Mannose-Receptor Targeting with Liposomes: A good idea…and an old one!
We are continuing to review publications related to mannosylated liposomes. This discussion will be updated with any pertinent information as we find it or whenever new information is published.
Mannose receptor binding is one of the oldest proposed mechanisms for liposome targeting through surface modification of the liposomes. The increasing number of both cell-surface-associated and soluble mannose-binding proteins that have since been discovered reveal the complexity of the mechanism(s) involved in enhanced uptake of mannosylated liposomes . Yet, several publications present convincing data on the ability of liposomes with various different conformations of surface bound mannose molecules to target mannose-receptor positive cells both in vitro and in vivo. Unfortunately, most of these proven formulations have not been applied to selective macrophage depletion by encapsulation of clodronate in these liposomes. The clodronate liposome field, in particular, has been plagued by the introduction of a liposome formulation prepared in the presence of p-aminophenyl mannose, but without covalently coupling this modified sugar to the liposome surface. The profusely cited reference to this liposome preparation claimed to target mannose-receptor positive cells in the brain as well as deliver liposome cargo across the blood-brain barrier. We find the data in this paper to be very weak and cannot confirm that the formulation was ever appropriately characterized with respect to the presence of mannose capable of targeting the liposome to mannose-binding proteins. While some stronger data suggesting that this formulation may target mannose-receptor positive cells has been published, others find no evidence of targeting by this formulation. We conclude that studies specifically designed to characterize and assess the targeting ability of this formulation are required to justify the use of this formulation in targeting mannose-receptor positive phagocytes. Until these studies are completed and published, we recommend that formulations containing covalently-anchored mannose and demonstrating the ability to selectively bind mannose receptors be utilized when experimental design calls for mannose receptor targeting.
Many different mannosylated lipids have been synthesized which effectively display mannose residues on the surface of liposomes suitable for binding to mannose receptors as judged by Con A agglutination and competitive inhibition of mannose receptor binding by mannose or mannose-containing macromolecules. These demonstrations, however, are more difficult to interpret in vivo because of the multiple mechanisms by which phagocyte uptake of mannosylated liposomes is enhanced. The structures of these mannosylated lipids are shown on this page along with a brief discussion of the published characterization and behavior of liposomes containing these derivatives. While the rationale for the design, synthesis and biophysical properties of these molecules may not seem directly applicable to clodronate liposome studies, there is a great deal of useful information in these papers on the rational design of surface-modified liposomes which will be invaluable when interpreting mannosylated clodronate liposome depletion data as well as when attempting to deplete specific mannose-receptor-positive phagocyte subsets.
Despite the large number of publications by those attempting to deplete mannose-receptor-positive phagocyte populations with liposomal clodronate prepared in the presence of p-aminophenyl mannose, combined with as many, or more, papers from those designing mannosylated liposomes for targeting cells displaying macrophage receptors, thus far, we have found only one paper describing mannosylated clodronate liposomes that were confirmed to target mannose receptors in vitro. Therefore, there is a great deal of work to be done with regard to evaluating the use and advantages of mannosylated clodronate liposomes.
We are awaiting publication of data from one group to whom we supplied covalently-coupled mannosylated liposomes; they have indicated achieving successful targeting and depletion of a mannose-receptor positive phagocyte subpopulation. By compiling the information on this page, it is our intention to equip everyone with the basic tools necessary for successful, unequivocal mannose receptor targeting using mannosylated clodronate liposomes.
Multiple studies have shown that modifying the surface of liposomes with mannose  increases the uptake of these liposomes in phagocytes bearing a mannose receptor mechanism [2, 3, 4, 15] as far back as the mid-1970’s . Obviously, this is not a new idea although Juliano’s first efforts actually produced a liposome with the lectin (carbohydrate-binding protein) on the surface rather than the sugar. While we are focusing on mannosylation, many other publications describe liposome targeting using other carbohydrates. The significant amount of data on mannose and other carbohydrate targeting should not be confused with being able to selectively target macrophage subpopulations which may express mannose receptors. Mannosylated-liposome uptake is enhanced by other mechanisms unrelated to mannose receptor binding, such as recognition by mannose-binding protein (MBP or MBL) , which activates complement and initiates uptake by complement receptor mediated phagocytosis [5, 11]. While all liposomes activate complement and bind other serum opsonins to some extent, certain types of liposomes (i.e. 100 nm DSPC/cholesterol liposomes) bind much less complement than others, thus exhibit extended circulation times post-i.v. injection in vivo. However, when mannose or other carbohydrates were covalently attached to the liposome surfaces, they are rapidly cleared from the bloodstream . This has even been shown for so-called “stealth” liposomes, known for their ability to avoid immediate phagocyte uptake .
Mannose is one of the carbohydrate components of many bacterial and viral cell surfaces, therefore the ever-efficient, highly redundant immune system has evolved multiple mechanisms for identifying pathogens based on mannose recognition. The animal and plant kingdoms likewise utilize carbohydrate recognition signalling mechanisms including mannose residues. Many publications evaluate other carbohydrates as targeting mechanisms for various cell types, however mannose targeting to phagocytes appears to be one of the more specific mechanisms identified to date . Mammalian cell surface identification molecules based on mannose binding, such as the ICAM family of leukocyte adhesion molecules, target the SIGN family of mannose receptors to accomplish self-recognition in vivo [12, 13, 14]. Several other mannose-binding molecules have been identified, but not all of them have been connected to their biologically relevant ligands and functions . There are also a large bodies of work focusing on using mannosylated liposomes as an adjuvant, for non-specific immunostimulation and for nucleic acid delivery, the mechanisms all of which are somewhat outside the focus of this discussion [15, 16, 17, 18, 19]. However, these applications have provided a multitude of methods for synthesizing mannosylated-lipid conjugates that can be included into mannosylated-liposomal clodronate formulations.
Mannose- and Mannose-Liposome-Binding Serum Components
The abundance of both cell associated and soluble mannose receptors in serum contributes to the lack of specificity of mannose targeting . As mentioned earlier, mannose-binding protein, present in mammalian serum, is very effective at complement activation when MBP is bound to mannose. Several studies describe the soluble mannose receptor (sMR) which is the extracellular portion of the macrophage mannose receptor (MR) released by macrophages for transferring macrophage-processed antigens to antigen-presenting cells (APC) . While MBP requires the association of several subunits, each possessing a single carbohydrate binding domain (CRD) for high affinity binding, sMR possesses 8 CRDs on a single molecule. Therefore sMR has no requirement for self-association in serum before binding occurs as does MBP.
MBP is closely related to surfactant proteins A and D (SP-A, SP-D) which also bind mannose in the lung. Therefore, intrapulmonary administration of mannosylated liposomes will also result in binding of mannose-specific proteins that will compete with MR+ alveolar macrophages and other MR+ pulmonary phagocytes for the mannose on the liposome surface . The receptors and functions of SP-A and SP-D have been the subject of conflicting reports and are still not clearly understood , furthermore these soluble mannose receptors are found in several locations outside the lung  making it somewhat more difficult to take into account how these lectins will affect mannosylated liposomes.
Complement component, C3 , has been shown to covalently bind mannose during its activated state, although this has not been demonstrated in vivo . Antibodies bind both specifically (anti-phospholipid antibodies) and non-specifically to liposomes in vivo . While not directly addressed in any of these publications, it seems that antibodies generated from previous exposure to various mannose-displaying pathogens may recognize some mannosylated liposomes due to their similar surface properties. We highlight these soluble mannose receptors, because when designing a targeted liposome, as with any other liposome for use in vivo, the first consideration is how the liposome will interact with the biological milieu before it reaches its potential target cell.
In summary, the propensity for serum components binding to liposomes is enhanced by the presence of both mannose-specific and non-specific components when the liposomes are surface-modified by mannose. Most of this enhanced binding correlates to enhanced uptake of liposomes in the liver since removing foreign particles from the bloodstream is one of the liver’s primary functions. In fact, Ghosh, et. al. and other groups have demonstrated that mannosylated liposomes are preferentially phagocytosed by non-parenchymal liver cells, including Kupffer cells, while galactosylated liposomes tend to be associated with hepatocytes .
Cell-Surface Mannose Receptors
When developing a mannose-targeting rationale, the macrophage mannose receptor (formerly MMR, now MR or CD206) is usually the targeting goal. This receptor was originally discovered and characterized from macrophages, but this receptor has since been found on dendritic cells, certain endothelial cells and as a non-cell-bound soluble form which has been shown to migrate from macrophages to antigen-presenting cells (APC) as discussed above [43, 44]. A structurally and functionally similar receptor, Endo-180 , was first identified on fibroblasts, but has now been identified on macrophages and endothelial cells in all tissues. Endo-180 appears to function in regulation of extracellular matrix catabolism and structuring. DEC-205 , a mannose receptor first identified in maturing dendritic cells and one of the receptors which binds the soluble from of phospholipase A2, M-PLA 2 R , also specifically binds mannose-containing molecules. This family of cell surface mannose receptors all operate by the clathrin-mediated (coated-pit) endocytic mechanism which would limit the size of the liposome that could be internalized to a few hundred microns. However, MR itself is unique in that it also initiates actin-mediated phagocytosis although some speculate that MR requires cooperation with other receptors to accomplish phagocytosis. DEC-205 and M-PLA 2 R are not likely to play a significant role in mannosylated liposome targeting, but their ability to bind specific mannose-containing molecules may recruit mannosylated liposomes to the cell surface increasing the likelihood that the liposome will come into contact with the phagocytic mannose-receptors.
In addition to MR and Fc (antibody) receptors which allow dendritic cells to phagocytose pathogens, a distinctly different receptor which recognized mannose was first identified on these cells. This receptor was found to mediate the interaction between dendritic cells and T cells which allows the transfer of antigenic peptides produced by dendritic cells (from ingested pathogens) to the T cells (for antibody production) through the membrane-associated MHC shuttle. The receptor ligand on T cells was identified as ICAM-3 and the receptor was designated DC-SIGN as it had only been identified on dendritic cells [13, 14]. Of course, related receptors ( SIGNRs ) were later identified in many tissues including splenic red pulp and peritoneal macrophages ( SIGNR1 ). Mannosylated-liposome targeting to this receptor demonstrates the requirement for complement receptor (CR3) participation in the phagocytosis of the liposomes suggesting that SIGN receptors do require cooperative support from other phagocytic receptors as has been speculated for MR . Langerin is another SIGN-like mannose-binding receptor speculated to be a target for mannosylated liposomes.
Clearly, a variety of potential mannose-targeting sites exist on the surface of many cell types including those susceptible to depletion by liposomal clodronate. Although all of these sites may not initiate phagocytosis of clodronate liposomes, mannosylated clodronate liposomes may bind to any of these sites ensuring that the liposomes are in close proximity to cell-surface phagocytic receptors thus increasing the probability of liposome phagocytosis.
Is there a Rationale for using Mannosylated Clodronate Liposomes?
Given that mannosylated liposomes are expected to demonstrate enhanced uptake by all (not only those with mannose receptors) accessible phagocytic cells due to their increased level of opsonization, it is difficult to predict if the pattern of macrophage depletion will be substantially different in vivo. Buiting, et. al. report no significant differences in the biodistribution of clodronate liposomes of differing lipid compositions including the p-aminophenyl mannose formulation as well as a colvalently linked mannosyl-DOPE conjugate . With regard to the covalently attached mannose liposomes, DOPE is a non-bilayer forming lipid which is known to be disruptive to liposomes under many conditions and is suspected to have demonstrated similar disruptive behavior as described for Weissig, et. al. below. Successful targeting with the PE-linked mannose studies employ DPPE or other bilayer-preferring PE species. Additionally, this mannosyl-DOPE conjugate does not include a flexible spacer which extends the mannose away from the liposome surface, the importance of which is also discussed below. Furthermore, most mannosylated-lipid studies report that two or more mannose residues per lipid are required for optimal mannose-receptor binding while this conjugate has a single mannose molecule per lipid. Taken all together, we believe that this covalently-linked mannose liposome formulation was not optimized for successful targeting. Yet, many mannose-conjugated lipid derivatives (shown below) do demonstrate selective uptake in vitro and different biodistribution patterns in vivo , and there is a study showing that mannosylated (ManDog) clodronate liposomes are effective for depleting immature dentritic cells in vitro , while non-manosylated liposomes have no effect as discussed in detail below. These data suggest that depleting atypical subsets of phagocytes with well-characterized mannosylated clodronate liposomes is an intriguing possibility worth investigating.
A Strategy for Targeting Phagocytes Resistant to Clodronate Liposome Depletion
If mannosylated clodronate liposomes are being considered for phagocyte depletion, then we assume that the target phagocyte population is not being sufficiently depleted by standard clodronate liposomes. While liposome mannosylation has been shown to target different phagocyte populations when compared to non-mannosylated liposome formulations outside the clodronate liposome literature, the applicability to macrophage depletion using clodronate liposomes is far from established as discussed in detail below. The general result of dosing mannosylated-liposomes intravenously is that liposomes are removed from the bloodstream faster and a larger proportion of the liposomes are taken up by the liver specifically by Kupffer cells. However, Kupffer cell depletion with standard clodronate liposomes is already quite effective and enhancing Kupffer cell uptake by liposome mannosylation will only further reduce the probability that the liposomes will not reach their target if it is outside the liver. Furthermore, when taking into account that mannosylated liposomes are more highly opsonized with presumably higher binding affinities due to the specific interactions of mannose with soluble mannose receptors, it seems even less likely that a mannosylated liposome will reach a cellular mannose receptor at all. Ghosh et. al. reported that mannosylated liposomes composed of egg PC/chol/mannosylated egg-PE (7:2:2) MLV encapsulating radioiodinated γ-globulin were preferentially (7X hepatocyte uptake but only 2x control liposome uptake) taken up by Kupffer cells post-i.v. injection and that the pre-injection of mannan reduced the uptake by about 60% (coincidentally, the pre-injection of mannan increased hepatocyte uptake 2x so that the uptake was similar between the two cell types in the presence of mannan). Although the authors take this as proof-positive that the mannan competitively inhibited the liposomes’ interaction with the Kupffer cell MR, we wonder if this could also be the result of the mannan binding/occupying soluble mannose-receptors so that the mannosylated liposomes were less opsonized. Note that these experiments were carried out before soluble MR or MBP were identified. Given more recently developed methods for detecting the degree of liposome opsonization, this hypothesis could be tested now. However, we are not aware of any data that demonstrates in vivo targeting of cellular mannose-binding receptors post-i.v. injection of mannosylated liposomes other than the increased uptake in the liver (and sometimes spleen and lung) which may be due to increased opsonization. There is also some data on enhanced uptake of mannosylated-liposomes by alveolar macrophages post-intrapulmonary administration discussed later on this page, however the lung and CSF represent different environments when compared to the bloodstream.
To reach bloodstream-accessible, mannose-receptor positive cells outside the liver, a significant number of liposomes will have to escape first-pass uptake by the liver and spleen, so that the target cells are exposed to a higher concentration of mannosylated liposomes from the blood. One strategy that has been used to ensure that liposomes escape the liver and spleen is known as RES blockade  in which animals are pre-dosed with a sufficient quantity of liposomes to temporarily saturate the phagocytic cells of the blood, liver and spleen, also known as the reticuloendothelial system (RES) or the mononuclear phagocyte system (MPS). This sufficient quantity is dependent upon the liposome type and composition as well as the species being dosed; the pre-dosed liposomes do not necessarily need to be the same type or composition as the therapeutic or diagnostic liposomes avoiding the RES. Soon after this pre-dose is cleared from the bloodstream (usually within a couple of hours), the liposomes of interest are dosed. Since the RES is involved in digesting the previous dose of liposomes, the subsequently dosed liposomes will remain in the circulation much longer thus be much more likely to bind to their target site outside the RES including those phagocytic cells which are accessible, but are not usually exposed to a higher concentration of liposomes.
While RES blockade is usually thought of as saturating phagocytic cells, it has been shown that opsonin-binding by liposomes is a saturable phenomenon . Therefore, part of RES blockade may involve serum depletion of complement and other opsonins known to coat liposomes. In the current application, removal or reduction in the concentration of soluble mannose-receptors may further increase the probability of a mannosylated liposome being able to interact with mannose receptors on the target cell. Therefore, if the goal is to deplete a target subset of mannose-receptor+ cells which may not normally be exposed to a substantial number of mannosylated liposomes, pre-dosing with mannosylated clodronate liposomes, in order to both saturate the blood, liver and spleen phagocytes and reduce the concentration of opsonins including soluble mannose receptors, should increase the number of subsequently dosed mannosylated clodronate liposomes available to this target subset hypothetically resulting in increased uptake and depletion by these targeted cells. We are not aware of any published applications proving this hypothesis and a series of experiments will be required to determine the appropriate RES blocking dose, timing of second dose, etc., however it is a reasonable approach to targeting a difficult-to-deplete subset of mannose-receptor+ cells. This rationale can also be applied to non-mannosylated liposomes or other routes of administration. The key is determining the optimal doses and timing between the two doses for the clodronate liposomes.
Liposomes Prepared in the Presence of p-Aminophenyl Mannose
While the primary message from the Buiting, et. al. paper is the the p-aminophenyl mannose formulation or other reformulations did not demonstrate a significantly different biodistribution or behavior compared to unmodified clodronate liposome formulations, we do not believe that the authors addressed the issue of clodronate liposome-induced macrophage depletion when analyzing their results. While this should not alter the earlier sampling data, it is not typical to observe a decrease in liposome content in the liver as early as 3 hours. The authors attribute this observation to the instability of clodronate liposomes in serum, which we agree is significant, however the release of clodronate after introduction of the liposomes into serum is very fast. The authors seemed to fail to appreciate the difference between liposome leakage in a test tube and the “infinite dilution” experienced by the liposomes in the bloodstream. The majority of the clodronate released from the liposomes in vivo must have occured well before 3 hours post-injection, therefore we believe that macrophage depletion also contributed to the atypical biodistribution results for unmodified clodronate liposomes. Since the authors did not directly assess macrophage depletion in this paper, we cannot draw any conclusions regarding the relative efficacy of the p-aminophenyl mannose clodronate liposomes with regard to macrophage depletion.
Van Rooijen and Sanders later re-evaluated the p-aminophenyl mannose clodronate liposome formulation compared to the unmodified clodronate liposome formulation by investigating the effects of the amount of clodronate incorporated into the liposomes on liver macrophages in vivo . They reported the number of ED2+ (resident) macrophages from histological samples of rat livers taken 2 days post-intravenous injection of clodronate liposomes.
Note on clodronate concentrations in van Rooijen and Sanders (1998)
Note that the clodronate concentrations reported are the initial concentrations of clodronate added to the lipid film which is not necessarily (not even likely) the final concentration of clodronate inside the lipsomes. The standard clodronate liposome preparation method used by van Rooijen utilizes 690 mM clodronate. In this paper, van Rooijen is also comparing the efficacy of another drug, propamidine, in macrophage depletion, and the maximum solubility of propamidine is 230 mM. Therefore, the maximum clodronate concentration used to prepare the liposomes is also 230 mM rather than the typical 690 mM. The unacknowledged variable introduced here is the effect of the osmotic strength of the initial clodronate concentration on the morphology and serum stability of the resulting liposomes. The possible ramifications are discussed elsewhere on this site.One of the biggest problems in analyzing the clodronate liposome literature is the lack of analytical data on the clodronate liposomes including no report of the clodronate and lipid concentrations in many papers.
As expected, the extent of macrophage depletion decreased with decreasing clodronate concentrations below 77 mM. The highest concentration used, 230 mM, as well as 77 mM, resulted in complete macrophage depletion, while lower concentrations provided what appears to be a linear dose response. Unfortunately, the actual clodronate concentration is not determined for any of the liposome preparations, so we can only assume that each different liposome preparation contained comparable amounts of clodronate (this is always a dangerous assumption, similar to assuming that control groups will not respond to a treatment and therefore unnecessary).The p-aminophenyl mannose formulation demonstrated a depletion profile indistinguishable from the standard clodronate liposome formulation.
…and the uptake of liposomes by macrophages can be enhanced by incorporation of, e.g., mannose in their bilayers (Gregoriadis, 1992). Such molecules have to be anchored in the bilayers by conjugation to strongly hydrophobic molecules. The best approach is through conjugation with the phospholipids themselves (e.g. phosphatidylethanol- amine). The resulting amphipathic conjugates are incorporated in the bilayers… from van Rooijen and Sanders (1994).In an earlier paper, van Rooijen and Sanders acknowledge the requirement for a stable lipid anchor for attachment of molecules to the liposome surface  leaving us to wonder why p-aminophenyl mannose was ever considered to be an appropriate molecule for liposome surface modification since it clearly does not meet the stated criteria (see left).
The bulk of the clodronate liposome work involving the p-aminophenyl mannose formulation does not include a control comparing the efficacy of the unmodified clodronate liposome formulation and we question the actual presence of bindable mannose on the surface of these liposomes as discussed in detail later. We have not yet completed a review of each publication, but so far, only one other paper has been identified which makes a rather unsatisfying comparison. Huitinga, et. al. did treat two groups of animals within the same experiment with the p-aminophenyl mannose clodronate liposome formulation and the standard formulation, but only clinical symptomatic assessment of the disease was reported. The p-aminophenyl mannose group did appear to experience fewer clinical symptoms and limited disease progression compared to the standard clodronate liposome group, but no data on macrophage depletion or other distinguishing histology between the two groups was specifically reported. Therefore, we consider this data anecdotal as it is a clinical observation without objective data for support.
p-Aminophenyl Mannose Liposomes Outside the Clodronate Liposome Field
Chono, et. al. published 3 papers containing data on enhanced alveolar macrophage uptake of p-aminophenyl mannose liposomes . The liposomes were dosed “just before the bronchus” into the lungs of rats using a Penn-Century microspray device inserted through the nasal cavity. In the latter of the the papers, the radiolabeled (HSPC/DOPC/chol/dicetylphosphate-3.5:3.5:2:1 with or without 9 mole% p-aminophenyl mannose) liposomes encapsulated ciprofloxacin (CPFX) , were extruded through 1 µm filters and were prepared/suspended in a pH 5.2 buffer; the resulting data demonstrated about 2X higher uptake in alveolar macrophages (AM) harvested 2 hours post-treatment. The absolute uptake is difficult to ascertain as the authors extrapolated the amount of drug or lipid assayed in the AM and lung epithelial lining fluid (ELF) to amount per ml of source tissue (to calculate PK parameters?). Since the total volume of AM is nowhere near 1 ml, the values are extremely high (1,000-3,000 % initial dose/ml tissue) and the parameters used for this calculation are not available to discern the original values. The data does suggest that the CPFX leaks from the p-aminophenyl mannose liposomes faster than the unmodified liposomes in vivo (perhaps the p-aminophenyl mannose is affecting the liposome stability?). We cite this paper first because the liposomes were prepared at pH 5.2. The pK of the amino group of the p-aminophenyl mannose is 4.74, therefore about half the p-aminophenyl mannose is positively charged. Since the liposomes contain dicetyl phosphate, a negatively charged lipid, it is reasonable to assume that the p-aminophenyl mannose is electrostatically associated with the liposome at the time of dosing. The route of administration also provides a very different environment for the liposomes as opposed to intravenous administration where the liposomes are effectively diluted in an “infinte volume” as well as being subjected to significant shear forces from the blood flow. The fluid content in the lung, while not stagnant due to breathing motion, is more similar to a test tube environment in which the diffusion rate of molecules leaving or attaching to the liposome is slower. Hypothetically, if the p-aminophenyl mannose is electrostatically associated upon dosing, it could remain near the liposome and initiate phagocytosis. Given that these are multilamellar liposomes prepared in the presence of p-aminophenyl mannose, the mannose derivative could be slowly released in the vicinity of the liposomes for some time. Prior studies indicate that free mannose added to the medium increases liposome phagocytosis in vitro presumably through the engagement of a mannose receptor by mannose in the vicinity of liposomes (aka “piggyback uptake”) . This mechanism obviously is more likely to contribute to enhanced uptake in vitro or in a contained compartment in vivo such as the lung or CSF and we believe that this may be the mechanism by which the p-aminophenyl mannose enhances uptake in these papers.
The liposomes described in the earlier Chono, et. al. papers are prepared at pH 7.4, rather than 5.2, and contain HSPC/chol/dicetylphosphate (7:2:1) with or without 9 mole% p-aminophenyl mannose but without CPFX. One paper  primarily evaluates the effect of liposome size on uptake and the other  looks at the effect of varying the amount of p-aminophenyl mannose added to the liposomes. The methods are identical to the previous paper except for the absence of CPFX. The data is presented at “% dose/mg protein” in these papers, so while are not seeing 3000% of the dose in the AM, we still cannot determine the %dose found in the AM per animal As with the previous paper, the p-aminophenyl mannose formulation increases uptake by about 2X over unmodified liposomes. We are not intentionally being contrary to this group’s data presentation, however we would like to know the magnitude of AM uptake. Are they seeing an increase from 1% to 2% of the dose going to the AM—not likely to change clodronate liposome results very much—or is the increase from 20% to 40% which could dramatically change the efficacy of depletion for clodronate liposomes?
Increasing the liposome size from 100 nm to 1 µm increases the uptake of unmodified liposomes in vitro, while uptake continues to increase up to 2 µm and perhaps larger in vivo ; liposomes larger than 2 µm were not assayed for uptake. The authors did not repeat the experiments for the p-aminophenyl mannose liposomes and chose 1 µm as the optimal size for this formulation. This observation regarding enhanced phagocytsis as the liposome size increases up to a few microns is in agreement with many such studies on liposomes performed in various systems and cell types.
There is clearly an increase in uptake as the level of p-aminophenyl mannose increases up to 9.1% in vitro. Furthermore, the enhanced uptake is inhibited by the addition of free mannose to the medium (not p-aminophenyl mannose). Any enhancement in uptake due to the presence of free mannose in the medium discussed earlier is detectable by varying the concentration of free mannose added, however only one concentration of free mannose, 20 mM, was assayed in this system. 20 mM mannose also inhibited uptake of mannosylated liposomes in the Shao and Ma paper cited earlier. Maximum inhibition in that study was observed at 10 mM free mannose; the mannosylated liposomes began to recover from the free mannose inhibition above 30 mM mannose and the unmodified liposomes also exhibited and increase in uptake in the 30-60 mM free mannose range. The effective inhibitory and excitory mannose concentrations will, no doubt, be different depending on specific experimental conditions including cell type, liposome type and lipid concentration, to name a few. Similar theories and demonstrations of “piggyback endocytosis” have been described for both liposome and non-liposome systems [55, 56, 57].
Although there are several controls as well as extensive liposome characterization to be done in this system, the data presented in these papers do indicate that these particular p-aminophenyl mannose formulations enhance liposome uptake by AM by about 2X when delivered directly to the lung. As stated above, knowing the magnitude of uptake would give us a better idea of how helpful this could be if we use them to deplete AM. Note that the formuations in the latter papers, in particular, are very different from the p-aminophenyl mannose formulaton used in the clodronate liposome studies. The Chono, et. al. formulation, HSPC/chol/dicetylphosphate/p-aminophenyl phosphate (7:2:1:1), is comprised of saturated phospholipids (more rigid bilayer) and contains ~9 mole% negatively charged lipid while the neutral van Rooijen formulation, EPC/chol/p-aminophenyl mannose (7:2:1), has an unsaturated, more fluid bilayer structure. We would expect to see a different binding profile of the p-aminophenyl mannose between theses two types of liposomes, so while the Chono p-aminophenyl mannose liposomes do demonstrate enhanced uptake by AM, we cannot conclude that the p-aminophenyl chlodronate liposome formulation used in many studies performed similarly.
Saturated and negatively charged liposomes, used in the Chono formulations, are known to be more readily phagocytosed, but it has been shown for clodronate liposomes that unless the negatively charged lipids interact with serum components (mainly Ca 2+ and Mg 2+ ) so that the liposome membrane is destabilized, the liposomal clodronate is ineffective in killing macrophages (see detailed discussion on the Intrapulmonary Administration page under “Clodronate liposomes are inherantly “leaky”…and should be”). We include this to emphasize that simply changing the standard clodronate liposome formulation from an unsaturated to a saturated lipid may not give the desired response and that the dicetylphosphate would have to perform similarly to DSPG in terms of divalent ion destabilization of the liposomes for this formulation to be effective as a clodronate delivery system.
Again, careful characterization of each formulation is necessary to determine the actual mechanism of enhanced uptake. To reiterate our hypothesis as to the mechansim of enhanced uptake observed with p-aminophenyl mannose liposomes as described in the Chono, et. al. papers, we believe that these results can be explained by the observation that free mannose appears to enhance the uptake of unmodified liposomes, as well as inhibit the uptake of mannosylated liposomes depending on the free mannose concentration under certain conditions such as the limited diffusion environments found in vitro and in more isolated, self-contained organ systems in vivo such as the lumen of the lungs or the CSF. The CNS specifically evolved the blood-brain barrier to very selectively control the environment of the brain preventing its exposure to systemic toxins or pathological changes in the homeostatic balance in other parts of the body, while the lungs have been designed to prevent airborne toxins or pathogens from entering the systemic circulation across the blood-gas barrier. These environments are different from the bloodstream or the peritoneum.
Who first claimed that this p-aminophenyl mannose liposome formulation targets mannose receptors? Hundreds of clodronate liposome studies, as well as some non-clodronate liposome studies, have been performed using liposomes prepared from EPC/cholesterol/p-aminophenyl mannose (7:2:1) with the assumption that the p-aminophenyl mannose is incorporated into the liposome bilayer at 100% and that the mannose residue presents on the liposome surface such that the liposome will preferentially target mannose receptors on cell surfaces. This assumption appears to be based upon a 1988 paper claiming that this formulation targets the mannose receptor in the brain . While we do not have conclusive data that these liposomes do not target the brain or the mannose receptor, the data contained in this publication cannot support any conclusions as discussed in this review of the paper. We encourage everyone to read the paper and draw conclusions as to the veracity of the claims.
Critical review of Umezawa and Eto (1988)
- The data contradicts itself within the paper. The conclusion of the paper is based upon Figure 1 in which a spike of probe (the authors assume the probe is liposome-associated, discussed later) appears at 48 hours. Yet, the tabular data looking at the cellular and subcellular distribution of the probe with time, does not demonstrate this spike at 48 hours. In both tables 1 and 2, which only assess the mannose-liposome treated animals, the maximal probe accumulation occurs at 24 hours and is back to baseline by 48 hours. The data in the graph and the tables were all collected from mice receiving i.p. injections of the same dose of liposomes. Why is the maximum brain accumulation at 48 hours in figure 1 but 24 hours in both tables as the authors acknowledge? Cellular or subcellular fractionation of the brain tissue versus lipid extraction of the brain tissue should not produce dramatically different results if the lipid extraction truly represents extraction of liposomal lipids and intact lipid probe, yet the authors don’t seem to notice that their “efficient” incorporation into the brain is not reproducible. Additionally, the majority of the probe after the subcellular fractionation of the brain appears in the supernatant suggesting that the radioactivity is no longer lipid associated. While we are not well-versed in brain fractionation, these results do not appear to represent an intact lipid probe. These observations alone renders the paper meaningless; the data is not reproducible within a single published experiment.
- Regarding table 2 showing the cellular location of the lipid probe taken up by the brain, “The cellular fraction in which liposomes were most incorporated was glial cells rather than neuronal cell. The greatest amount of radioactivity was found in the myelin fraction, however, this is probably due to co-migration of free liposomes with the myelin.” So, the majority of intact liposomes migrate across the BBB into the brain tissue remaining intact for up to 96 hours while escaping phagocytosis? And what mannose receptor interaction mediates the migration of intact liposomes directly into the extracellular space of the brain?
- The data presentation implies n=1 for each time point in the figure with additional animals being dosed and sacrificed at various times to produce the data in the tables. (i.e. Table 2 represents the cellular distribution of the probe in a single mouse brain from one animal sacrificed at each of the given time points.) Even in the absence of the other issues discussed here, n=1 does not support any conclusion especially in biodistribution studies.
- The methods state that after the organs were removed, “the lipids were extracted and the radioactivity measured.” Therefore, the data in the figure only represents the radioactivity that is lipid associated. Was the extraction method optimized for liposomal phospholipids and cholesterol or the galactocerebroside probe?
- In order for the spikes in liposome accumulation to occur as shown in figure 1, a sufficient fraction of liposomes would have had to remain in circulation for 24 hours, then suddenly accumulate in brain tissue between 24 and 48 hours followed by complete metabolism and washout at 72 hours. It is difficult, if not impossible, to postulate a mechanism for this liposome behavior. This is why its critical to collect and assay blood as well as organs in a biodistribution study . A gradual brain accumulation might have been more convincing, but the salvage pathway (see below) of sphingolipid catabolism makes the 24 hour washout of the high level of probe an equally inexplicable phenomenon.
- The methods state that 200,000 cpm in 0.5 ml was intraperitoneally injected into each mouse. Figure 1 indicates that the data is presented as cpm/mg wet tissue. This means that the spike in probe reported at 48 hours (10,000 cpm/mg wet tissue) amounts to 4 million cpm in the whole mouse brain (~400 mg). Even if we assume a printing mistake and the y-axis represents cpm/g wet tissue, the reported data cannot be convincingly reconciled.
- Once a liposome leaves the bloodstream or the peritoneal cavity and becomes tissue-associated, presumably by phagocytosis, the intact liposome does not reenter the circulation, rather it is metabolized within the phagocyte. Therefore, when probe activity peaks in an organ and begins to decrease at subsequent time points, the liposome and the probe are no longer associated; the liposome has been degraded and there is no longer a liposome to follow. The probe is only liposome-associated as long as the liposome remains in the bloodstream and this assumption is only valid if the probe has been verified not to exchange with serum components in the bloodstream. Clearly this data, if it could interpreted, represents the redistribution of the 3 H-galactocyl-ceramide or its metabolites after the liposomes were digested intracellularly in the liver and spleen. This is further supported by the coincidental spike in kidney accumulation; probe accumulation in the kidney is a usually result of metabolite excretion although galactocerebroside itself has been shown to accumulate in the kidney in certain pathological conditions. Liposomes are not typically retained in the kidney; probe found in the kidney or urine is considered no longer liposome associated.
- Galactocerbroside (galactosyl-ceramide) is found in large quantities on the outer leaflets of neuronal cell membranes. While the contribution of the salvage pathway (generates sphingolipids from components of partially degraded molecules rather than de novo synthesis) of sphingolipid catabolism varies among cell types, it accounts for at least half of the overall production of galactocerebrosides [25, 26, 27, 28]. Therefore it is reasonable to expect that the partially degraded 3 H-galactocerebroside released from the spleen and liver could accumulate in the brain. To reiterate, this probe is no longer liposome-associated; the authors are tracking 3 H-galactocerebroside metabolites.
Biodistribution studies must contain controls like any other experiment. The behavior and metabolism of the probe must be known. Any biodistribution data must be accompanied by blood levels…we could go on regarding the experimental design, data presentation and conclusions from this paper. But, it is clear that this publication cannot be used as a justification for this liposome formulation being utilized to target mannose receptors or deliver compounds across the BBB. This paper has been validated by hundreds of references from the clodronate liposome community, among others, and we hope that this critique will convince those considering using this formulation to carefully reevaluate the data in the Umezawa and Eto paper and strongly consider choosing a different clodronate liposome formulation until the targeting ability and characterization of these liposomes is verified.
While the contents of the paper is reason enough to question the use of this formulation, there are additional issues with the formulation itself which require, at a minimum, that this formulation be carefully evaluated for its actual targeting ability. It has been the assumption for all of whom have employed this formulation that the p-aminophenyl mannose fully incorporates into the bilayer and displays mannose on the surface of the liposome such that the mannose is accessible for mannose receptor binding. As liposome scientists, who relatively recently began to investigate the liposome component of the hundreds of clodronate liposome publications, this premise is intuitively contrary to at least 2 basic tenets of liposomology as discussed below. Furthermore, we do not find any data verifying neither the stable incorporation of p-aminophenyl mannose nor the resulting liposome’s ability to bind mannose receptors. All the publications discussed below which report the synthesis and incorporation of lipid-anchored mannose into liposomes minimally demonstrate the ability of these liposomes to bind mannose-specific lectins (concanavalin A). Many also demonstrate specific binding to cellular mannose-receptors through competitive inhibition with mannan or other mannose-containing molecule which was briefly addressed in one p-aminophenyl mannose liposome paper described above. The effort and expense involved in attempts to target mannose-receptors in vivo certainly warrant more rigorous performance criteria for the mannosylated liposomal clodronate formulations. This data is also necessary to support any hypothesis specifically invoking mannose-receptor binding. Without characterization of each liposome formulation to ensure that mannose-receptor binding is at least possible for that formulation, conclusions on mannose-receptor mediated uptake is one of many explanations for any enhanced uptake observed. More commonly in the clodronate liposome field, negative results lead the investigators to believe that mannose-receptor uptake is not a factor in their model, when it is quite possible that the liposomes do not have bindable mannose on their surface, thus mannose-receptor contributions have not actually been evaluated.
With respect to the “liposomological” assessment of the p-aminophenyl mannose liposome formulation, it is well known that small lipophilic compounds which readily incorporate into the bilayer quckly exchange out of the bilayer upon dilution or the addition of serum or blood (or many individual components of these) providing competitive binding options for these compounds . Even cholesterol will redistribute itself when mixed with cholesterol-deficient liposomes or cell membranes [30, 31, 32]. Longer single-chain fatty acids also quickly exchange out of liposomes in vitro and in vivo . Cholesterol and fatty acids are obviously highly soluble in the lipid membrane while p-aminophenyl mannose has a negative logD (HLB or octanol/water partition coefficient) at all pH values indicating that it is prefers an aqueous environment rather than a hydrophobic environment such as the acyl-chain region of the bilayer. While it may not be unbelievable that the phenyl group could insert into the hydrophobic region of the bilayer allowing p-aminophenyl mannose to show some binding to liposomes, it is much more difficult to envision this interaction to be maintained once the liposomes are exposed to biological media. Many groups have attempted to use various non-convalently attached molecules (often oligosaccharides and other polymers), which are known to interact with the liposome surface, as targeting mechanisms. Some level of targeting can be achieved, but including an anchoring mechanism through covalent modification with a lipid or strong electrostatic interaction is usually required to obtain substantial and reproducible targeting . We believe this will be the case when liposomes containing p-aminophenyl mannose are directly compared to liposomes with covalently bound mannose on their surface. The only data of which we are aware that verifies any p-aminophenyl mannose association with liposomes was published by Weissig, et. al. . Click on the tab below for a detailed discussion of the data in this paper, but their data suggests that p-aminophenyl mannose may not associate with liposomes at the concentration used in the preparation of the clodronate liposomes. However, there are differences in liposome composition and preparation methods which may contribute to p-aminophenyl mannose association with liposomes. This further emphasizes the need for detailed characterization of the clodronate liposomes prepared with p-aminophenyl mannose.
Weissig, et. al. Covalent coupling of sugars to liposomes
The following table contains the data from this paper on the association of p-aminophenyl mannose to egg PC/cholesterol (1:1) liposomes containing various amounts of n-glutaryl-DOPE. After preparation of the liposomes increasing amounts of p-aminophenyl mannose is added either in the presence or the absence of the coupling reagent, carbodiimide. The amount of mannose covalently coupled to the liposomes is determined by subtracting the amount of p-aminophenyl mannose associated with the liposomes in the absence of coupling reagent from the amount associated in the presence of the carbodiimide (“net” column). Coupling is confirmed by a new spot appearing on TLC of intermediate migration between the two reactants. The amount of n-glutaryl-DOPE incorporated into the liposomes was limited to 4.5 µmoles due to the fact that increasing amounts of this lipid appeared in the supernatant (outside the liposomes) when larger amounts were added. We believe that this is due to the fact that DOPE is a non-bilayer forming lipid which often exhibits anolomus behavior in liposomal formulations. The authors do not speculate on the results other than observing that the binding of p-aminophenyl mannose appears to be cooperative due to the dramatic increase in binding when the amount of p-aminophenyl mannose in the suspension is increased 4X corresponds to a 6-8X increase in bound sugar. They also reiterate that the DRV method of liposome preparation can produce non-liposomal structures and bilayer anomalies [39, 40], although this preparation seems to retain most of the encapsulated carboxyfluorescein throughout the coupling process indicating stable liposome structure.
Perhaps the most relevant observation for application to clodronate liposomes containing p-aminophenyl mannose is that binding in the absence of coupling is not detectable until the p-aminophenyl mannose concentration reaches 3.6 mM or 0.98 mg/ml. The concentration of p-aminophenyl mannose in the clodronate liposome formulations is 3.3 mM or 0.9 mg/ml. While the clodronate liposomes differ from those used in this paper in lipid composition and preparation method, this data suggests that little or no p-aminophenyl mannose would remain on the surface of the clodronate liposomes after removal of the unencapsulated clodronate. Experiments quantitating the amount of p-aminophenyl mannose on the surface of the clodronate liposomes must be performed with the clodronate liposome formulation to confirm or disprove this hypothesis.
One other possibility which was not investigated is that some p-aminophenyl mannose may cross the membrane and remain inside the liposome during the course of the experiment above rather than being surface associated. Thus, the p-aminophenyl mannose is certainly liposome associated, but not on the surface, thus rendering any binding most unlikely. We are not aware of any data on the encapsulation properties of p-aminophenyl mannose and consider this question to be one of the most important unknowns to be elucidated in the p-aminophenyl mannose formulation arena. However, Ghosh, et. al. demonstrated that the p-aminophenyl mannose migrates into egg PC/chol/egg PE MLVs  under the conditions they used for coupling the sugar to PE. The sugar was added to the preformed liposomes followed by glutaraldehyde, the coupling reagent. After dialysing the uncoupled p-aminophenyl mannose and gluataraldehyde from the mannosylated liposomes and dissolving the mannosylated liposomes in Triton-X 100, they determined that 80-85% of the total egg PE was conjugated to the p-aminophenyl mannose while only 27% of the total sugar was on the outside of the liposome using Con A agglutination. Therefore the gluaraldehyde and p-aminophenyl mannose must have accessed the inner lamellae of the liposomes in order to access the PE on the inside of the liposomes. The authors cite Torchillin, et. al. who developed this method for coupling proteins to liposomes and reported that encapsulated 3 H-deoxyglucose did not leak from the liposomes during the coupling process  suggesting that the glutaraldehyde process did not cause the liposomes to become leaky. If p-aminophenyl mannose does cross liposomal membranes, a slow release of encapsulated p-aminophenyl mannose from multilamellar liposomes could explain some of the properties of this formulation through “piggyback endocytosis” discussed earlier.
Another rule of liposome-cell binding addresses the ability of liposomes to bind cell-surface receptors. The liposome-bound ligand must be able to come into close proximity of the cell surface for binding to occur. Many studies involving antibodies and various other ligands, including mannose, have shown that a spacer which extends the ligand at least 3-4 carbons away from the surface of the liposome allows or enhances liposome-ligand to cell-surface-receptor binding [36, 37]. The spacer alleviates steric hindrance which prevents the two surfaces (liposome and cell) from coming into close contact with each other. While a 3-4 carbon spacer is often adequate to prevent steric hindrance, it is advisable to evaluate a series of spacers in order to determine the spacer length which demonstrates optimal liposome-cell binding preferably in vivo. Antibody targeting often performs best when PEG spacers of tens, hundreds or more monomers (several nm) are used. The n-glutaryl-PE lipid anchor synthesized in the Weissig, et. al. paper with its 4-carbon equivalent spacer arm is frequently of sufficient extension and is now commonly used in its unsaturated and monounsaturated forms (to avoid the problems attributable to the diunsaturated species as discussed above) as a general lipid anchor for many types of surface modifications, including mannosylation; it is commercially available ( Avanti Polar Lipids , Alabaster, AL, USA) simplifying the mannose-coupling chemistry.
This is the structure of the Man-Glu-DOPE conjugate next to a POPC molecule showing the relative extension of the mannose beyond the surface of the bilayer. But, if we try envision the aminophenyl group of the uncoupled sugar extending down into the hydrophobic region of the bilayer (the probable binding moiety) as pictured here, it appears that the mannose would be buried amongst the headgroups. As was experimentally demonstrated for the palmitoylated mannose compounds by Engel, et. al.  (discussed below), when the mannose is buried in the headgroups, the mannose is not available for binding at reasonable concentrations. Given this model in combination with the hypothesis that little, if any, p-aminophenyl mannose is associated with the liposome at the concentration used to prepare clodronate liposomes, it seems unlikely that this formulation will effectively target cells displaying the mannose-receptor. Nonetheless, we have made several assumptions which may not prove true in experiments designed to assess the association of p-aminophenyl mannose with clodronate liposomes.
Clodronate Encapsulation into Covalently Linked Mannosylated Liposomes. One group has published data indicating that immature dendritic cells (iDC) only respond to clodronate in vitro when it is encapsulated in liposomes including novel synthetic lipid-anchored mannose derivatives designated as ManXDOG where X=1, 2 or 4 mannose molecules [45, 76]. The structure of Man2DOG and Man4DOG is shown here.
The authors intended to use the liposomes for antigen delivery to iDC and were very thorough in their evaluation and optimization of these liposomes, however their selection of lipid composition and liposome size probably tempered their binding results as the authors acknowledged. Despite this handicap, they demonstrate that even with small unilamellar liposomes (70-100 nm) composed of unsaturated lipids (egg PC and egg PG) without cholesterol, the liposomes were taken up by 80-90% of the iDC while little (<10%) or no uptake was observed with unmodified liposomes. They further demonstrated a stunning dose response to clodronate when it was encapsulated into these liposomes; again, no response to free clodronate, unmodified control clodronate liposomes, or control ManXDOG liposomes without clodronate. While the Man2DOG liposomes performed better than the Man1DOG liposomes at the same total mannose concentrations, there was little difference between the Man2DOG and Man4DOG liposomes. The authors attribute this to the ability of the Man2DOG to phase separate or “cluster” so that the surface presentation of Man2DOG is not different from the Man4DOG demonstrating their intimate understanding of lipids, bilayers and liposomes. While the Man2DOG has not been compared to other mannosylated lipids for its relative targeting ability, we highly recommend this paper as a model for those unfamiliar with developing effective liposomal formulations due to their comprehensive evaluation, liposome characterization/optimization and explanation of the mechanism of mannose-targeting by these derivatives. Of course, further work would be required (other lipid compositions, liposome sizes, etc.) to complete the formulation development, but with respect to evaluating a new lipid derivative for its targeting ability, Espuelas, et. al. discuss and design experiments to address many, if not all, of the issues that we have raised with the p-aminophenyl mannose liposome formulations. Several papers have been published on the use of this mannosylated lipid in enhancing adjuvant activity in vivo , but we are not aware of any which specifically address the cell population(s) targeted by these derivatives [58, 59].
Other Publications on Mannosylated Liposomes. Several mannosylated lipids have been synthesized and incorporated into liposomes in addition to those discussed elsewhere on this page as well as in the recent Kelly, et. al. review . Representative papers for some of these mannosylated lipids are discussed later on this page.
Some groups have historically argued that cholesterol is a better anchor for both carrying a lipid soluble label (fluorescent, radioactive, etc.) and surface modifications. Although phospholipids are now more commonly used to anchor surface modifications, the cholesterol anchor is still a popular choice for radiolabeling liposomes. The development of simple phosphatidylethanolamine coupling methods which can be performed either before or after the liposomes are prepared probably trumped any advantage of cholesterol anchors as we are not aware of any proven methods of liposome surface modification post-preparation of liposomes for cholesterol-anchored derivatives. Other lipid anchors have also been investigated as discussed elsewhere on this page. While the distribution of modified lipids on both sides of the lamellae is often not an issue, there are circumstances in which it is either desirable or necessary to only modify the surface bilayer leaflet of the liposome, rather than the inner leaflet or internal lamellae. Therefore, most of the surface-modification methods currently used may be carried out on intact liposomes containing the reactive lipid anchor. Lipid anchors other than cholesterol and PE have also been investigated; the most important factors in selecting an anchor is ensuring that the anchor does not readily exchange out of the lipid bilayer and that the anchor allows enough conformational freedom for the ligand to adopt the appropriate orientation for binding its receptor.
The fact that liposomes and cells have a minimum distance from which other particles and macromolecules can approach their surfaces is addressed in many of these papers. This is of particular concern when designing a liposome intended to interact with a cell through a specific binding mechanism since cells also demonstrate a minimum distance from their surface inside of which steric hindrance inhibits binding. The structures of the derivatives are shown aligned with phosphatidylcholine molecules to give an idea of the distance that the mannose residue extends from the surface of the liposome. Orr, et. al. did a minimal proof-of-concept study looking at two spacers of different lengths, but a more thorough evaluation was performed by Engel, et. al.  who systematically evaluated a series of 9 derivatives of mannose palmitate each containing 0 up to 8 ethylene glycol (EG) spacers. We provide a diagram of the approximate locations of the mannose moiety of each of those conjugates within the bilayer, so it may be useful to first look at the Engel tab below to get an idea of how the positioning of the mannose group dictates receptor binding as we touched upon briefly in the p-aminophenyl mannose discussion. This group concluded that spacers longer than 8 EG’s may further enhance cell binding. Frisch, et. al.  later published a method (“click chemistry”) for attaching a variety of molecules to the liposome surface; their example demonstrating “click chemistry” was a lipid-anchored mannose containing 12+ ethylene glycol spacers, but no binding data was included. However, as Engel, et. al. discussed, it has been demonstrated that spacers can be too long for effective targeting, further emphasizing the need to evaluate a series of spacer arm lengths for each particular application. We would also emphasize that spacer arm lengths must be evaluated within the particular environment of their intended use, since this paper also shows that different spacer lengths perform optimally depending on the species of mannose receptor being targeted.
The lipid anchor for the Frisch, et. al. conjugate was a disaturated C-16 analogue of DOG lipid-anchor previously employed by Espuelas, et. al. The diacyl lipid anchor greatly reduces the potential for exchange of the mannosylated lipid out of the liposome as opposed to the hexadecanol (monoacyl) anchor used by Engel, et. al. Lipids composed of a single acyl chain (i.e. fatty acids) are well known for their propensity to exchange into or out of bilayers when compared to diacyl lipids of similar acyl chain lengths. This tendency is exacerbated when the headgroup is modified with a large, highly water-soluble compound such as the EG(x)-mannose. Thus, the PEG in PEGylated (stealth) liposomes is anchored to DSPE (diacyl anchor) rather than stearic acid (monoacyl anchor) ; PEG-stearate was shown to be completely ineffective in increasing the blood half-life of liposomes due to the fact that it quickly exchanges out of the liposome post-injection.
When visualizing the presentation of the mannose on the surface of the liposome, it is also important to remember that the membrane is a not a stationary, fixed surface like a polystyrene sphere. The membrane undulates along its surface (envision ocean water rippling along its surface) and the individual lipid molecules rotate and move laterally within the membrane. Additionally, the motion within the individual phospholipid molecules is significant primarily due to the C—C rotations within the acyl chains. Lipids are leaving and entering the membrane at various rates depending on the liposome composition. While not energetically favorable in most systems for most lipids, some lipids can even move from the outer leaflet to the inner leaflet (flip) or vice versa (flop) although lipids with larger modified headgroups, such as these mannosylated lipids do not spontaneously flip-flop. Water and other small anions (i.e. chloride) are migrating through the bilayers while cations are attracted (but not bound) to the surface due to the electronegative phosphate group. Hydrogen-bonds are forming, breaking and reforming everywhere regulating much of this activity. The magnitude and speed of these collective motions are, of course, affected by temperature and other parameters, but the most important factor is the lipid composition of the membrane including surface modifications. This means that the lipid composition of the liposomes in which the mannosylated lipids are inserted plays a role in the binding behavior of the liposomes. Therefore any indirect comparisons of binding behaviors from paper to paper must be viewed as a function of liposome composition and type as well as the structure of the mannosylated lipid. If mannose receptor binding is demonstrated in mannosylated, saturated lipid liposomes without cholesterol, it is very unlikely that the same binding profile will be obtained with the same mannosylated lipid in egg PC/cholesterol liposomes. Both liposome types may provide sufficient binding for the particular application, or neither may work well enough, but it is almost certain that the relative binding properties of the two preparations will be different.
While reading and contemplating the results in each of these papers may seem irrelevant to the world of clodronate-liposome-mediated macrophage depletion, the strategies and hypotheses presented are exactly those which must be considered when designing targeted liposomes. And, if the clodronate liposome application calls for depletion of a specific clodronate-sensitive phagocyte population not normally substantially depleted with the standard EPC/cholesterol clodronate formulation, mannose-receptor targeting certainly provides a valid rationale for reaching such a population as demonstrated by Espuelas, et. al. for dendritic cells. Furthermore, Encapsula has formulated and provided covalently mannosylated clodronate liposomes to a client who reports effective targeting of a mannose-receptor positive alveolar macrophage subset and we eagerly await the publication of this data.
The selected publications below describing various mannosylation strategies for liposome targeting are presented in chronological order, however, as suggested earlier, the Engel, et. al. paper may provide the best basic explanation of the importance of the relative distance of mannose from the surface of the bilayer. However, while this may be a good starting point, many other important factors are discussed in each of the papers.
Publications Describing Some Mannosylated Lipids and Their Applications
- Ref 37. Orr, et. al.
- Ref 41. Ghosh, et. al.
- Ref 63. Baratt, et. al.
- Ref 62. Muller & Schuber
- Ref 68. Sugimoto, et. al.
The synthesis of one of the older mannosylated lipid derivatives using cholesterol as the lipid anchor was presented in this paper. In fact, this seems to be the oldest reference to a completely synthetic mannosylated lipid derivative studied in liposomes albeit by only a few months. The authors provide an excellent justification for the use of cholesterol anchors in relation to the surface presentation of the ligand. Both the shorter and longer spacers allow lectin binding to the mannose presented on egg PC SUV, but the longer spacer demonstrates a higher rate of binding at a lower % incorporation of the derivative. These conjugates were generated for the purpose of incorporation into cell membranes, thus the subsequent use of this particular derivative seems to have been limited to the study of enzymes involved in the removal of sugar residues from glycosylated membrane lipids rather than in vivo targeting.
A few months later, Ghosh, et. al. published their work with a mannosylated-phosphatidylethanolamine along with other glycolipids and their ability to enhance liver uptake of egg PC/chol/glycolipid (7:2:2) multilamellar liposomes containing radioiodinated-γ-globulin in mice post i.v.-injection of about 40 mg/Kg lipid. Liver uptake at 15 minutes post-injection was enhanced about 1.6X, lung uptake was enhanced by almost 3X, while splenic uptake was reduced by 2.3X. Furthermore, the enhanced uptake was obliterated when mannan, but not other glycosides, were co-injected with the liposomes. This group later demonstrated that mannosylated lipids preferentially targeted non-parenchymal liver cells, including Kupffer cells, while galactosylated liposomes targeted parenchymal (hepatocytes) cells .
Banerjee, et. al. used this conjugate to increase the efficacy of liposomes carrying antibiotics for the treatment of leishmaniasis .
This group isolated mycobacterial mannosylated-phosphatidyl- myo inositol (PIM) and prepared liposomes from this lipid mixed with 33 mole% cholesterol. The authors report an important result from passing these liposomes through 3 µm syringe filters (Swinnex filter holder). While the size of the liposomes was smaller (700-2300 µm depending on filter type), almost 70% of the lipid was retained by the filters*. This was not exclusive to the mycobacterial lipid-containing liposomes; PC/PS liposomes behaved similarly.
One goal of this group was to prepare stable liposomes for encapsulating immunomodulators and selective targeting to alveolar macrophages. They demonstrated about a 4X increase in uptake in vitro in isolated rat alveolar macrophages when compared to PC/PS (7:3) liposomes.
The specific structure(s) of the purified PIMs from Bacillus were not shown, but the authors state that the mixture consisted of PIMs with varying numbers of mannose residues. Drieson, et. al. more recently demonstrated the selective binding of PIMs from M. tuberculosis and mutants to DC-SIGN specifically looking at the relative affinities of PIM 2 , PIM 4 and PIM 6 . PIM 6 demonstrated a high binding affinity, but the lesser mannosylated PIMs showed little binding to DC-SIGN . This result would argue that conjugates containing either many branched mannose residues or conjugates that can cluster in the liposomal membrane will increase specificity for the SIGN receptors.
*We have often emphasized he difference between high-pressure extrusion, which reduces the size of liposomes by larger liposomes disrupting and reforming into smaller liposomes, and low-pressure filtration through a filter support attached to a standard syringe (not gas-tight syringes) which simply retains larger liposomes or stable aggregates. This example clearly demonstrated the difference by the bulk of the lipid being retained on the filter during low-pressure syringe filtration. Little or no retention of lipid occurs during successful high-pressure extrusion.
Muller and Schuber utilized a coupling method introduced by Martin and Papahadjoupolos  for coupling IgG fragments to preformed liposomes. As shown in the figure, the mannose is extended from the surface of the liposome by a 15 or so atom spacer arm, the longest used up until this publication. The DPPE-MPB lipid anchor synthesized by this and previous groups has since become commercially available ( Avanti Polar Lipids , Alabaster, AL, USA) with both saturated and unsaturated acyl chains for incorporation into all types of liposomes. In addition to the mannose conjugate shown here, the authors synthesized an analogous conjugate of similar spacer arm length using phenyl derivatives of mannose such that the thioether sugar linkage shown here was substituted with the phenyl sugar group. REV (100-500 nm, oligolamellar) liposomes were prepared from EPC/cholesterol/MBP-egg PE (10:7:1 or 10:7:1.5) followed by coupling of the sugar residues to the resulting liposomes. Note that the disaturated-DPPE derivative is shown in the figure rather than the predominantly monounsaturated-egg PE derivative used in this publication. Both sugar conformations demonstrated enhanced binding to primary mouse peritoneal macrophages and Kuppfer cells, but the thioether sugar conjugate uptake was about 2X higher than the phenyl sugar conjugate uptake which was about 2X higher than control liposome uptake. Therefore most of the studies were performed with the thioether conjugate when the added advantages of previously reported enhanced uptake of non-liposomal thioether mannose derivatives, as well as higher hydrolyisis resistance for the thioether sugar, were taken into account. These mannosylated liposomes showed a concentration dependent binding to Con A competitively inhibited by mannan. Increasing concentrations of the liposome added to primary mouse peritoneal macrophages did not reach saturation at the highest lipid concentrations. Kupffer cell binding (at 4°C to prevent phagocytosis) of fluorescent liposomes was not inhibited by mannan although it was inhibited by non-fluorescent mannosylated liposomes but not non-mannosylated liposomes.
These liposomes were further reported to target tumors  and inhibit their growth when LPS was incorporated into them.
This series of mannosylated lipids were synthesized based on their widely observed occurrence at the termini of glycolipids found on various primarily human glycoproteins (IgG, transferrin, RNAse B, etc.) and their selective ability to bind MBP [81, 82]. Sugimoto, et. al. first tested these lipids as adjuvants when included into DPPC/chol (2:1) at about 6% by weight (2.3 mole %) for the Man5-DPPE conjugate . The authors do not specify the amount of the Man(x)-DPPE incorporated into the MLV, therefore we back-calculated based on the stated doses of lipid and Man(x)-DPPE per mouse. The encapsulated antigen was ovalbumin (OVA) and the mouse footpad swelling adjuvant assessment model was used. In this model the liposomes were injected into the mouse footpad. Only mannan-coated liposomal OVA solilicted a larger response than Man5-DPPE by about 40%. In subsequent experiments, Man2- and Man3-DPPE at 10 mole % also performed well as adjuvants. The significance of this finding was that mannan, although an excellent immunostimulator, was highly immungenic itself and was reported to be toxic when injected i.v. into mice. The authors assume that the Man(x)-DPPE liposomes will be neither toxic nor immunogenic since the Man(x) was derived from human glycoproteins.
Several papers have since been published [16, 67, 68, 83, 84] on the use of these mannosylated liposomes as safe adjuvants for both immunization against infectious organisms as well as tumor cells. However, we did not find information on the biodistribution of these liposomes although one paper did demonstrate Con A agglutination, preferential uptake into CD11b+ (Mac-1) cells post-i.p. injection and inhibition of uptake by α-methylmannose (but not galactose) by Man3- (0.07 moles) and Man5-DPPE (0.09 moles) in DPPC/chol (1:1) MLV.
An Encapsula client selected Man3-DPPE for incorporation into clodronate liposomes targeting a subpopulation of MR+ cells. The client reported successful targeting and will be publishing the data which we will review ASAP. We can provide many of these different mannosylated lipid species in clodronate liposomes should any of our clients prefer a different mannosylated lipid species for targeting mannose receptors on cells. However, we cannot guarantee the performance of any of these mannosylated clodronate liposomes for macrophage depletion until more data on their behavior becomes available.
This mannose cholesterol conjugate was incorporated into DSPC/chol/man-C4-chol (60:35:5) 100 nm liposomes and pharmacokinetic parameters calculated post-i.v. injection into mice. Liver uptake was about 4X higher (75% initial dose) for the mannosylated liposomes compared to non-mannosylated liposomes. When added to DOPE or DOPE/chol liposomes followed by complexation to nucleic acids at 6 hours post-i.v. injection into mice, the levels of luciferase were significantly higher in the lung, liver and spleen (but not kidney and heart) in animals dosed with the mannosylated lipoplex than with the non-mannosylated lipoplex .
Note that partial inhibition of the DSPC/chol/man-C4-chol liposome response by mannan was demonstrated in all of these studies evaluating man-C4-chol targeting.
When DSPC/chol/man-C4-chol (60:35:5) 100 nm liposomes encapsulating muramyl dipeptide (MDP, an immunomodulator) were dosed i.v. to mice to which CD-26 (metastatic tumor) cells had been administered, the man-C4-chol liposomes reduced the number of metastatic lesions in the liver by 8X compared to non-mannosylated liposomes and >10X compared to the free MDP control . These results were indeed striking.
Dexamethasone palmitate (DP) was incorporated into the same liposome formulation at 9-10 mole % and dosed intratracheally ( Penn-Century microspray device ) to rats pretreated with LPS. The difference in cytokine suppression and neutrophil infiltration was only significant at 3 hrs post administration for the man-C4-chol DP liposomes, otherwise both liposome treatments were equally effective at -3 hr and 1 hr post-LPS administration . The man-C4-chol DP liposome inhibition of NFκB activation of p38MAPK phosphorylation in both alveolar macrophages and lung tissue was dramatic when compared to non-mannosylated DP liposomes and free DP.
The effect of man-C4-chol concentration in this liposome formulation was evaluated for selective uptake of the liposomes by alveolar macrophages in vitro and in vivo by intratracheal administration in mice . DSPC/chol liposomes containing 2.5, 5, and 7.5 mole % man-C4-chol showed a concentration dependent increase in AM uptake which appeared to be nearing saturation at the highest man-C4-chol concentration. The authors further demonstrated that the uptake is specific to AM and not lung epithelial cells (type II) which had been shown to be the predominant cellular target of “surfactant-like liposomes”.
This branched mannose was selected for conjugation because it is the terminal region of the major surface glycoprotein of Leishmania mexicana amazonensis. Its use has been limited to incorporation into DOPE/DC-chol liposomes for targeting the human IL-10 gene to mouse peritoneal macrophages [88, 89].
This paper provided an excellent demonstration of the impact of the position of the mannose relative to the bilayer. Moreover, they further demonstrated the importance of evaluating mannosylated liposomes in various systems rather than just defaulting to, for example, the conjugate which performed best in the Con A agglutination assay. The authors nomenclature for these derivatives was based upon the number of ethylene glycol (EG) spacers incorporated between the hexadecanol (C16) lipid anchor and the mannose . Therefore Man0 represents mannose directly conjugated to hexadecanol and Man1 incorporates 1 EG between the hexadecanol and mannose. Liposomes containing HSPC/DPPG/Man(x) (63:7:30) extruded through 200 nm filters showed no Con A agglutination when the Man0 and Man1 conjugates were incorporated into the liposomes.
Looking at this figure, it is clear that the mannose residue is buried within the lipid headgroups for these 2 molecules. As expected Man2, Man3 and Man4 liposomes were agglutinated by ConA, but Man2 demonstrated the highest agglutination. When incubated with HL-60 cells (no MR) the percent of these liposomes taken up decreased dramatically with increasing spacer length from almost 120% for Man0 to less than 10% for Man4. The authors explained this by showing that propensity of the liposomes to bind to the cells showed an identical response; the liposomes with the longer spacers demonstrated an inhibition of binding compared to Man0 or control liposomes. Many small molecules conjugated to the surface of liposomes have been shown to decrease phagocyte uptake as well as large polymers such as the most commonly used PEG 2000 . Since these experiments were performed under serum-free conditions, opsonization did not play a role in phagocytosis or binding. <p>However, when the liposomes (including those containing the Man6 and Man8 lipids) were incubated with primary human pericardial monocytes/macrophages isolated from patients undergoing heart surgery, also under serum-free conditions and in suspension, the response was the opposite for MR+ cells. The Man6 and Man8 liposomes showed a ~100% and ~250% increase, respectively, in uptake for MR+ cells (macrophages), but not MR- cells (monocytes), when compared to control liposomes. The percentage of macrophages (MR+) compared to monocytes (MR-) varied from 15% to 57% depending on the patient. After incubation with human primary peritoneal macrophages, the uptake by Man6 and Man8 liposomes were ~4X and 5X control while the shorter spacer-containing liposomes’ uptake was about 2X control. Thus, the response of cells in vitro is dependent upon the presence of the macrophage MR, furthermore the response by MR- cells is the inverse of the response by MR+ cells.
Also note that while the Con A agglutination confirms the presence of accessible mannose molecules on the surface of the liposomes, therefore important for confirming the success of producing a liposome with mannose bound to its surface, it is not predictive of mannosylated liposome performance with regard to cellular uptake of the liposomes.
The authors’ second publication on the liposome incorporation of these mannosylated lipid evaluated the effect of the liposome composition on ConA agglutination . When incorporated into soy PC/chol (7:3) liposomes, the agglutination response required significantly more of the mannosylated lipid incorporated into these unsaturated liposomes compared to the saturated PC/PG liposomes as used in the first study. The authors conclude that phase separation (clustering) of the mannosylated lipids found in the saturated liposomes provided a better ligand presentation than when the mannosylated lipid was evenly distributed around the surface of the liposomes. This interpretation is in agreement with other studies describes on this page showing that liposome presentation of multiple mannose binding residues on a single lipid anchor is preferable to a single mannose residue per lipid anchor.
This group used these manosylated lipids for studying lyophilized liposomes and we did not find any information on the in vivo performance of these lipids from this group, although we suspect that these conjugates, especially those with longer, water-soluble EG spacers would exchange out of the liposomes too quickly for effective in vivo application as previously discussed.
One group did study cetylmannoside (equivalent to Man0) liposomes of various compositions for the delivery of monocyte/macrophage activators [79, 80] several years before the Engel, et.al. publication. Despite the authors’ acknowledgement that monocytes had been shown not to express MR, their target cells are monocytes and they demonstrate mannose-competitive uptake by monocytes. It was not specified as to the presence or absence of serum during the incubations. Mannosylated liposomes (PC/Man0/DCP/chol-2:3:1:4) showed ~1.5-2X higher uptake than control liposomes in vitro in isolated human monocytes.
See above discussion on ManDOG. This book chapter describes a method for preparing the ManDOG derivatives with the DOG anchor already incorporated into preformed liposomes, although this particular description is for the disaturated (distearoylglycerol or DSG) anchor. The EG spacer is also elongated to 12 EG monomers. These conjugates are undergoing continued investigation as liposomal adjuvants.
mannose receptor targeting by mannosylated liposomes has been demonstrated for a variety of mannosylated lipid conjugates in a variety of liposome morphologies and compositions in several different in vitro and in vivo models. Unfortunately, only one of these publications addressed phagocyte depletion by encapsulating clodronate into the covalently mannosylated liposomes, while hundreds of publications utilize clodronate liposomes prepared in the presence of p-aminophenyl mannose under the assumption that these liposomes target mannose receptors. We are quite skeptical about the validity of this assumption for a number of reasons outlined above and recommend that researchers requiring mannose receptor targeting by clodronate liposomes select a formulation for which, minimally, mannose surface presentation by the liposomes can be verified by Con A agglutination or other mannose receptor binding in vitro . Competitive inhibition of binding by mannose or mannose-containing macromolecule further validates the specificity of mannose-receptor binding by the liposomes. We will review the results of ongoing covalently mannosylated clodronate liposome depletion studies as soon as they are published. Until a substantial collection of reproducible data on macrophage depletion using well-characterized mannosylated clodronate liposomes is established, phagocyte depletion with mannosylated clodronate liposomes requires well-designed and controlled experiments focused on the behavior of the liposomes in order to establish the value of these targeting liposomes to phagocyte depletion.
- On this page, “mannose”, “manosyl”, “mannosylated” or “mannosylation” will comprehensively refer to monomeric, oligomeric or branched configurations of mannose. The various structures are shown and discussed later on this page. Also the term “mannose receptor” may refer to any of several proteins both in solution and on cell surfaces which have mannose-selective binding domains. “MR” is used when specifically referring to the macrophage mannose receptor (CD206).
- Cryan S-A. Carrier-based strategies for targeting protein and peptide drugs to the lungs. AAPS J. 2005 Mar;7(1):E20–E41.
- Irache JM, Salman HH, Gamazo C, Espuelas S. Mannose-targeted systems for the delivery of therapeutics. Expert Opinion on Drug Delivery. 2008 Jun;5(6):703–24.
- Sihorkar V, Vyas SP. Potential of polysaccharide anchored liposomes in drug delivery, targeting and immunization. J Pharm Pharm Sci. 2001 Aug;4(2):138–58.
- Yan X, Scherphof GL, Kamps JAAM. Liposome Opsonization. Journal of Liposome Research. 2005;15(1-2):109–39.
- Kawakami S, Wong J, Sato A, Hattori Y, Yamashita F, Hashida M. Biodistribution characteristics of mannosylated, fucosylated, and galactosylated liposomes in mice. Biochimica et Biophysica Acta (BBA) – General Subjects. 2000 Dec;1524(2-3):258–65.
- Semple SC, Chonn A, Cullis PR. Interactions of liposomes and lipid-based carrier systems with blood proteins: Relation to clearance behaviour in vivo. Advanced Drug Delivery Reviews. 1998 Jun 8;32(1–2):3–17.
- Juliano RL, Stamp D. Lectin-mediated attachment of glycoprotein-bearing liposomes to cells. , Published online: 20 May 1976; | doi:10.1038/261235a0. 1976 May 20;261(5557):235–8.
- East L, Isacke CM. The mannose receptor family. Biochimica et Biophysica Acta (BBA) – General Subjects. 2002 Sep 19;1572(2–3):364–86.
- Ishida T, Harashima H, Kiwada H. Liposome Clearance. Bioscience Reports. 2002;22(2):197–224.
- Schlesinger LS. Macrophage Phagocytosis of Virulent but Not Attenuated Strains of Mycobacterium Tuberculosis Is Mediated by Mannose Receptors in Addition to Complement Receptors. J Immunol. 1993 Apr 1;150(7):2920–30.
- Takagi H, Numazaki M, Kajiwara T, Abe Y, Ishii M, Kato C, et al. Cooperation of Specific ICAM-3 Grabbing Nonintegrin-Related 1 (SIGNR1) and Complement Receptor Type 3 (CR3) in the Uptake of Oligomannose-Coated Liposomes by Macrophages. Glycobiology. 2009 Mar 1;19(3):258–66.
- Steinman RM. DC-SIGN: A Guide to Some Mysteries of Dendritic Cells. Cell. 2000 Mar 3;100(5):491–4.
- Geijtenbeek TB., Torensma R, van Vliet SJ, van Duijnhoven GC., Adema GJ, van Kooyk Y, et al. Identification of DC-SIGN, a Novel Dendritic Cell–Specific ICAM-3 Receptor that Supports Primary Immune Responses. Cell. 2000 Mar 3;100(5):575–85.
- Ahsan F, Rivas IP, Khan MA, Torres Suárez AI. Targeting to macrophages: role of physicochemical properties of particulate carriers—liposomes and microspheres—on the phagocytosis by macrophages. Journal of Controlled Release. 2002 Feb 19;79(1–3):29–40.
- Fukasawa M, Shimizu Y, Shikata K, Nakata M, Sakakibara R, Yamamoto N, et al. Liposome oligomannose-coated with neoglycolipid, a new candidate for a safe adjuvant for induction of CD8+ cytotoxic T lymphocytes. FEBS Letters. 1998 Dec 28;441(3):353–6.
- Shimizu Y, Takagi H, Nakayama T, Yamakami K, Tadakuma T, Yokoyama N, et al. Intraperitoneal immunization with oligomannose-coated liposome-entrapped soluble leishmanial antigen induces antigen-specific T-helper type immune response in BALB/c mice through uptake by peritoneal macrophages. Parasite Immunology. 2007 May 1;29(5):229–39.
- Gregoriadis G, Taylor M, Summerfield J. Mannose-mediated targeted immunoadjuvant action of liposomes. Immunology. 1988;64(4):743.
- Kawakami S, Sato A, Nishikawa M, Yamashita F, Hashida M. Mannose receptor-mediated gene transfer into macrophages using novel mannosylated cationic liposomes. Gene therapy. 2000;7(4):292–9.
- Düffels A, Green LG, Ley SV, Miller AD. Synthesis of High‐Mannose Type Neoglycolipids: Active Targeting of Liposomes to Macrophages in Gene Therapy. Chemistry – A European Journal. 2000 Apr 17;6(8):1416–30.
- Martınez-Pomares L, Mahoney JA, Káposzta R, Linehan SA, Stahl PD, Gordon S. A Functional Soluble Form of the Murine Mannose Receptor Is Produced by Macrophages in Vitro and Is Present in Mouse Serum. J. Biol. Chem. 1998 Sep 4;273(36):23376–80.
- Turner M. The role of mannose-binding lectin in health and disease. Molecular Immunology. 2003 Nov;40(7):423–9.
- Law SKA, Minich TM, Levine RP. Binding reaction between the third human complement protein and small molecules. Biochemistry. 1981;20(26):7457–63.
- Umezawa F, Eto Y. Liposome targeting to mouse brain: Mannose as a recognition marker. Biochemical and Biophysical Research Communications. 1988 Jun;153(3):1038–44.
- Tettamanti G. Ganglioside/glycosphingolipid turnover: New concepts. 2003;20(5):301–17.
- Jungalwala FB. Synthesis and Turnover of Cerebroside Sulfate of Myelin in Adult and Developing Rat Brain. J. Lipid Res. 1974 Mar 1;15(2):114–23.
- Kitatani K, Idkowiak-Baldys J, Hannun YA. The sphingolipid salvage pathway in ceramide metabolism and signaling. Cell Signal. 2008 Jun;20(6):1010–8.
- Merrill AH. Sphingolipid and Glycosphingolipid Metabolic Pathways in the Era of Sphingolipidomics. Chem. Rev. 2011;111(10):6387–422.
- Sihorkar V, Vyas SP. Potential of polysaccharide anchored liposomes in drug delivery, targeting and immunization. J Pharm Pharm Sci. 2001 Aug;4(2):138–58.
- McLean LR, Phillips MC. Cholesterol desorption from clusters of phosphatidylcholine and cholesterol in unilamellar vesicle bilayers during lipid transfer or exchange. Biochemistry. 1982;21(17):4053–9.
- Pagano RE, Weinstein JN. Interactions of Liposomes with Mammalian Cells. Annual Review of Biophysics and Bioengineering. 1978;7(1):435–68.
- Lange Y, Molinaro AL, Chauncey TR, Steck TL. On the Mechanism of Transfer of Cholesterol Between Human Erythrocytes and Plasma. J. Biol. Chem. 1983 Jun 10;258(11):6920–6.
- van Rooijen N, Sanders A. Manipulation of Kupffer cells by liposome encapsulated clodronate and propamidine—synergistic and antagonistic effects of liposomal phospholipids and drugs. International Journal of Pharmaceutics. 1998 Mar 20;162(1–2):51–8.
- Huitinga I, Van Rooijen N, De Groot C, Uitdehaag B, Dijkstra C. Suppression of experimental allergic encephalomyelitis in Lewis rats after elimination of macrophages. The Journal of experimental medicine. 1990;172(4):1025.
- Ulrich AS. Biophysical Aspects of Using Liposomes as Delivery Vehicles. Bioscience Reports. 2002;22(2):129–50.
- Engel A, Chatterjee SK, Al-arifi A, Riemann D, Langner J, Nuhn P. Influence of Spacer Length on Interaction of Mannosylated Liposomes with Human Phagocytic Cells. Pharmaceutical Research. 2003;20(1):51–7.
- Orr GA, Rando RR, Bangerter FW. Synthetic Glycolipids and the Lectin-Mediated Aggregation of Liposomes. J. Biol. Chem. 1979 Jun 10;254(11):4721–5.
- Weissig V, Lasch J, Gregoriadis G. Covalent coupling of sugars to liposomes. Biochim. Biophys. Acta. 1989 May 15;1003(1):54–7.
- Antimisiaris SG. Preparation of DRV Liposomes. In: Weissig V, editor. Liposomes [Internet]. Humana Press; 2010 [cited 2012 May 8]. p. 51–75.
- Škalko N, Bouwstra J, Spies F, Stuart M, Frederik P., Gregoriadis G. Morphological observations on liposomes bearing covalently bound protein: Studies with freeze-fracture and cryo electron microscopy and small angle X-ray scattering techniques. Biochimica et Biophysica Acta (BBA) – Biomembranes. 1998 Mar 6;1370(1):151–60.
- Ghosh P, Bachhawat B. Grafting of different glycosides on the surface of liposomes and its effect on the tissue distribution of 125 I-labelled γ-globulin encapsulated in liposomes. Biochimica et Biophysica Acta (BBA) – General Subjects. 1980 Nov;632(4):562–72.
- Castagna M. Blood Content of Rat Liver Homogenates. Nature. 1965 Feb 27;205(4974):905–7.
- Linehan SA, Martínez-Pomares L, Stahl PD, Gordon S. Mannose Receptor and Its Putative Ligands in Normal Murine Lymphoid and Nonlymphoid Organs: In Situ Expression of Mannose Receptor by Selected Macrophages, Endothelial Cells, Perivascular Microglia, and Mesangial Cells, but Not Dendritic Cells. J Exp Med. 1999 Jun 21;189(12):1961–72.
- Linehan S. The mannose receptor is expressed by subsets of APC in non-lymphoid organs. BMC Immunology. 2005;6(1):4.
- Espuelas S, Thumann C, Heurtault B, Schuber F, Frisch B. Influence of Ligand Valency on the Targeting of Immature Human Dendritic Cells by Mannosylated Liposomes. Bioconjugate Chemistry. 2008 Dec 17;19(12):2385–93.
- Kelly C, Jefferies C, Cryan S-A. Targeted Liposomal Drug Delivery to Monocytes and Macrophages. Journal of Drug Delivery. 2011;2011:1–11.
- Huitinga I, Van Rooijen N, De Groot CJ, Uitdehaag BM, Dijkstra CD. Suppression of Experimental Allergic Encephalomyelitis in Lewis Rats After Elimination of Macrophages. J Exp Med. 1990 Oct 1;172(4):1025–33.
- Buiting AMJ, Zhou F, Bakker JAJ, van Rooijen N, Huang L. Biodistribution of clodronate and liposomes used in the liposome mediated macrophage “suicide” approach. Journal of Immunological Methods. 1996 Jun;192(1-2):55–62.
- Chono S, Tanino T, Seki T, Morimoto K. Efficient drug targeting to rat alveolar macrophages by pulmonary administration of ciprofloxacin incorporated into mannosylated liposomes for treatment of respiratory intracellular parasitic infections. Journal of Controlled Release. 2008 Apr;127(1):50–8.
- Chono S, Tanino T, Seki T, Morimoto K. Uptake characteristics of liposomes by rat alveolar macrophages: influence of particle size and surface mannose modification. Journal of Pharmacy and Pharmacology. 2007 Jan 1;59(1):75–80.
- Chono S, Kaneko K, Yamamoto E, Togami K, Morimoto K. Effect of surface-mannose modification on aerosolized liposomal delivery to alveolar macrophages. Drug Development and Industrial Pharmacy. 2010 Jan;36(1):102–7.
- Derksen JTP, Morselt HWM, Scherphof GL. Processing of different liposome markers after in vitro uptake of immunoglobulin-coated liposomes by rat liver macrophages. Biochimica et Biophysica Acta (BBA) – Molecular Cell Research. 1987 Oct 22;931(1):33–40.
- Shao J, Ma JKH. Characterization of a mannosylphospholipid liposome system for drug targeting to alveolar macrophages. Drug Delivery. 1997 Jan;4(1):43–8.
- De Diesbach P, N’Kuli F, Delmée M, Courtoy PJ. Infection by Mycoplasma Hyorhinis Strongly Enhances Uptake of Antisense Oligonucleotides: A Reassessment of Receptor‐Mediated Endocytosis in the HepG2 Cell Line. Nucl. Acids Res. 2003 Feb 1;31(3):886–92.
- Goldman R, Raz A. Concanavalin A and the in vitro induction in macrophages of vacuolation and lysosomal enzyme synthesis. Experimental Cell Research. 1975 Dec;96(2):393–405.
- Fidler IJ. Targeting of immunomodulators to mononuclear phagocytes for therapy of cancer. Advanced Drug Delivery Reviews. 1988 Oct;2(1):69–106.
- Thomann J-S, Heurtault B, Weidner S, Brayé M, Beyrath J, Fournel S, et al. Antitumor activity of liposomal ErbB2/HER2 epitope peptide-based vaccine constructs incorporating TLR agonists and mannose receptor targeting. Biomaterials. 2011 Jul;32(20):4574–83.
- Heurtault B, Frisch B, Pons F. Liposomes as delivery systems for nasal vaccination: strategies and outcomes. Expert Opinion on Drug Delivery. 2010 Jul;7(7):829–44.
- Ghosh P, Das PK, Bachhawat BK. Targeting of liposomes towards different cell types of rat liver through the involvement of liposomal surface glycosides. Archives of Biochemistry and Biophysics. 1982 Jan;213(1):266–70.
- Driessen NN, Ummels R, Maaskant JJ, Gurcha SS, Besra GS, Ainge GD, et al. Role of Phosphatidylinositol Mannosides in the Interaction Between Mycobacteria and DC-SIGN. Infect. Immun. 2009 Oct 1;77(10):4538–47.
- Muller CD, Schuber F. Neo-mannosylated liposomes: Synthesis and interaction with mouse Kupffer cells and resident peritoneal macrophages. Biochimica et Biophysica Acta (BBA) – Biomembranes. 1989 Nov 17;986(1):97–105.
- Barratt G, Tenu J-P, Yapo A, Petit J-F. Preparation and characterisation of liposomes containing mannosylated phospholipids capable of targetting drugs to macrophages. Biochimica et Biophysica Acta (BBA) – Biomembranes. 1986 Nov 6;862(1):153–64.
- Banerjee G, Nandi G, Mahato SB, Pakrashi A, Basu MK. Drug delivery system: targeting of pentamidines to specific sites using sugar grafted liposomes. Journal of Antimicrobial Chemotherapy. 1996;38(1):145.
- Kawakami S, Sato A, Nishikawa M, Yamashita F, Hashida M. Mannose receptor-mediated gene transfer into macrophages using novel mannosylated cationic liposomes. Gene therapy. 2000;7(4):292–9.
- Düffels A, Green LG, Ley SV, Miller AD. Synthesis of High‐Mannose Type Neoglycolipids: Active Targeting of Liposomes to Macrophages in Gene Therapy. Chemistry – A European Journal. 2000 Apr 17;6(8):1416–30.
- Shimizu Y, Nakata M, Matsunuma J, Mizuochi T. Simultaneous quantification of components of neoglycolipid-coated liposomes using high-performance liquid chromatography with evaporative light scattering detection. Journal of Chromatography B: Biomedical Sciences and Applications. 2001 Apr 15;754(1):127–33.
- Sugimoto M, Ohishi K, Fukasawa M, Shikata K, Kawai H, Itakura H, et al. Oligomannose-coated liposomes as an adjuvant for the induction of cell-mediated immunity. FEBS Letters. 1995 Apr 17;363(1–2):53–6.
- Frisch B, Hassane FS, Schuber F. Conjugation of Ligands to the Surface of Preformed Liposomes by Click Chemistry. In: Weissig V, Walker JM, editors. Liposomes [Internet]. Humana Press; 2010 [cited 2012 May 18]. p. 267–77.
- Allen TM, Hansen C, Martin F, Redemann C, Yau-Young A. Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochimica et Biophysica Acta (BBA) – Biomembranes. 1991 Jul 1;1066(1):29–36.
- Nag A, Ghosh PC. Assessment of Targeting Potential of Galactosylated and Mannosylated Sterically Stabilized Liposomes to Different Cell Types of Mouse Liver. Journal of Drug Targeting. 1999 Jan;6(6):427–38.
- Martin FJ, Papahadjopoulos D. Irreversible Coupling of Immunoglobulin Fragments to Preformed Vesicles. An Improved Method for Liposome Targeting. J. Biol. Chem. 1982 Jan 10;257(1):286–8.
- Dumont S, Muller CD, Schuber F, Bartholeyns J. Antitumoral properties and reduced toxicity of LPS targeted to macrophages via normal or mannosylated liposomes. Anticancer Res. 1990 Feb;10(1):155–60.
- Proffitt RT, Williams LE, Presant CA, Tin GW, Uliana JA, Gamble RC, et al. Liposomal Blockade of the Reticuloendothelial System: Improved Tumor Imaging with Small Unilamellar Vesicles. Science. 1983 Apr 29;220(4596):502–5.
- Torchilin VP, Goldmacher VS, Smirnov VN. Comparative studies on covalent and noncovalent immobilization of protein molecules on the surface of liposomes. Biochemical and Biophysical Research Communications. 1978 Dec 14;85(3):983–90.
- Espuelas S, Haller P, Schuber F, Frisch B. Synthesis of an amphiphilic tetraantennary mannosyl conjugate and incorporation into liposome carriers. Bioorganic & Medicinal Chemistry Letters. 2003 Aug 4;13(15):2557–60.
- Engel A, Chatterjee SK, Al‐Arifi A, Nuhn P. Influence of spacer length on the agglutination of glycolipid‐incorporated liposomes by ConA as model membrane. Journal of Pharmaceutical Sciences. 2003 Aug 25;92(11):2229–35.
- Schmidt M, Chatterjee S., Dobner B, Nuhn P. New modified single chained glycolipids. Part 1: synthesis of deoxy and partially O-methylated glycolipids with or without a sulfur containing spacer. Chemistry and Physics of Lipids. 2002 Feb;114(2):139–47.
- Hiroshi K, Hitochi N, Yuichi F, Sunju Y, Yuriko K. Application of Synthetic Alkyl Glycoside Vesicles as Drug Carriers. I. Preparation and Physical Properties. Chemical & pharmaceutical bulletin. 1985 Feb 25;33(2):753–9.
- Yamashita C, Sone S, Ogura T, Kiwada H. Potential Value of Cetylmannoside‐modified Liposomes as Carriers of Macrophage Activators to Human Blood Monocytes. Cancer Science. 1991 May 1;82(5):569–76.
- Childs RA, Drickamer K, Kawasaki T, Thiel S, Mizuochi T, Feizi T. Neoglycolipids as probes of oligosaccharide recognition by recombinant and natural mannose-binding proteins of the rat and man. Biochem J. 1989 Aug 15;262(1):131–8.
- Mizuochi T, Loveless RW, Lawson AM, Chai W, Lachmann PJ, Childs RA, et al. A Library of Oligosaccharide Probes (neoglycolipids) from N-Glycosylated Proteins Reveals That Conglutinin Binds to Certain Complex-Type as Well as High Mannose-Type Oligosaccharide Chains. J. Biol. Chem. 1989 Aug 15;264(23):13834–9.
- Shimizu Y, Yamakami K, Gomi T, Nakata M, Asanuma H, Tadakuma T, et al. Protection against Leishmania major infection by oligomannose-coated liposomes. Bioorganic & Medicinal Chemistry. 2003 Apr;11(7):1191–5.
- Kojima N, Biao L, Nakayama T, Ishii M, Ikehara Y, Tsujimura K. Oligomannose-coated liposomes as a therapeutic antigen-delivery and an adjuvant vehicle for induction of in vivo tumor immunity. Journal of Controlled Release. 2008 Jul 2;129(1):26–32.
- Opanasopit P, Sakai M, Nishikawa M, Kawakami S, Yamashita F, Hashida M. Inhibition of liver metastasis by targeting of immunomodulators using mannosylated liposome carriers. Journal of Controlled Release. 2002 Apr 23;80(1–3):283–94.
- Wijagkanalan W, Higuchi Y, Kawakami S, Teshima M, Sasaki H, Hashida M. Enhanced anti-inflammation of inhaled dexamethasone palmitate using mannosylated liposomes in an endotoxin-induced lung inflammation model. Molecular pharmacology. 2008;74(5):1183.
- Wijagkanalan W, Kawakami S, Takenaga M, Igarashi R, Yamashita F, Hashida M. Efficient targeting to alveolar macrophages by intratracheal administration of mannosylated liposomes in rats. Journal of Controlled Release. 2008 Jan;125(2):121–30.
- Fellowes R, Etheridge CJ, Coade S, Cooper RG, Stewart L, Miller AD, et al. Amelioration of established collagen induced arthritis by systemic IL-10 gene delivery. Gene Therapy. 2000 May 25;7(11):967–77.
- Miller AD. Unlocking Mechanisms in Gene Therapy, Stress and Proteomics. In: Schneider MP, editor. Chemical Probes in Biology Science at the Interface of Chemistry, Biology and Medicine [Internet]. Springer Netherlands; 2004 [cited 2012 May 23]. p. 297–316.
- Ahl PL, Bhatia SK, Meers P, Roberts P, Stevens R, Dause R, et al. Enhancement of the in vivo circulation lifetime of l-α-distearoylphosphatidylcholine liposomes: importance of liposomal aggregation versus complement opsonization. Biochimica et Biophysica Acta (BBA) – Biomembranes. 1997 Oct 23;1329(2):370–82.
- Kishore U, Greenhough TJ, Waters P, Shrive AK, Ghai R, Kamran MF, et al. Surfactant proteins SP-A and SP-D: Structure, function and receptors. Molecular Immunology. 2006 Mar;43(9):1293–315.
- Kerrigan AM, Brown GD. C-type lectins and phagocytosis. Immunobiology. 2009 Jul;214(7):562–75.