Cover Image

PAPERBACK
$29.00



View/Hide Left Panel
Click for next page ( 156


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 155
MEMBRANE MODELS: E\tOL{JTION FROM IRE EWIDMOSAIC STANDBY ROY HAMMERS TEDT INTRODUCTION Postulation of the fluid-mosaic model for membrane structure by Singer and Nicholson (1972) represented a thoughtful and reasoned synthesis of information obtained from observations extending over loo years. These many contributions, summarized by Robertson (1981), began with the implicit recognition by Schleiden and Schwann that a membrane structure Rust bound the cell, and includes contributions from physicists, chemists and morphologists. Other key elements were: use of evolving techniques to highlight new aspects of the complex structure leg., electron microscopy); careful quantitative analyses of the sizes and numbers of the postulated elements of the membrane (protein, sterol and phospholipids); and characterization of the physical features of membranes, either as isolated or after resolution into individual components. The resulting model has served as an framework upon which a host of experiments have been conducted to further illuminate the role of the interface between the cell interior and its environment or between compartments within the cell. Since the presentation of the fluid-mosaic model, continuous application of pre-existing plus new techniques has resulted in additional quantitative details that must be integrated into the model. Careful rereading of this classic paper, as opposed to simple inspection of the much imitated schematic Figures (see Figure 1, left panel), reveals that the authors clearly understood the limitations to their proposal and anticipated some of the modifications proposed later by others (Figure 1, right panel). Figure 1. Schematic representation of bilayer membrane. The progressive evolution of our concepts of the structure of the lipid bilayer is illustrated by the initial presentation (Singer and Niclolson, 1972) on the left and an "updated" version (Cullis and Hope, 1985) on the right. Figures reprinted with permission of the publishers. The fluid mosaic model retained earlier concepts leg., a lipid bilayer by Danielle and Davson) while providing a new way of interpreting the distribution of protein. Proteins were considered to penetrate into the lipid ~ in an iceberg fashion) leaving most of the inner and outer surfaces as naked lipid. This mosaic pattern provided a portion of the name. The second aspect, fluidity, encompassed the observations of Frye and Edidin (1970) on the ability of fluorescent dyes to move freely in the membrane. These two aspects (fluidity and mosaic) remain dominant features of the evolving membrane model. 155

OCR for page 155
Contributions of others will be reviewed herein to illustrate: an increased awareness of the complexity of the outer surface of the plasma membrane; the asymmetric distribution of lipid components across the bilayer, and the effect that these distributions have on the physical properties of the membrane; and the lateral organization (domain structure) of the membrane. These revisions are especially important for understanding of membrane events on the surface, or plasma, membrane. Final comment" will be directed to the design and interpretation of experiments related to reproduction and will emphasize the dynamics of time frame and spatial relationships. COMPLEXITY OF THE CELL SURFACE The concept of integral ( intrinsic) and peripheral . . . (extrinsic) proteins found in the original fluid-mosaic model illustrates an awareness that the external face of the membrane was complex, and the differential staining of the outer surface of cells (Revel, et al, 1960) provided strong evidence for the specialized distribution of carbohydrate on the cell surface. The growing array of biophysical techniques available to study the properties of cell surfaces (pp 31-48 of Rao, 1987 ~ provide a large body of evidence that the surface membranes of cells differ, and that ~layers" of unique molecules extend from the bilayer into the surrounding media. Molecules in this succession of layers (Figure 2) "shield" the bilayer from the surrounding solution. Exposure to external phospholipases resulted in greatly different extents of hydrolysis of phospholipids of erythrocytes and erythrocyte ghosts (Zwall and Roelofson (1976) and Ottolenghe (1973) ) . This was because of shielding, as illustrated in Figure 2. Figure Z. Schematic representation of the layers found on the cell surface. This scale drawing illustrates the types of molecules, and . their spacla1 relatlonsnlps, Anal can exist on one surface or a cell. The frequency of appearance of the proteins that extend from the bilayer to form the glycocalyx or are adsorbed is unknown, and probably differs among cell types. The relative sizes of soluble components of molecular weights of 200 and 60,000 are illustrated. Adapted from Susko-Parrish, et al., 1985. 156 -

OCR for page 155
PUS'M ~EMB~ PHO~HOUPlD PLAYER it, 1 IS :~ O-YCOCALYX "SORSED . PG3ON PROTEIN ,~ 25~ 1~ - ~ . ~ i: - _~ - - ~St=~ ~! - 1~ -- .14 7 157 O O.7nm ( - Snm

OCR for page 155
This succession of layers undoubtedly exists on the surface of many-to-most cells, but the distribution of the components and their density on the surface is difficult to establish. Adsorbed protein senre'; as ~ general descriptor for many of these macromolecules, and the necessity of washing cells before analysis wil 1 remove loosely bound components before the analysis begins. These complicating factors make interpretation of data obtained from such suspensions very dif f icult . Hormone-receptor binding analyzes often reveal alterations in the number of receptor site" as a function of physiological state. In many cases, but certainly not all, subsequent analyses established that biosynthetic and degradative processes resulted in removal of (or addition to) receptors from the bilayer. Inspection of Figure 2, which is drawn to scale, shows how access of an extracellular ligand (the large sphere represents ~ molecule of about 60,000 daltons) to the bilayer is dependent on the disposition of other surface components exterior to the bilayer. In contrast, the approach of a small molecule (eg. glucose; "mall sphere) is less likely to be impeded e Such hinderances often are not considered in the analysis of surface binding phenomena. If the ligand for study has only one class of binding sites, the binding properties can be accurately assessed. Unfortunately, multiple classes of binding sites introduce complexities unrecognized by most investigators. The reader should consult the articles of Klotz (1982, 1983) for examples and details. One example of multiple sites is the analysis of lectin binding mites on the sperm surface. Susko-Parish et al. (1985) tested the use of iodinated lectin and classical Skatchard analysis for this purpose; no reliable quantitative data were obtained. Subsequently, Magargee et al (1988) utilized flow cytometry, under carefully standardized conditions, for cell-by-cell analysis of the same phenomenon and were able to guantitate relative binding to the sperm surface for a variety of lectins. ASYMMETRY OF CELL SURFACES This concept also was considered in the original fluid-mosaic model, where the pictorial displays clearly indicate that components in the membrane were unequally distributed relative to the exterior and interior faces (Figure 11. Further, based on staining properties mentioned previously, carbohydrate components (eg., gangliosides) were selectively enriched on the exterior face of the bilayer. The major contribution of the past decade was the delineation of the orientation of the phospholipids of the bilayer. These concepts are lucidly described in a review by Cullis and Hope (1985) and the texts by Houslay and Stanley (1982) and Jain (19873; the reader is encouraged to study these contributions for the primary references and exposure to the limitations to the general statements presented below. t Analysis of the lipid components in membranes involves: isolation of the cell (s) of interest free of ~significant" contaminants; removal and purification of membranes, including providing proof of their point of origin in the cell; extraction of the lipids from the recovered membrane vesicles; and analysis of the lipid classes. Methods to accomplish these interdependent steps have been developed, and the 158

OCR for page 155
resulting data base describes the mayor lipids for a variety of cell and membrane types. The large variety specific molecular forms leg., phospholipid types with a variety of acyl substituents) has been bewildering, in that no general structural solution to the common biologic.] problem of assembly of a stable bilayer was apparent. Major contributions of the last decade have been in identifying common features of these components to allow general statements about membrane structure. The phospholipid asymmetry of plasma membranes for five cell types is presented in Figure 3. For a membrane of the dimensions of ~ cell, a stable bilayer will have the phospholipids equally divided between the outer and inner surfaces (monolayers or leaflets). However, the individual phospholipid types need not be equally distributed toward the inner and outer monolayers. This certainly is true for the plasma membrane, where phosphatidy~choline (PC) and sphingomyelin (SPM) are preferentially oriented ~out" with phosphatidylethanolamine (PE), phosphatidylserine (PS) and phosphatidylinositol (PI) preferentially oriented gins. Since the phospholipid classes differ in the general types of acyl substituents, with PC and SPM having the more saturated members and PE, PS and PI having the more unsaturated components, this asymmetry in lipid by head group also introduces asymmetry in the hydrocarbon portion of the bilayer. To a first approximation, this asymmetry between outer and inner leaflets could result in differences in flexibility gradient (freedom of motion proceeding from the bilayer face into the interior) for the exterior and interior faces. Model system studies also establish that cholesterol (CHOL) preferentially interacts with SPM and PC relative to PE, PS and PI, thereby introducing the possibility of a asymmetric distribution of the sterols. Definitive proof of the asymmetric distribution of cholesterol in cells has been difficult because of a dearth of methods to establish sterol localization, but the data for model systems argue strongly for this feature. Figure 3: Phospho] ipid distributions for surface membranes . The percentage of the phosphor ipids that are oriented to the outer vs the inner faces of the plasma membrane are shown for : A, human erythrocyte; B. rat liver sinusoidal surface; C, rat liver continuous surface; D, pig platelet; and E, BHK-21 cells. Abbreviations for the lipids are presented in the text. Taken from Cullis and Hope (1985) with permission of the publisher. ~: ;1 _ _ - 159 -

OCR for page 155
Detailed studies of model lipid dispersions have proven most useful to the resolution of these difficulties, in that precise spectroscopic and microscopic methods can be applied to the analysis of lipid assembly into phases. A clear representation of several possibilities (Figure 4 considers two general cases . First and most predominant, a situation where molecules cannot exhibit total freedom of motion in all dimensions is classified as an anisotropic systems (bilayer and hexagonal (Hal) ~ . The second, where complete freedom of motion can occur, is an isotropic mode. Specific examples include a bilayer (orientation of headgroup out, and respective acyl side chains toward each other and intercalated), the hexagonal phase (polar head group'; in and acyl side chains out) and any of 'several orientations (vesicles, micellar, etc. ~ where ache members have isotropic motion. Each is recognizable by virtue of its nuclear magnetic resonance (~) spectroscopic features and the images gathered by freeze fracture electron microscopy (EM) analysis. Detailed studies of many types of lipids have established that each has a preference for the phase it will adopt when placed in an aqueous environment. Some lipids (PC, SPM, PS (at pH > 3 ), PI) prefer the bilayer while others (PE , PS flat pH ~ 3 ), CHOL) prefer the hexTI orientation. If a membrane was composed of only one-lipid, reasonably accurate predictions of the phase of the membrane formed could be made. Even then, predictive ability is limited as evident from observations that addition of secondary components ~ cholesterol or long-chain unsaturated fatty acids ~ to l ipids preferring the bilayer form will induce a transition to a hexagonal orientation. [inure 4. Unique NMR spectra and freeze-fracture patterns for various phospholipid phases. Preparation of lipid dispersions to yield putative bilayer, hexanonal and isotropic phases, followed by analyses by NMR (center panel) and electron microscopy (right panel) established the unique features of each phase. Taken from Cullis and Hope (1988) with permission of the publisher. P. ',l~li~ - _ C. ala. 81, '4~' a_C~ a I. V~ - 1 Id - Ca At 5. ~ "~ - C~_~_ ~ a,~c_ ~ ~ j~ - .'_ ~ if_ I V_ I_ I_ ~3 - 160 -

OCR for page 155
Considerable attention has been given to the changes in the degree of molecular ordering observed when temperature of the lipid phase is altered. At low temperatures the bilayer exists in a crystalline state, where the acyl chains are ordered. Raising the temperature then results in an alteration of state (to the liquid-crystalline array) where the acy} chains are disordered. Much attention has been focused on these transitions in phase state, most often characterized using techniques such as differential scanning calorimetry (Figure 53. Abrupt changes in molecular order of the acyl groups with increasing temperature are observed for pure components, but addition of other lipids leg., cholesterol) eliminate these sharp transitions. The net effect for actual cellular membranes is that no such sharp transition occurs, and crystalline domains are unlikely to exist. Figure 5, Relationship between lipid composition and phase transition. Careful heating of lipid preparations during analysis by differential scanning calorimetry tests for the temperature of conversion from the crystalline to the liquid-crystalline state. Pure lipids (DPPC, diphosphatidylphosphatidy~choline) undergo this transition at: an unique temperature, but introduction of increasing amounts of a second 1 ipid (cholesterol ~ results in a much broader range in the transition temperature. Adapted from a presentation by Houslay and Stanley (19821. {~-RRR??R 3~S - ~ 86688b mar. ~ . - e ~~ - r _,_,___ -_" L~S~E "~ SOLID Lull PLUM Lo 1 OILERS +~ - Lo ~ _~ ~ ~ TOOLS ~TEROl_ 80 ~% , , , ,~ + SO IDOLS C~StE~ t90 "O TE~RAT~ (~1 - 161

OCR for page 155
To recapitulate, detailed biophysical studies have established that quite di';~;iailar lipids, when place in water and allowed to assemble into aggregate';, assume one of a limited Amber of overall orientations. Since the hexagonal phase would not provide the essential permeability barrier for demarcating the cell boundary, this; form cannot Provide the Thor form in the cell membrane. The bilayer form is _ preferred. Recent studies, however, are consistent with the conclusion that localized and transitory formation of hexagonal forms may be important in the function of the intact membrane. These studies of model systems have uncovered another feature of lipids that may further simplify analysis of the interaction of lipids in the membranes. This is the concept that lipid interactions, and therefore patterns of assembly into large scale structures, can be interpreted in terms of the geometric shape of individual lipid molecules. - The primary presentations (Israelachvili, et al., 1980, Ruypers, et al., 1984 and Gruner, 1985) establish that the dominant features are the molecular volume, head group area, and depth of the hydrocarbon side chains. From these considerations, the lipids can be grouped into molecular shapes and ultimately to predict the Phase into which they will assemble (Figure 63. _ A allayer could be envisioned as a collection of snapes, some with a predisposition to assume one of the other phase forms (Figure 7~. Figure 6. Types of lipid packing shapes assumed by various lipids. Lipids can be grouped according to molecular parameters of molecular volume, surface area and depth of penetration into the bilayer. These considerations yield a limiting number of geometric shapes (cylinders, cones and inverted cones ~ which, in turn, can be related to the type of polymorphic phase form the system will assume when placed into water. See text for further details and references. POLYMORPHIC GEOMETRIC LIPID PHASE F - M SHAPE TtOYLC~L 1 - _ TIOYLSE ltlNE ~~, ) ~ ,,, _ IDYLB405'TOL \) i ~ I, l; l,' ~ ~ ' ~ ~ ~ TID - GLYCEROL OF . .... &TIOdC ACIO C-~010LI-N ILAYER eYLI~ICaL . POSIT IDY L ~~ - LA-lNE CARDIOLIPIN-COZ4 ~ ~ I' PROS - UTlOIC ACID-Co24 ~~ _~ _OIDYLSE. - E ~ .. . (as MID) X&CO - AL {~} __ LYIO~PIOS ~ V ~ICLLaR I - ERTE0 CONE - 162 l -

OCR for page 155
Eigure 7. Schematic representation of the assembly of lipid into the bilayer. A collection of lipid shapes, selected from those presented in Figure 6, have been assembled to form a bilayer. Such a description using rigid geometrical forms i'; imperfect in that gaps between lipids would exist {see outer leaflet). These gaps could be accomodated by the flexibility of the lipids as well as by the insertion of proteins into the membrane. OUTER LEAFLET INNER LEAFLET MODEL MEMBRANE (WITHOUT PROTEIN ~ Results of recent studies on differentiation of stem cells to erythrocytes (Rayler et al., 1985) reveal changes in the "inner vs outer" display of the polar headgroups of phospholipids (Figure 8~. Such changes provide one mechanism to greatly alter the surface during cellular differentiation. A second striking example (Kuypers, et al. 1984; Op den Kamp et al., 1985) is the x~ vitro incubation of erythrocytes, under conditions where surface lipids were exchanged without,tranabilayer movement from inner to outer bilayer leaflet, and a resulting gross alteration of the shape of the erythrocyte. This alteration can be explained by exchanging 1 ipid molecules residing in the outer leaflet of the bilayer. Such a 'substitution would result in an alteration of the surface area of that leaflet, and force a change in the overall shape of the cell (Figure 9~. Figure 8. Schematic representation of the changes in the orientation of E~osphol~ ids during mouse erythroid differentiation. Differentiation of stem cells into erythrocytes has been intensively studied, with Friend cells considered by some investigators as a model for an intermediate stage in the process. Data provided by Rayler, et al. (1985) establish that the cell types differ in their orientation of phospholipids toward the outer and inner leaflets. - 163

OCR for page 155
~ so - lo 25 IL lo 1,~. Z 25 ALSO C f-RYTHROID DIFFERENTIATION (MOLISE) STEM CELL ~ | VIRUS ~ tFRIEND CELLS1 BURSt FORMING COLONY fORMING TOTAL ~OUTSIDE. INSIDE PC+SM P PSond Pl n ~ ma, ~~--- .4 50 OUTER LEAFLET 25 O 25 50 INNER LEAFLET ,.,. ~% . (ERYTHRocyTs I Figure 9. Effect of substitution of lipids of the outer leaflet of erythrocytes on the overall shape of the cell. Incubation of erythrocytes with donor vesicles of containing unique phosphatidylcholine molecules (differing in the acyl substituents, 16:0 vs 18:2) and phospholipid exchange proteins allows selective alteration of the outer surface of the bilayer (provided the time of the experiment is limited to less that that required for transhilayer movement). Such treatment converts the normal erythrocyte (center, bottom) to distinct forms. Taken from Op den Kamp, et al. (198S) with permission of the publisher. OUr5IDE . not_ 6:0,18:1 PC MY t6:0,t8:2;PC Q In '-~" Qua Q dlC te:0 PC | = t) C) = ~dICl8:2 PC ,~ it_ it_

OCR for page 155
Controlled fusion of membranes represents a common membrane event, often associated with fundamental processes critical for reproduction. Careful studies of model systems are providing suggestions for the mechanisms) by which these processes occur; no single mechanism has yet emerged fro. these studies (see Ohki et al., 1987~. One potential mechanism involves inverted micelle forms, transiently produced by relocation of selected lipid components within the bilayer, to provide those essential features (localized points of stress, integration and relief of stress) necessary for formation of the new membrane forms (Figure 10~. Figure lo. Model fox the role of minor lipids in the fusion of lipid vesicles. Lipid vesicles (panel a) can come together (panel b) and bring their lipid bilayers into molecular contact. Minor lipids , with a propensity for formation of hex II structures, could play a key role in the interaction of the membrane bilayers (panel c) before separation or fusion (panel d). Taken with Cullis and Hope (1988) with permission of the pub] isher. .~.~ttI~ (S~ Existing data clearly establish that phospholipid asymmetry exists on cellular membranes and that this condition can persist over the lifetime of the cell. Such an unequal distribution carries with it the implication of energy expenditure to maintain disequilibrium. This could be accomplished by processes such as direct enzymatic action (phospholipid exchange enzymes moving lipids preferentially to one side of the bilayer) or synthesis and retention of proteins possessing unique binding specificities for the phospholipid head groups. Data support both concepts, but no evidence uniquely supports any single mechanism. To summarize this section, asymmetric orientation of phoppholipids on the separate bilayer leaflets exists. Their tendencies to form unique molecular aggregates can result in atlered cell functions and shape. These distributions phospholipids provides an excellent framework for the design and interpretation of experiments dealing with a variety of membrane related events. - 16S -

OCR for page 155
LIPID DOMAINS This concept, also anticipated by Singer and Nicholson, focuses on the extent to which the membrane is considered to be ~fluid". Recent data clearly establish that the term must be used carefully. Fluidity is the inverse of viscosity. The implicit assumption, derived from the dictionary definition, is that the membrane would have an equivalent bulk viscosity in all dimensions. Limitations to use of the term for membrane structures have been discussed (see Lands, 19803. The predominant restriction is that freedom of motion in the membrane is very d~.ension-dependent in that lateral (two-dimensional on the surface), translational (across the bilayer) and rotational (around the axis of the acyl groups) motions each have their own unique features. This results in situations where molecules (eg., phospholipids of the bilayer boundary) can have great freedom of lateral motion but highly restricted translational motion. The general texts cited in the preceding section, as well a" the review by Wolf (1988) and sections of the book by Loew (1988), provide an excellent introduction to the topic. Even after acceptance of these 1 imitations , it now is apparent that more stringent restrictions to movement of individual components within the membrane exist. They include restrictions due to phase segregation, lipid-lipid interactions, lipid-protein interactions, or morphologically recognized demarcation between portions of the cell. Examples of phase separations are found in the data gathered from model systems of two and three lipid component mixtures. The predictions from the Gibbs' phase rule establish that at a given temperature a mixture of components will segregate to form unique gel phases (each enriched in specific components of the original mixture). These could tend to form domains within the membrane. The second form, lipid-lipid interactions, was mentioned previously; an example would be the apparent preferential interaction of cholesterol with selected phospholipids that occur as part of the total lipid mixture. Specif ic lipid-protein interactions also occur, although direct experimental proof of the extent of the interactions and the factors responsible for the interaction is lacking. The last feature is recognition of morphologically distinct portions of the membrane, where points of restriction (eg., tight junctions) preclude movement from one section of the surface to the other. Direct proof of these aspects, much less the molecular basis for the restrictions, has been difficult to accumulate. However, the current implication that a variety of important cellular processes (membrane bound enzyme activation, sperm capacitation, receptor movements) are regulated in part by these membrane features assures that the topic will be intensively studied in the near future.- RELATIVE SPEED OF MEMBRANE EVENTS Progress in the study membrane events related to reproductive processes undoubtably will proceed as a form of derivative science, where data from other areas (eg., erythrocyte and lymphocyte membrane - 166 -

OCR for page 155
structure-function; model liposome characterization) will ';erve as examples for experimental design. This is desirable. Progress with blood cells is easier because these cells have simplified membrane features and they are actively studied by a relative abundance of experienced investigators. The next two sections outline precautionary atatoments regarding this approach. First, what is the most appropriate display of data to establish cause-and-effect relationships (Figure 113? The tendency is to array data in terms of the maximum total response, leading to the assignment of order of events of A before B before C (in the left panel of Figure 113. This conclusion implies a known stoichiometry of interaction between the successive elements in the series and that no "overshoot" in the production of enzyme products occurs. While this may be true for soluble enzymes, the factors that regulate the activity of membrane-bound enzymes are less well known. Indeed, since they operate in a mileau where conformational adj ustment~ may be slowed by 'surrounding molecules, it in highly probable that production of product beyond that needed for initiation of the next step wil 1 occur . The essential relationship is to supply ~enough" product to satisfy the binding needs for the next step in the pathway. One example involves the sequence of events that follow activation of adenylcyclase, where absolute amount of cAMP produced in the first step is not important once ~ concentration sufficient to saturate the binding proteins involved in the next step of the sequence is achieved. The power of retrospective analysis, possible after detailed study of the sequence, reveals the quantitative coupling between individual steps. This provides the revised interpretation of data in the right panel of Figure 11. Thus, hypotheses should not prematurely die because of our lack of understanding of the stoichiometry of membrane related events. The next point of consideration is the effect of the time axis on analysis of data (Figure 11~. Testing of a hypothesis requires that experimental observations be made with sufficient rapidity to clearly establish the kinetic relationships between factors of interest. The elegant hypotheses related to events of reproduction often are tested by evaluating features related to changes in membrane permeability, movement of molecules in/on the membrane or modification of surfaces. A listing of such events (Figure 12), when scaled to a common axis (logarithmic, in seconds), is embarassingly informative. It alerts us to the extreme rapidity of membrane events described in other portions of this paper and to the fact that all too often the true ';cale of measurements is inappropriate for the events to be studied. 167 -

OCR for page 155
Figure 11. Effect of selection of axis dimensions on the _assignment of sequence in a metabolic process. For ~ simulated system, data for response of three enzymes (evaluated by assay of products of each) after addition of a stimulant are given (left panel). For each enzyme, the cats were normalized to the maximu" Amount of product detected over the total time of the experiment. Precursor-product assumptions would suggest the relationship of A before B before C. It is possible that a detailed examination of the stoichiometry of the individual steps relative to overall pathway needs (see text for examples) would reveal that C is produced in vast excess, and A in moderate excess relative to B. Such considerations would be consistent with the sequence C before A before 8. 100 x ~ so I. a o A ~ B TIC - 158 ___' ' _' ~ j , , -- C- HA ~ ~ o TIME 0F INTEREST

OCR for page 155
n9,~7 ~.,~ the time ~c.te-D! evened ~cc~rin~ on membranes to those of reproduction. Events intern membranes differ greatly in time, ranging from picoseconds (10 ~ to months (10 seconds) while molecular and organismal events of reproduction are confined to a much smaller range of time. The effects of such considerations on the design of experiments are discussed in the text. Adapted from a presentation by Jain (1987~. In -~ - _~e . ~~ ~ At_ tVtitT~ OUTHIT_T~ ,~O rear - - - r art no) ~ "Ttl"L 0~1 me AXIAL ale Maytag "YtR - ~ ~ J At~ ~ Actual OF - . 4_L ~ 1~L~. tow "L=tS CYCLE OF - .tK. ~P - - ul C~ , A_ O- ~#. ~ C~L "OTtlU .~_ ' Atetat ~~t l~Yt. At - - U~t Of CtLL At_ [~ - ~ 1~ - ~ - It) CORAL ~ItAC7 - , .~u Blew -~UCLt" ream c - aatar" Of apt" 1~ CcENTa~r rovers INSET `,rest CLEa^ - S or r - To c" , - ~~" T.ANS~ t~ TO ~~. "XUAL ~~ - - <- . - ~" "U ~ ~LEO ~ ~17~ I. .~) - The comparison in Figure 11 will be used for two illustrations. In the first, very unique experimental designs are essential for any direct test of a postulated relationship of regulation of a cellular event involved in reproduction via a mechanism such as alteration of Nat - 169

OCR for page 155
movement through channels . Since these processes dif fer in speed by at least four orders of magnitude, no single incubation will test the desired relationship. Since direct test is impossible, elimination of alternative possibilities is most difficult. The required experimental design, use of several experiments featuring overlapping analyses of several cellular events to provide the desired temporal relationship, is rarely used. The second illustrates an opposite effect. Review of many discussions of tranabilayer movement of phospholipids leaves the impression that these processes are so slow that they may be of minimal physiological significance. This may be untrue for events such as the epididymal maturation of sperm, in that the 7-14 days required for the process are sufficient to allow for extensive movement of phospholipids between the inner and outer leaflets of the bilayer. It is evident that study of membrane events is very immature, especially for complex membrane processes related to reproduction. Great care must be taken when adapting hypotheses derived from studies on more simple systems to assure that the extrapolation is consistent with additional information about the biological process. SPATIAL RELATIONSHI PS ON THE MEMBRANE SURFACE Cell-to-cell interactions, such as sperm-egg binding and fusion, reflect the end point of an extensive differentiation process whereby the surfaces are modified to allow specific molecular interactions. Katz, Cone and colleagues (Katz , et al ., 1987 ; 1988 ~ have calculated the forces necessary to tether a motile sperm to the oocyte, and provide an order-of-magnitude estimate of a requirement of up to 10 noncovalent bonds between the respective surfaces. This serves as a minimum estimate, since mouse sperm binding to the zone pellucida may ultimately involve 10-50 x 10p molecules per sperm head (see review by Wassarman, 1987~. Since it is possible to induce specific membrane vesiculation processes, related to the acrosome reaction, using liposome surfaces that are much smaller than those of the egg (Graham et al, 1986; Graham and Hammerstedt, 1988), the following questions merit consideration. What are the spatial requirements for a specific interaction, with special emphasis on relating the body of microscopic data with the emerging biochemical description of the cell surface? Given the large difference in size of a liposome and an oocyte, are comparisons of events occurring on each surface directly comparable? The scale drawings in Figure 13 illustrate both comparisons. On the left is a model of bovine spend binding to an oocyte, while in the center (at 10X increase in scale) is binding of a liposome to a bovine sperm. At this level, direct comparison of events between the sperm surface and an oocyte vs a 1 iposome appear inappropriate . However, when viewed from the perspective of the bonds to be formed (right panel, at a further 100X increase in scale) it is clear that significant interactions of the sperm surface can occur with either surface. The 10 bonds needed for tethering wil 1 come from portions of the surface that are indistinguishable using microscopic techniques. These illustrative examples serve to show the difficulty of direct comparisons of data gathered using diverse techniques ranging from microscopy to biochemical and biophysical analyses. 170

OCR for page 155
Figure 13, Size relationships of the cells and structures involved in ~ _ sperm egg fusion. In panel A a bovine sperm is shown superimposed on a bovine egg, in panel B a liposome is shown bound to the sperm head and in panel C the liposome is shown relative to the bilayer of sperm surface (see Figure 21. The marker bars illustrate the lOOOX increase in magnification. The effects of these considerations on the design and interpretation of experiments is presented in the text. A ~ ~ BO,OOOnm ; \ . B', (~?11~ (a 2,000nm ~ 1 SUMMARY Introduction of the fluid-mosaic model of a membrane in 1972 remains a landmark contribution to cell biology. As to be expected, subsequent contributions have refined the model by reemphasizing and clearly depicting the modes of molecular heterogeneity possible on any given surface. These unique orientations and motional properties of the molecules ultimately will allow discrimination among superficially similar, but uniquely regulated, membrane events. Data from model systems aid understanding the potential characteristics of membrane surfaces involved in fundamental processes of reproduction. However, the challenge to those interested in reproduction is to make appropriate extrapolations from a model systems to the complex cell-cell interactions of interest. The objective of this overview is to introduce recent contributions that provide a startpoint for interpretation of data on changes in membrane composition. As an example, Parks and Ha:mmerstedt ~1985 ) described changes, associated with epididymal transit, in 1 ipid composition of the plasma membrane overlying the acrosome of ram sperm. Concomitant with this passage, sperm acquire the capacity to undergo the acrosome reaction (Williams, et al., 19873. The challenge is to devise an appropriate means to test for a relationship between changes in structure (lipid and protein) and changes in function (ability for specific vesiculation). Unique interpretations have not been provided, since if the reported lipid changes are considered in terms of the concepts of membrane fluidity, ~ prediction of an increase, decrease or no change in this parameter can be made! The conclusion drawn depends on the model system accepted for comparison. The alteration in 1 ipid composition should introduce a change in the asymmetry of lipids across the bilayer. This could have extreme ef feats on the interaction of thi s - i71 -

OCR for page 155
sperm 'surface with other surfaces. Distinctions among the various possibilities requires careful extrapolation from model studies, with a greater emphasis on direct evaluation of the physical features of the membranes of interest. ACKNOWLEDGEMENTS Financial support was provided by the National Institutes of Health (HD 13099) and the United States Department of Agriculture (GAM 85025353. Dr. James Graham provided essential suggestions for Figures 12 and 13. - 172 -

OCR for page 155
REFERENCES Cullis, P. R. and M. Hope 1985 Physical properties and functional roles of lipids in membranes. pp 25-72 in Biochemistry of Lipids and Membranes, D. Vance and J. Vance eds. Menlo Park, CA: Benjamin/Cummings Publishing. Cullis, P. R. and M. J. Hope 1988 Lipid -polymorphism, lipid asymmetry and membrane fusion. in Molecular Mechanisms of Membrane Fusion, S. Ohki, D. Doyle, T. D. Flanagan, S. W. Hui, and E. Nayhew eds. Plenum Press New York pp 37-51. Frey, L. D. and M. Edidin, 1970 The rapid intermixing of cell surface antigens after formation of mouse-human heterokaryons . J . Cel 1 Sci. 7: 319-335. Graham, J. K., R H. Foote and J . J . Parrish 19 8 6 Ef feet of dilaurylphosphatidylcholine on the acrosome reaction and subsequent penetration of bull spermatozoa into zong-free hamster eggs. Biol Reprod 35: 413-427 . Graham, J. F. and R. H. Hammerstedt 1988 Induction of motility, the acrosome reaction and egg penetration in epididymal ram sperm by liposomes. Society Study Reproduction, 37 (Suppl 1) Abst 309. Gruner, S. M. 1985 Intrisic curvature hypothesis for biomembrane lipid composition: A role for nonbilayer lipids. Proc. Natl. Acad. Sci. (USA) 82: 3665-3669 . Houslay, M. D. and Stanley, K. K. 1982 Dynamics of Biological Membranes New York: John Wiley and Sons. Israelachvili, J. N., S. Marcel ja, and R. G. Horn 1980 Physical principles of membrane organization. Quart. Rev. Biophysics 13: 12 1-2 00 . Jain, M., K. 1987 Introduction To Biological Membranes, 2nd Edition, John Wiley and Sons, New York. Katz, D. F., E. Z. Drobnis, G. N. Cherr, J. Baltz, A. I. Yudin, R. A. Cone, and L. Y . Cheng 1987 The biophysics of sperm penetration of the cumulus and zone pellucida. Pp 275-285. in New Horizons in Sperm Cell Research, H. Mohri ed. Tokyo: Japan Sci . Soc . Press . Ratz, D. F., E. Z. Drobnis, J. Baltz, and N. DeMestre 1988 Analysis of forces generated by sperm against the zone pellucida. Society Study Reproduction 38 (Suppl 1) Abstract 285 Rlotz, I.M 1982 2lumber of receptor sites for Scatchard graphs: facts and fantasies. Science (USA) 217 :1247-1249. Klotz, I. M. 1983 Nu~her of receptor sites from Scatchard and Klotz graphs: a constructive critique. Science (USA) 220: 981. 173

OCR for page 155
Ruypers, F. A., B. Roelofsen, W. Berendsen, J. A. F. Op den Ramp, and L. L. M. van Deenen 19 8 4 Shape changes in human erythrocytes induced by replacement of the native phosphatidylcholine with species containing various fatty acids J. Cell Biol. 99: 2260-2267. I`ands, W. E . H. 19 8 0 Fluidity of Membrane Lipids Pp 69 -7 3 in Membrane Fluidity eds M. Mates and A. Kukses Humana Press, Clifton NJ Lowe, L. M. 1988 Spectroscopic Membrane Probes. Vol 1 Boca Raton FL: CRC Press. Magargee , S . ~ ., M. E . Runze and R. H . Hammerstedt 19 8 8 Changes in lectin-binding features of ram sperm surfaces associated with epididymal maturation and ejaculation. Biol. Reprod. 38: 667-685. Ohki, S., D. Doyle, T. D. Flanagan, S. W. Hui, and E. Mayhew 1987 Molecular mechanisms of membrane fusion. New York: Plenum Press. Op den Kamp, J. A. F., B. Roelofsen , and L. L. M. van Deeden Structural and dynamic aspects of phosphatidylchol ine in the human erythrocyte membrane. Trends Biochem. Sci. 10: 320-323. Ottolenghi, A. 1973 Preparation and characterization of mouse intestinal phospholipase. Lipids 8: 415-425. Parks, J. E. and R. H. Hammerstedt 1985 Developmental changes occurring in the lipids of ram epididymal spermatozoa plasma membrane. Biol. Reprod. 32: 653-668. Rao, R. V. 1987 Role of cell surface in development. Vol 1 Boca Raton FL: CRC Press. Ray~er, A., P. H. van der Schaft, B. Roelofsen, J. A. F. Op den Ramp 1985 Phospholipid localization in the plasma membrane of Friend erythroleukemic cells and mouse erythrocytes. Biochem. 24: 1777-1783 e Revel, J. P., L. Napolitano and D. W. Fawoett 1960 Identification of glycogen in electron micrographs of thin tissue sections. J. Biophys. Biochem. Cytol. 8: S75-589 . Robertson, J. D. 1981 Membrane structure. J. Cell Biol. 91: 189s-204s. Singer, S. J. and G. L. Nicholson 1972. The fluid mosaic model of the structure c~f cell membranes. Science (USA) 175: 720-731. Susko-Parish, J. S., P. L. Senger and R. H. Hammerstedt 1985 Binding of 12S T-succinylated concanavalin A to bovine spermatozoa. Biol. Reprod. 32: 129-136. - 174 -

OCR for page 155
Wassarman, P. M. 1987 Early events in mammalian fertilization. Ann. Rev. Cell Biol. 3: 109-142. Williams, R. M., J. E. Graham and R. H. Hammerstedt 1987 In vitro induction of the acrosome reaction and hamster egg penetration in epididymal and ejaculated ram sperm. Society Study Reproduction 36 (Supp] I} Abstract 68. Wolf, D. E. 1988 Probing the lateral organization and dynamics of membranes . Pp 193-22 0 in Spectroscopic Membrane Probes, L. M. Loew, ed Boca Raton, FL: CRC Press. Zwall, R. F. A. ~ B. Roelofsen' 1976 Applications of pure phospholipases in membrane studies. in Biochemical Analysis of Membranes. Pp 352-377 ed A. H. Maddy New York: John Wiley and Son 175 - .