Hermanson G. Bioconjugate Techniques, Second Edition

Подождите немного. Документ загружается.

860 22. Preparation of Liposome Conjugates and Derivatives

The conﬁ guration the bilayers can assume also can be complex. A bilayer may be in a spheri-

cal form, having one layer of hydrophilic head groups pointing outward toward the surrounding

solution and the second hydrophilic layer pointing inward toward a compartment of aque-

ous solution sequestered within the sphere. The morphology of a liposome may be classiﬁ ed

according to the compartmentalization of aqueous regions between bilayer shells. If the aqueous

regions are segregated by only one bilayer each, the liposomes are called unilamellar vesicles

(ULV) ( Figure 22.2 ). If there is more than one bilayer surrounding each aqueous compartment,

the liposomes are termed multilamellar vesicles (MLV). ULV forms are further classiﬁ ed as to

their relative size, although rather crudely. Thus, there can be small unilamellar vesicles (SUV;

usually thought of as less than 100 nm in diameter, with a minimum of about 25 nm) and large

unilamellar vesicles (LUV; usually greater than 100 nm in diameter, with a maximal size of about

2,500 nm). With regard to MLV, however, the bilayer structures cannot be as easily classiﬁ ed due

Figure 22.2 The highly varied morphologies of lipid bilayer construction.

to the almost inﬁ nite number of ways each bilayer sheet can be associated and interconnected

with the next one. MLVs typically form large complex honeycomb structures that are difﬁ cult

to categorize or exactly reproduce. However MLVs are the simplest to prepare, most stable, and

easiest to scale up to large production levels.

Small lipid groupings or monomers also may fuse into bigger, inverted micelles, wherein

their hydrophobic tails point outward toward other inverted micelle lipid tails. The individ-

ual structures are usually hexagonal in shape, but typically they exist as large groupings of

inverted micelles, the outer edge of which contains partial inverted micelles, exposing their

inner hydrophilic heads to the surrounding aqueous environment.

The most useful form of liposomes for bioconjugate applications consists of small, spher-

ical ULVs that possess layers of hydrophilic head groups on their inner and outer surfaces.

The inside of each vesicle can contain hydrophilic molecules that are protected from the outer

environment by the lipid shell. The outside surface can be derivatized to contain covalently

attached molecules designed to target the liposome for speciﬁ c interactions.

1.2. Preparation of Liposomes

Mixtures of phospholipids in aqueous solution will spontaneously associate to form liposomal

structures. To prepare liposomes having morphologies useful for bioconjugate or delivery tech-

niques, it is necessary to control this assemblage to create vesicles of the proper size and shape.

Many methods are available to accomplish this goal, however all of them have at least several

steps in common: (1) dissolving the lipid mixture in organic solvent, (2) dispersion in an aque-

ous phase, and (3) fractionation to isolate the correct liposomal population.

In the ﬁ rst stage, the desired mix of lipid components is dissolved in organic solvent (usually

chloroform:methanol (2:1 by volume)) to create a homogeneous mixture. This mixture will

include any phospholipid derivatized to contain reactive groups as well as other lipids used to

form and stabilize the bulk of the liposomal structure. During all handling procedures using

lipids or their derivatives, it is essential that the solutions be protected from oxidation and

excessive exposure to light, especially the sun. Organic solvents should be maintained under a

nitrogen or argon atmosphere to prevent introduction of oxygen. Water and buffers should be

degassed using a vacuum and bubbled with inert gas before introducing lipid components.

The correct ratio of lipid constituents is important to form stable liposomes. For instance, a

reliable liposomal composition for encapsulating aqueous substances may contain molar ratios of

lecithin:cholesterol:negatively charged phospholipid (e.g., phosphatidyl glycerol (PG)) of 0.9:1:0.1.

A composition that is typical when an activated phosphatidylethanolamine (PE) derivative is

included may contain molar ratios of phosphatidylcholine (PC):cholesterol:PG:derivatized PE of

8:10:1:1. Another typical composition using a maleimide derivative of PE without PG is PC:male-

imide-PE:cholesterol of 85:15:50 (Friede et al., 1993). In general, to maintain membrane stability,

the PE derivative should not exceed a concentration ratio of about 1–10 mol PE per 100 mol of

total lipid.

An example of a lipid mixture preparation based on mass would be to dissolve 100 mg of PC,

40 mg of cholesterol, and 10 mg of PG in 5 ml of chloroform/methanol solution. When using

activated PE components, inclusion of 10 mg of the PE derivative to this recipe will result in a

stable liposome preparation.

1. Properities and Use of Liposomes 861

862 22. Preparation of Liposome Conjugates and Derivatives

Once the desired mixture of lipid components is dissolved and homogenized in organic sol-

vent, one of several techniques may be used to disperse the liposomes in aqueous solution. These

methods may be broadly classiﬁ ed as (1) mechanical dispersion, (2) detergent-assisted solubiliza-

tion, and (3) solvent-mediated dispersion.

Probably the most popular option is mechanical dispersion, simply because the greatest

number of methods that utilize it have been developed. When using mechanical means to form

vesicles, the lipid solution ﬁ rst is dried to remove all traces of organic solvent prior to dis-

persion in an aqueous media. The dispersion process is the key to producing liposomal mem-

branes of the correct morphology. This method uses mechanical energy to break up large lipid

agglomerates into smaller vesicles having the optimal size and shape characteristics necessary

for encapsulation or bioconjugation.

Mechanical dispersion methods involve adding an aqueous solution (which may contain

substances to be encapsulated) to the dried, homogeneous lipid mixture and manipulating it to

effect dispersion. Major methods of mechanical dispersion include simple shaking (Bangham

et al., 1965), non-shaken aqueous contact (Reeves and Dowben, 1969), high-pressure emulsi-

ﬁ cation (Mayhew et al., 1984), sonication (Huang, 1969), extrusion through small-pore mem-

branes (Szoka et al., 1980), and various freeze–thaw techniques (Pick, 1981). Some devices

are available commercially which automate the mechanical dispersion process, usually by high-

pressure emulsiﬁ cation or sonication (Branson Ultrasonics Corp.).

Most of these methods result in a population of vesicles ranging from SUVs of only 25 nm

diameter to very large MLVs. Classiﬁ cation of the desired liposomal morphology may be

done by chromatographic means using columns of Sepharose 2B or Sepharose 4B, by density-

gradient centrifugation using Ficoll or metrizamide gradients, or by dialysis.

Liposome formation by detergent-assisted solubilization utilizes the amphipathic nature of

detergent molecules to bring more effectively the lipid components into the aqueous phase for

dispersion. The detergent molecules presumably bind and mask the hydrophobic tails of lipids

from the surrounding water molecules. Detergent treatment may take place from a dried lipid

mixture or after formation of small vesicles. Usually, nonionic detergents such as the Triton X

family, alkyl glycosides, or bile salts such as sodium deoxycholate are employed for this pro-

cedure. The immediate structures which form as the detergent molecules solubilize the lipids

from a dried state are small micelles. Upon removal of the detergent from the solution, the lipid

micelles aggregate to create larger liposome structures. Liposomes of up to 1,000 Å containing a

single bilayer may be formed using detergent-assisted methods (Enoch and Strittmatter, 1979).

Unfortunately, some detergent-removal processes also may remove other molecules that were to

be entrapped in the liposomes during formation.

Solvent-mediated dispersion techniques used to create liposomes ﬁ rst involve dissolving the

lipid mixture in an organic solvent to create a homogeneous solution, and then introducing this

solution into an aqueous phase. The solvent may or may not be soluble in the aqueous phase to

effect this process. There also may be components dissolved in the aqueous phase to be encap-

sulated in the developing liposomes.

Perhaps the simplest solvent dispersion method is that developed by Batzri and Korn (1973).

Phospholipids and other lipids to be a part of the liposomal membrane are ﬁ rst dissolved in

ethanol. This ethanolic solution then is rapidly injected into an aqueous solution of 0.16 M KCl

using a syringe, resulting in a maximum concentration of no more than 7.5 percent ethanol.

Using this method, single bilayer liposomes of about 25 nm diameter can be created that are

indistinguishable from those formed by mechanical sonication techniques. The main disadvan-

tages of ethanolic injection are the limited solubility of some lipids in the solvent (about 40 mM

for PC) and the dilute nature of the resultant liposome suspension. However, for the prepara-

tion of small quantities of SUVs, this method may be one of the best available.

Other solvent dispersion methods utilize solvents that are insoluble in the aqueous phase.

The key to the production of liposomes by this procedure involves the formation of a “ water-in-

oil ” emulsion. To create the proper reverse-phase emulsion, a small quantity of aqueous phase

must be introduced into a large quantity of organic phase containing the dissolved liposomes.

The result is a milky dispersion containing the “ homogenized ” liposomes. A number of tech-

niques have been developed to perform this procedure (Kim and Martin, 1981; Kim et al., 1983;

Pidgeon et al., 1986). The emulsiﬁ cation process in each of these solvent-dispersion techniques

involves the use of mechanical means (shaking, stirring, or sonication) to effect the formation of

small droplets of aqueous solution uniformly dispersed in the lipid–organic phase.

For the preparation of large quantities of liposomes, mechanical dispersion using a commer-

cially available emulsiﬁ er is probably the best route. For limited quantities, the use of simple

shaking or ethanolic dispersion techniques works well.

Regardless of their method of fabrication, most liposome preparations need to be further

classiﬁ ed and puriﬁ ed before use. To remove excess aqueous components that were not encap-

sulated during the vesicle formation process, gel ﬁ ltration using a column of Sephadex G-50 or

dialysis can be employed. To fractionate the liposome population according to size, gel ﬁ ltra-

tion using a column of Sepharose 2B or 4B should be done.

Small liposome vesicles often aggregate upon standing to form larger, more complex struc-

tures. Therefore, long-term storage in aqueous solution is usually not possible without major

transformations in liposome morphology. Freezing also fractures the liposomal membrane,

releasing any entrapped substances. The inclusion of cryoprotectants such as sugars or poly-

hydroxylic-containing compounds can overcome the structural degradation problems upon

freezing (Harrigan et al., 1990; Talsma et al., 1991; Park and Huang, 1992). Presumably, the

hydroxyl groups in cryoprotectants can take the place of water in hydrogen bonding activities,

thus providing structural support even under conditions in which water is removed. A procedure

by Friede et al. (1993) allows freezing and lyophilization of SUVs in the presence of 4 percent

sorbitol with complete retention of liposome integrity upon reconstitution. Thus freeze-drying

may be the best method for the long-term storage of intact liposomes.

1.3. Chemical Constituents of Liposomes

The overall composition of a liposome—its morphology, chemical constituents (including a

large variety of phospholipids and other lipids or fatty acids), charge, and any attached func-

tional groups—can affect the properties of the vesicle both in vitro and in vivo (Allison and

Gregoriadis, 1974; Alving, 1987; Therien and Shahum, 1989). Although there are literally doz-

ens of lipid components that potentially can be included in a liposomal recipe, only a handful

are commonly used.

Phospholipids are the most important of these liposomal constituents. Being the major com-

ponent of cell membranes, phospholipids are composed of a hydrophobic, fatty acid tail, and

a hydrophilic head group. The amphipathic nature of these molecules is the primary force that

drives the spontaneous formation of bilayers in aqueous solution and holds the vesicles together.

1. Properities and Use of Liposomes 863

864 22. Preparation of Liposome Conjugates and Derivatives

Naturally occurring phospholipids can be isolated from a variety of sources. One of the

most common phospholipid raw materials is egg yolk. However, since the composition of egg

phospholipid is from a biological source and can vary considerably depending on age of the

eggs, the diet of the chickens, and the method of processing, newer enzymatic and synthetic

chemical methods now are being employed to manufacture the required phospholipid deriva-

tives in higher purity and yield.

Two main forms of lipid derivatives exist biologically: molecules containing a glycerol back-

bone and those containing a sphingosine backbone. The most important type for liposomal

construction is a phosphodiglyceride derivative, which consists of a glycerol backbone that links

two fatty acid molecules with a polar head group ( Figure 22.3 ). The fatty acids are acyl bonded

in ester linkages to the Nos. 1 and 2 carbon hydroxyls of the glycerol bridge. The No. 3 carbon

hydroxyl of the glycerol group is phosphorylated and possesses a negative charge at physiologi-

cal pH. This basic phosphodiglyceride construct of two fatty acids and one glycerylphosphate

group is called phosphatidic acid. This is the simplest form of phospholipid available.

The fundamental phosphatidyl group also can be further derivatized at the phosphate to

contain an additional polar constituent. Several common derivatives of phosphodiglycerides are

naturally occurring, including PC; commonly called lecithin, phosphatidyl ethanolamine (PE),

phosphatidyl serine (PS), phosphatidyl glycerol (PG), and phosphatidyl inositol (PI) ( Figure

22.4 ). All of these phospholipids have polar groups that are linked to the phosphatidyl moiety

in a phosphate ester bond. The most abundant of these derivatives in biological cell membranes

is PC—the trimethyl derivative of PE, possessing a positive charge at physiological pH. Some or

all of these phosphodiglyceride derivatives can be mixed to create a particular liposomal recipe.

The fatty acid constituents of phosphodiglycerides can vary considerably in nature among

a number of different chain lengths and points of unsaturation. A given isolated phosphatidyl

Figure 22.3 The basic construction of phosphodiglyceride molecules within lipid bilayers. The fatty acid chains

are embedded in the hydrophobic inner region of the membrane, oriented at an angle to the plane of the mem-

brane surface. The hydrophilic head group, including the phosphate portion, points out toward the hydrophilic

aqueous environment.

derivative from a biological source usually possesses a range of fatty acid components, varying in

chain length from C16 to about C24. Some of the fatty acids also may contain points of unsatu-

ration—one or more double bonds between certain carbon atoms within the chain (Matreya

and Avanti Polar Lipids, suppliers). For instance, egg lecithin is not a single compound, but con-

tains a mixture of PC containing about 31 percent saturated fatty acid having a chain length

of 16 carbons, 16 percent saturated fatty acid with 18 carbons, about 48 percent also with 18

carbons but having at least 1–2 points of unsaturation, and the rest a variety of other fatty acid

constituents. Naturally occurring, unsaturated fatty acids typically are of the cis conformation,

not trans. The existence of unsaturation within a fatty acid is usually abbreviated as the chain

length followed by a colon and the number of double bonds. For instance, cis-9-hexadecenoic

acid (palmitoleic acid) contains one double bond at carbon 9 and it is abbreviated as C16:1.

By contrast, a given synthetic preparation of a major phospholipid possesses fatty acid constit-

uents all of identical chain length and unsaturation. A synthetic PC derivative can be purchased

that contains only, for instance, 1,2-dimyristoyl (C14) fatty acid substitutions on its glyceryl

Figure 22.4 The head-group construction of the six commonly encountered phosphatidyl derivatives.

1. Properities and Use of Liposomes 865

866 22. Preparation of Liposome Conjugates and Derivatives

Figure 22.5 The three fatty acid components commonly used in liposome construction.

backbone (Genzyme). The use of synthetic rather than natural phospholipids for making lipo-

somes thus produces reagents of known chemical purity, which is very important for regulatory

requirements surrounding the introduction of products used topically or in vivo .

Despite the large variety of potential fatty acid components in natural-occurring phosphod-

iglycerides, only three major fatty acid derivatives of synthetic phospholipids are commonly

used in liposome preparation: (1) myristic acid ( n-tetradecanoic acid; containing 14 carbons),

(2) palmitic acid ( n-hexadecanoic acid; containing 16 carbons), and (3) stearic acid ( n -octade-

canoic acid; containing 18 carbons) ( Figure 22.5 ).

The nomenclature for associating individual fatty acid groups with particular phosphodig-

lyceride derivatives is straightforward. For instance, a phosphatidic acid (PA) derivative which

contains two myristic acid chains is commonly called dimyristoyl phosphatidic acid (DMPA).

Likewise, a PC derivative containing two palmitate chains is called dipalmitoyl phosphatidyl

choline (DPPC). Other phosphodiglyceride derivatives are similarly named.

The second form of lipid derivative that occurs naturally in membrane structures is derived

from sphingosine. Unlike the phosphodiglyceride derivatives discussed above, sphingo-

lipids contain no glycerol backbone. Instead, these lipids are constructed from a derivative of

4-sphingenine, containing an N-acyl-linked fatty acid group and possibly other constituents

off the No. 1 carbon hydroxyl group ( Figure 22.6 ). Sphingolipids are highly similar in their

construction to glyceryl lipids, in that there are two hydrophobic tails present on a 3-carbon

backbone (one of them contributed from 4-sphingenine itself and the other from the attached

fatty acid), and there also exists a hydrophilic head group. This creates the typical amphipathic

properties common to all lipid membrane components.

The simplest form of sphingolipid, ceramide, contains a fatty acid group, but no additional

components on the No. 1 hydroxyl. Major derivatives of ceramide at the 1-hydroxyl position

include a positively charged phosphocholine compound, called sphingomyelin, a glucose deriv-

ative, called glucosylcerebroside, and other complex carbohydrate derivatives, termed gan-

gliosides ( Figure 22.7 ). Gangliosides are involved in various cellular recognition phenomena,

including being part of the blood group determinants, A, B, and O, in humans.

The use of sphingolipids in liposome formation is possible due to the natural amphipathic

properties of the molecules. Some sphingolipids can lend structural advantages to the integrity

Figure 22.6 Sphingolipids are constructed of sphingosine derivatives containing an acylated fatty acid and a

head group attached to the hydroxyl.

Figure 22.7 Common sphingolipid derivatives include small and highly complex head groups.

868 22. Preparation of Liposome Conjugates and Derivatives

Figure 22.8 The orientation of cholesterol in phospholipid bilayers.

of liposomal membranes. Sphingomyelin, for example, is capable of hydrogen bonding with

adjacent glyceryl lipids, thus increasing the order and stability of the vesicle construction. This

stability may translate into a lower potential for passage of molecules through the membrane

bilayer, forming vesicles that are better able to retain their contents than more ﬂ uid liposome

constructions. The temperature of phase transition in sphingolipid-containing membranes is

often greater than membranes constructed of only phosphodiglyceride derivatives. Liposomes

containing sphingomyelin or gangliosides also have prolonged lifetimes in vivo (Gregoriadis

and Senior, 1980; Allen and Chonn, 1987) and may be advantageous for creating liposome

immunogen complexes.

The main disadvantage of incorporating sphingolipids in liposomes is their high cost.

Puriﬁ ed phosphodiglyceride derivatives may be obtained in bulk quantities and in highly

deﬁ ned synthetic preparations, whereas sphingolipid derivatives are not so readily available in

similar purity.

Another signiﬁ cant component of many liposome preparations is cholesterol. In natural cell

membranes, cholesterol makes up about 10–50 percent of the total lipid. For liposome prepa-

ration, it is typical to include a mole ratio of about 50 percent cholesterol in the total lipid rec-

ipe. The addition of cholesterol to phospholipid bilayers alters the properties of the resultant

membrane in important ways. As it dissolves in the membrane, cholesterol orients itself with

its polar hydroxyl group pointed toward the aqueous outer environment, approximately even,

in a three-dimensional sense, with the glyceryl backbone of the bilayer ’s phosphodiglyceride

components ( Figure 22.8 ). Structurally, cholesterol is a rigid component in membrane construc-

tion, not having the same freedom of movement that the fatty acid tails of phosphodiglycerides

possess. Adjacent phospholipid molecules are restricted in their freedom of movement through-

out the length of their fatty acid chains that are abutting the cholesterol molecules. However,

since the cholesterol components have the effect of creating spaces in the uniform hydrophobic

morphology of the bilayer, the portion of the fatty acid chains below the abutted regions are

increased in their freedom of movement.

Cholesterol ’s presence in liposome membranes has the effect of decreasing or even abolish-

ing (at high cholesterol concentrations) the phase transition from the gel state to the ﬂ uid or

liquid crystal state that occurs with increasing temperature. It also can modulate the permeabil-

ity and ﬂ uidity of the associated membrane—increasing both parameters at temperatures below

the phase transition point and decreasing both above the phase transition temperature. Most

liposomal recipes include cholesterol as an integral component in membrane construction.

1.4. Functional Groups of Phospholipids

For the production of liposomal conjugates, lipid derivatives must be incorporated into the

bilayer construction that contain available functional groups able to be chemically crosslinked

or modiﬁ ed. A number of phosphodiglyceride compounds can be employed for conjugation

purposes. Each of these components contains a head group that can be directly derivatized or

chemically modiﬁ ed to contain a reactive group. For instance, several lipid derivatives contain

amine groups that can be utilized in nucleophilic reactions with crosslinkers or other modiﬁ ca-

tion reagents. These include PE, PS, and stearylamine. Carboxyl-containing molecules include

all the individual fatty acids as well as PS. These can be coupled to amine-containing mole-

cules by the use of the carbodiimide reaction (Chapter 3, Section 1). Hydroxyl-containing lipids

include PG, fatty acid alcohols, PI, and various gangliosides and cerebrosides of sphingolipid

derivation. Lipids possessing hydroxyl groups on adjacent carbon atoms, such as those contain-

ing sugar constituents, may be oxidized with sodium periodate to produce reactive aldehyde

residues (Chapter 1, Section 4.4). Coupling aldehydes to amine-containing molecules can be

accomplished by reductive amination (Chapter 3, Section 4). Finally, the phosphate groups of

PA residues may be used to conjugate with amine-containing molecules in a method similar

to modiﬁ cation of the 5 -phosphates of DNA probes. This is done through the use of the car-

bodiimide reaction with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) (Chapter 27,

Section 2.1). Figure 22.9 shows the structures and reactive sites of these lipid functional groups.

When liposomes are used as part of a conjugate system, the targeting molecules usually are

attached covalently to these head group functional groups using standard crosslinking chemis-

tries (Derksen and Scherphof, 1985). Although all of the above-mentioned functional groups

on lipid molecules can be used for the conjugation process, most often the derivatization reac-

tion is done off the PE constituents within the liposomal mixture. The primary amine modiﬁ ca-

tion off the glycerylphosphate head of PE provides an ideal functional group for activation and

subsequent coupling of targeting or detection molecules (Shek and Heath, 1983). Liposomes

may be constructed with reactive groups already prepared on their PE constituents, all set to

be conjugated with selected molecules having the correct functional group. Stock preparations

of activated liposomes may even be prepared and lyophilized to be used as needed in coupling

macromolecules (Friede et al., 1993). All of the amine-reactive conjugation methods discussed

in this section may be used with PE-containing liposomes.

2. Derivatization and Activation of Lipid Components

Two approaches for the activation of lipid components may be used to create reactive groups in

liposomes. A puriﬁ ed lipid may be activated prior to incorporation into the bilayer construction

2. Derivatization and Activation of Lipid Components 869