A fast and effective determination of the biodistribution and subcellular localization of fluorescent immunoliposomes in freshly excised animal organs

In most preclinical biodistribution studies, researchers generally assume that the accumulation of drugs or fluorescent probes in the liver or kidneys portray their degradation and excretion via the hepatobiliary or urinary routes [12, 23]. In such studies organs are analyzed at a single time point (e.g. 6 h or 8 h) post probe application. However, vital information on the subcellular distribution and possible damage to the organs due to long retentions remain undocumented in such studies. The underlying results demonstrates the importance of monitoring the biodistribution at many time points in order to make reliable conclusions on probe distribution, retention and their later elimination from the system. A longer retention of some contrast agents in several organs could result in adverse side effects [13, 14, 27] depending on the subcellular localization, since different tissue cells react differently to different agents. Thus, it is important to know which subcellular factors are responsible for the retention of probes in different organs, for example excretory organs such as the liver, kidney and also sensitive organs not involved in biodegradation and excretion, such as the lungs, brain and heart. A longer retention in the liver could be due to a slower degradation, or unwanted affinity of the probes to molecular structures in the organs as seen with the mEnd-IL. Based on the contrast agent or therapeutic drug in question, this may cause adverse side effects. For example prolonged retention in the liver may implicate adverse effects of the probes on the kuppfer cells, or sinusoids of the liver or on the secretion to the bile. Likewise, longer retentions in the kidney due to formation of unfilterable aggregates may pose damages with time. A major reason why many researches do not include microscopic validation of the subcellular localization of probes is lack of suitable equipment. Thus, we verified if the use of simple and easily accessible microscopy setups could enable this.

Using macroscopic NIRF imaging, we could determine the biodistribution of targeted immunoliposomes based on the first-pass effect post injection and also their retention based on molecular targeting to different target cells. Interesting differences were seen between the FAP-IL and the mEnd-IL. Whereas, the FAP-IL distributed predominantly based on the first pass effect and subsequent elimination, the mEnd-IL revealed longer durations of retention in vital organs such as the lungs, liver and kidneys. The first-pass effect of FAP-IL was especially characterized by fluorescence of lungs which disappeared before 6 h post injection. This was coupled with a preliminary fluorescence of the gall bladder, liver, duodenum and kidneys at 1 h p.i., which increased in these organs after 6 h, and subsequently decreased 24 h after injection. Furthermore, a gradual movement of the fluorescence from the duodenum towards the jejunum, ileum and colorectum with increasing time post application indicated probe elimination via feces. The relatively high fluorescence signals of both immunoliposomes seen in the stomach could not directly be explained. We postulated previously [25] that this fluorescence may result from pancreatic and partial bile release of probes into the stomach like in humans, or due to reflux from the duodenum. This is supported by the fact that the fluorescence increases with time post injection and is very high even at time points when there is no fluorescence in the duodenum. Although the liposomes are not pH sensitive, and microscopic images demonstrate the secretion of the liposomal DY-676-COOH and NBD-DOPE in the bile canaliculi of the liver, which implies their eventual delivery to the gall bladder and stomach as individual components and not intact liposomes, the low pH of the stomach could possibly influence the DY-676-COOH and play a role in the high NIRF detected here. Contrary to FAP-IL, the combination of first-pass effect and molecular targeting by the mEnd-IL caused longer retention of the probe in many organs including the lungs, liver and kidneys as seen by macroscopic imaging. This retention could be detected based on the different time points considered and exposes the relevance of this consideration in biodistribution studies. Besides considering several time points, it is also important to pinpoint the sub cells responsible for the probe retention.

We therefore implemented a simple fresh organ microscopy setup to validate this. Although tissue autofluorescence interferes with in vivo fluorescence imaging, it has been exploited in defining several organs / tissue structures and to distinguish pathological changes in diseased tissues [16]. Thus, cellular and tissue autofluorescence originating from mitochondria, lysozymes, lipo-pigments and pyridinic (NADPH), flavin coenzymes, collagen, elastin, hemoglobin and melanin are successfully exploited for diverse applications such as in endoscopic imaging [28] and intravital microscopy [29, 30]. These tissue fluorophores absorb and emit light at different wavelengths which lie beyond the near-infrared optical window (650 nm – 900 nm) [17]. Consequently, the fluorescence of the liposomal encapsulated NIRF dye, DY-676-COOH (abs/em max. 674 /699 nm) could be easily distinguished from tissue autofluorescence of freshly excised organs. Whereas the lungs of FAP-IL treated mice revealed no detectable fluorescence signals, the liver and kidneys revealed distinct liposomal fluorescence at the different time points investigated. The kidneys showed mild fluorescence of the blood vessels and tubules of the cortex at 1 h p.i. and predominantly in the pyramids and pelvis with increased duration post injection. This indicates a partial, but gradual elimination of the probes in urine [25]. It was shown previously that DY-676 is highly hydrophobic and hence preferentially eliminated by the hepatobiliary route [12]. Consistent with this, the FAP-IL based fluorescence was located predominantly in kuppfer cells of the liver at all the investigated time points. The liver kuppfer cells are responsible for the host defense. When toxic or foreign substances are recognized by the system they opsonize the foreign substances, making them recognizable by macrophages which engulf them in the blood and migrate to the liver. In the liver the macrophages (both infiltrating and resident) are called kuppfer cells [31]. While in the liver they degrade the foreign substances which can then be secreted to the bile for elimination to the duodenum as was evident in the macroscopic images. However, the probes can also reach the liver directly through blood circulation. This is achieved by the first pass effect as well as repeated circulation of long circulating probes such as the immunoliposomes used herein. Notably, the FAP-IL undergoes only circulation, phagocytic uptake and degradation due to the lack of targets in the mice used, since FAP is exclusively expressed in diseased but not healthy tissues [21].

Opposed to FAP-IL, the mEnd-IL was detected in the lung, brain, kidney and liver endothelia cells. The fact that mEnd-IL localized in these endothelial cells substantiates the stability of the PEGylated liposomes in the blood circulation and their selectivity for murine endoglin. Hence, they enter the liver and other organs as intact vesicles which then specifically bind and are taken up and degraded by the respective endothelial cells, releasing the encapsulated DY-676-COOH. The free DY-676-COOH can be taken up by phagocytic cells but not by other cell types (see also Fig. 2), and is more rapidly eliminated than the long circulating liposomes in vivo [24]. Thus, the predominant green fluorescence and co-localization of the green and red fluorescence of liposomal NBD-DOPE and DY-676-COOH in organs such as liver and kidneys at 1 h p.i. for example, is indicative of the accumulation of intact liposomes, which eventually get degraded to release / activate the encapsulated DY-676-COOH in these organs (see Fig. 5b and Fig. 6a). This highlights the role of the non-quenched green fluorescent phospholipid, NBD-DOPE in tracking the intact versus degraded liposomes as was demonstrated in time course experiments earlier [24, 25] and also herewith (Fig. 2b, 4 °C). Based on earlier cytoxicity studies [19], it is known that the encapsulated DY-676-COOH is not cytotoxic. Considering that a cytotoxic substance used in its place could exert damages on the liver, lungs and brain endothelial cells upon long retentions, our results expose the importance of including microscopy as part of a biodistribution study. Furthermore, our results show that this is possible on freshly resected tissue. Thus, exploiting the tissue autofluorescence helps detect morphological changes that may result from adverse effects of the applied probes. This is especially feasible when characterizing the biodistribution of contrast agents coupled to a fluorophore with absorption and emission maxima in the near infrared optical window. Nevertheless, fluorescent dyes with lower wavelengths can be detected if their concentrations are much higher than that of the tissue autofluorescence. In our images for example, we could detect the green fluorescent phospholipid, NBD-DOPE (abs / em.: 480 nm / 530 nm), due to its high concentration and resulting strong signals which outshined the background autofluorescence. We therefore strongly recommend using this method when analyzing fluorescent probes with spectroscopic properties lying beyond the tissue autofluorescence range (650-900 nm). However, if the concentration of a probe is very high it could be detected similar to the NBD-DOPE used herein. In this case a comparison with a second dye would be of advantage, to avoid false interpretation.