies
Lipophilicity as expressed by the logD value varied between 0.34 and − 3.80 indicating a wide variation of lipophilicity (see Table 1). All compounds were stable after radiolabeling for > 2 h. Protein binding and stability over time for four selected compounds ([ 68 Ga]-TAFC, -TPFC, -DABuFC, -FSC(suc) 3 ) are shown in Table 2.

In Vitro Uptake of [ 68 Ga]Siderophores
The certain fluctuation of % uptake of [ 68 Fig. 1 in ESM).

Utilization of Siderophore by A. fumigatus
A double gene deletion mutant strain A. fumigatus, ΔsidA/ΔftrA, which lacks siderophore biosynthesis and reductive iron assimilation, is unable to grow in the absence of siderophore or Fe 2+ concentration above 2 mM. Therefore, this strain was used to investigate the utilization of external supplied [Fe]siderophores for its growth. This assay showed the utilization of all studied siderophores by A. fumigatus dependent on the concentration of the different [Fe]siderophores. Nevertheless, no significant difference in growth was observed at high concentration (50-100 μM), except for FSC(suc) 3 that showed a sporulation defect at 100 μM but no growth at 50 μM (Fig. 3).   3 displayed similar radioactivity accumulation in almost all tissues as illustrated in Fig. 4. Slightly lower retention was observed only in the blood, spleen, and pancreas for [ 68 Ga]DABuFC and in the intestine and kidneys for [ 68 Ga]FSC(suc) 3 . Higher uptake was found in intestine for [ 68 3 showed uptake in infected lung regions (0.3 %ID/g) comparable to non-infected tissue (see Fig. 5b).

Discussion
[ 68 Ga]TAFC has shown excellent targeting properties for A. fumigatus infections in vitro and in vivo [17,18], allowing molecular imaging of fungal infection using PET. TAFC is the main siderophore produced and excreted by A. fumigatus for iron utilization and growth. It is unknown which structural elements are required to achieve selective recognition of [Fe] complexes by MirB. Modification of TAFC may allow to particularly adjusting pharmacokinetics to optimize targeting of fungal infection in vivo, but also to introduce functionalities into the molecule for a variety of novel applications including theranostics of fungal infections, which was recently proposed for bacterial infections using multivalent siderophore-DOTAM conjugates [22]. We have previously shown that starting from FSC, the deacetylated variant of TAFC, selective substitution, and chemical modifications are possible. Mono-, di-, and trisubstituted FSC derivatives were prepared by acylation of the free amines [21,23,24]. We utilized this approach to introduce targeting vectors and even fluorescent dyes as basis for the preparation of novel Ga-68 and Zr-89 radiopharmaceuticals for oncological applications. In this study, we aimed to reveal some structure-activity relationships of fungal recognition by selective acylation of one, two, or three amine groups of the FSC molecule. We chose to introduce longer carbon substituents (propanoyl, butanoyl, benzoyl) instead of acetyl as well as positively and negatively charged moieties by using glycyl and succinyl substituents, respectively. Besides mono-substitution of the acetyl, we also prepared selected di-and trisubstituted derivatives. All compounds could be prepared straight forward in reasonable yields by starting from the deacetylated compounds, i.e., DAFC for mono-, MAFC for di-, and FSC for trisubstituents, protected by using the respective [Fe] complexes. After removal of iron, all compounds could be radiolabeled with gallium-68 in quantitative yields. Only for a couple of derivatives, radiolabeling was less efficient for unknown reasons, which required purification before further use. The tested compounds showed a wide variety of lipophilicity and various charges depending on the substituents.
In initial uptake assays, most Ga-68 compounds showed a specific uptake by A. fumigatus, when grown under iron starvation, indicating recognition by siderophore transporters. There was no indication for any changes in transporter recognition, except for FSC, which is known to be utilized not only by MirB. However, the efficiency in uptake was highly variable. Highest uptake was found for trisubstituted of propanoyl (tripropanoyl(FSC)) while tributanoyl(FSC) revealed a lower uptake which could be related to its increased lipophilicity. Comparable uptakes were observed in mono-substituted, i.e., DAFC derivatives, with a slight decrease in uptake with higher lipophilicity (e.g., DABzFC) or when a negative charge was introduced (DAFC(suc)). TAFC derivatives with di-and trisubstitutions with charged moieties resulted in loss of specific uptake by A. fumigatus. These uptake trends were confirmed by growth assays using an A. fumigatus mutant lacking siderophore production and reductive iron assimilation. This mutant only grows in the presence of external siderophores. Also in this assay, mono-substituted TAFC compounds showed comparable or only slightly impaired growth promotion as compared to TAFC, whereas with additional substitution, growth promotion was significantly impaired. This indicates that by modifying TAFC, not only uptake can be preserved but also A. fumigatus can utilize these modified siderophores as source of iron.
In vivo behavior was tested for three selected compounds ([ 68 Ga]DABuFC due to their higher lipophilicities. Therefore, these siderophores are not more efficient than TAFC itself in terms of imaging contrast but evidence the possibility of modifying TAFC without losing specific recognition by A. fumigatus.

Conclusions
Overall, we could show in this paper that modification of TAFC is possible without losing major properties of this promising siderophore for molecular imaging of A. fumigatus infections. Selective replacement of one acetyl group results in compounds that can be easily radiolabeled with gallium-68 and show high uptake by A. fumigatus and specific recognition by the MirB siderophore transporter as well as high stability both in vitro and in vivo. This knowledge enables us to design compounds with varying pharmacokinetics potentially improving targeting properties, but also to introduce functionalities such as fluorescent dyes to combine optical imaging with PET or to introduce moieties with antifungal activity opening new ways for theranostics also in the field of infectious diseases.

Supplementary Material
Refer to Web version on PubMed Central for supplementary material.  In vitro uptake of [ 68 Ga]siderophores in A. fumigatus under iron-starvation condition at 37 °C for 45 min. Normalized uptakes are expressed as ratio to [ 68 Ga]TAFC uptake from the same experiment (n = 3). Uptake under iron-starvation condition with TAFC blocking and under iron-replete condition, which were used as controls, was < 0.5 % of total activity added (data not shown).