<<

ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 10, No. 3 Copyright © 1980, Institute for Clinical Science, Inc.

Acanthocytosis—Biochemical and Physiological Considerations

JAMES J. BIEMER, M.D.

Pathology Department, St. Joseph’s Hospital, and University of South Florida, College of Medicine Tampa, FL 33677

ABSTRACT

Acanthocytosis represents an unusually pathological variant of red cell morphology which is encountered in a diverse group of inherited and ac­ quired disease states. While the morphological features are similar in all instances, the biochemical lesions frequently differ. Most demonstrable abnormalities involve lipids although those acanthocytes associated with the McLeod phenotype are probably due to an alteration in a membrane protein. Acanthocytes, regardless of their etiology, usually have a decreased survival in the circulation owing to splenic sequestration and destruction.

Introduction Young red cells, known as reticulocytes, with a frequently folded excess mem­ A great deal of interest has been gener­ brane, persist approximately two days in ated by observation of the wide array of the peripheral blood while their cyto­ pathological and physiological forms that plasmic organelles are discharged, and can be assumed by the normally bicon­ each undergoes remodeling with sym­ cave human , the discocyte. metrical membrane loss to assume its To survive its normal 100 to 120 days’ life normal discocyte configuration. As the span, the red cell must undertake a con­ cell ages, a series of changes occur includ­ tinuous circulatory journey, approximat­ ing additional membrane loss, increased ing 175 miles, frequently requiring corpuscular hemoglobin concentration negotiation of capillaries and slit-like because of water and cation loss, de­ spaces as small as 1/20 its diameter. Its creased enzyme activity, increased biconcave configuration with optimal methemoglobin content and decreased surface-to-volume ratio enhances its deformability. It is probably largely be­ ready deformability to negotiate those cause of the latter that these senescent tight spaces without damage, and also best primarily now spherical cells are detected serves its gas exchange function with its and destroyed by the as they at­ freely movable molecular hemoglobin tempt to pass its narrow passages. The vast content (figure 1). majority of normal circulating cells are, 238 0091-7370/80/0500-0238 $01.80 © Institute for Clinical Science, Inc. ACANTHOCYTOSIS 2 3 9

F ig u r e 1. T h e m a rk e d alteration of shape is por­ trayed as a normally deformable erythrocyte negotiates a tight interen- dothelial cell space of the spleen (10,000 x).

however, discocytes which, when ob­ The report of Brecher and Bessis9 has served while flowing in vivo, assume a well illustrated the stages of discocyte- vast variety of dynamic transitions in form eehinocyte transformation and elluci- largely related to flow rates and vessel dated many of the factors associated with size.31 formation. Extrinsic factors The following discussion pertains to de­ causing crenation include plasma incu­ formed red cells which are spiculated and bated at 37°C for 24 hours, lysolecithin, specifically known as acanthocytes. high levels of fatty acid and many others. These must be differentiated from a vari­ Intrinsic factors include aging of red cells ety of red cells known to have one or more which is probably related to depressed spiny projections. Many of these cells are adenosine triphosphate, (ATP) and wash­ of well defined pathogenesis, e.g., sickle ing cells in saline and the “glass effect” of cells associated with sickle cell hemoglo­ observing cells between slide and cover bin; schizocytes (helmet, fragmented or slip. Brecher and Bessis feel that echino- triangular cells), associated with micro­ cytes probably occur in various diseases angiopathic hemolytic anemia where the but that previous reports of such must be cells are injured by intravascular fibrin carefully reevaluated by examination of strands, diseased vessel walls, or by car­ fresh cells between plastic cover slips to diac valve prostheses; and tear drop cells exclude artifactual crenation. (), typically associated with Acanthocytes (spur or aeanthoid cells), myelofibrosis or abnormal hematopoiesis. on the other hand, are distinctly different Lastly, there is the echinocyte, (burr, cre- cells having 5 to 10 spicules of varying nated or berry cell), which is an ovoid or length irregularly distributed over the red spherical cell with 10 to 30 spicules cell surface (figure 2). The individual evenly distributed over its surface. spicules have knobby ends. Under the 2 4 0 BIEMER

which will then stand out in sharp contrast to acanthocytes which will be converted to acantho-echinocytes, rec­ ognizable by bifurcation of the spicules. Lastly, scanning electron microscopy fol­ lowing immediate fixation of fresh blood has provided the reference standard by which the subtle diferences in spiculated cells have been defined.4 While originally described as one of the diagnostic hallmarks of abetalipopro­ teinemia (ABL),3 acanthocytes have been described in a number of seemingly re­ lated and unrelated conditions, which

F ig u r e 2. Artistic composite illustrating a cen­ will be discussed (see table I). Regardless trally placed discocyte with acanthocytes at the 12,2, of their etiology, they almost invariably and 7 o’clock positions and echinocytes at 4 and 9 have a decreased life span although this o ’clock. may range from a severe hemolytic anemia requiring transfusions to a mild, light microscope on an air dried smear, barely detectable compensated hemolytic state. For a thorough discussion of the im­ acanthocytes and echinocytes may occa­ sionally be difficult to differentiate. How­ plications of altered red cell shape, their membranes, deformability, and hemolytic ever, typical echinocytes have a serrated outline with small projections more or less anemia, the reader is referred to the fol­ evenly spaced over the circumference of lowing comprehensive reviews; Weed,54 the red cells while acanthocytes have a Shohet and Ness,48 Lessin et al31 and few spicules of varying length and thick­ Mohandas et al.39 ness projecting irregularly from the cell surface. Red Cell Membranes—General Wet preparations have the advantage of Background allowing more of a three dimensional ap­ pearance, which is often helpful. Brecher The basic lesion in the acanthocyte ap­ and Bessis9 also suggest adding an pears to be limited to its membrane since, echinocytogenic substance, such as to date, there has been no abnormality plasma with lysolecithin, in doubtful demonstrated in hemoglobin, cation cases. This will convert the discocytes to transport, red cell antibodies or gluta­ thione and ATP levels.26 The more com­ monly encountered forms of acan­ T A BLE I thocytosis exhibit rather marked altera­ Diseases Associated with Acanthocytosis tions of the red cell membrane lipids al­ though there is increasing evidence that Majority of red cells are acanthocytes proteins play a major role in membrane 1. Spur cell anemia with severe disease 2. Abetalipoproteinemia properties. 3. Homozygous hypobetalipoproteinemia To date, methods for isolation and 4. McLeod phenotype characterization of membrane proteins Minority of red cells are acanthocytes 1. Certain neurological disorders without lipid are less well developed than those for abnormalities lipids, so our information regarding pro­ 2. Infantile pyknocytosis - vitamin E deficiency 3. Miscellaneous conditions teins is less complete.45 Hence, while most of the following discussion will ACANTHOCYTOSIS 241 dwell upon lipid alterations, it should be the various theories of membrane struc­ borne in mind that proteins may also be ture and lipid exchange, the reader is re­ changed either primarily or secondarily. ferred to several excellent recent reviews To illustrate this point, one instance of of the subject.11,35,45,54 acanthocytosis is cited, namely the McLeod phenotype, where a protein alt­ Spur Cell Anemia eration is the probable principle or pri­ mary lesion. Abnormalities in serum lipoproteins are The red cell membrane is composed of known to induce secondary changes in lipid and protein in approximately equal red cells by modifying the equilibrium of amounts by weight. Phospholipids and the passive exchange of lipids between free cholesterol constitute nearly 95 per­ plasma and erythrocytes.45 In liver disease cent of the total lipid and, on a molar basis, of both the obstructive or hepatocellular phospholipids and cholesterol are present type, the red cells may becom e laden w ith in nearly equal amounts.45 The molecular cholesterol and, to a lesser extent, phos­ arrangement of the membranes is un­ pholipid. Thus, the membranes exhibit a known although the various studies have marked increase in the cholesterol-phos- generally evolved into two basic con­ pholipid (C/P), ratio. When examined on cepts, namely; the membrane is vis­ air-dried blood smears, they are fre­ ualized as a sea of lipid containing islands quently “targeted,” reflecting the acquisi­ of protein47 or as a protein membranous tion of membrane surface area caused by skeleton containing lakes of lipid.35 Both the added cholesterol. concepts suggest that the zones of fluid Using osmotic fragility as an indirect lipid within the membrane form the envi­ measure of membrane surface, it has been ronment for membrane proteins. Much of found that these cells exhibit increased this lipid is in the form of a bilamelar leaf­ osmotic resistance as expected. To this ex­ let with the hydrophilic portions of the tent these changes are usually of little phospholipids facing the aqueous envi­ clinical significance. However, in ful­ ronment on either side, and the hy­ minating hepatocellular disease, particu­ drophobic portions situated within the larly that of advanced alcoholic cirrhosis, central core of the bilayer.11 this process may be exaggerated leading The membrane phospholipids consist to massive accumulation of cholesterol in predominatly of four classes which are excess of phospholipid leading to acan­ asymmetrically distributed. Phosphati- thocyte formation (spur cells). Under dylethanolamine and phosphatidylserine these circumstances, red cell survival is are preferentially on the inner lamella, markedly decreased and is associated and phosphatidylcholine (lecithin) and clinically with hemolytic anemia. The on the outer lamella. Free studies of Cooper et al16 have indicated cholesterol is envisioned as being packed that this abnormality is acquired by ma­ within the lipid phase of the membrane in ture erythrocytes during equilibration an intermediate position between parts of with serum lipoproteins which, because both the head groups and the hydrocarbon of hepatic dysfunction, are laden with tails of the phospholipids, serving to free cholesterol. “tighten-up” the lipid core. Cholesterol Cooper et al16 suggest that the produc­ and phospholipids are not synthesized by tion of spur cells (acanthocytes) occurs in the red cells, and the membranes and two phases. First is the selective acquisi­ serum lipoproteins are in general equilib­ tion of cholesterol which increases the rium so that the membrane lipids reflect membrane surface area and causes de­ their plasma counterparts. For details of velopment of a scalloped or regularly 2 4 2 BIEMER spiculated contour. This occurs in 12 to 24 through the vascular tree, including capil­ hours and can be demonstrated in vivo or laries having lumens considerably smal­ in vitro by incubation of normal cells in ler than the cell itself. This feature is par­ serum from patients with spur cell ticularly important during passage anemia. The second phase has been re­ through the spleen when red cells are ferred to as “splenic conditioning” and forced to traverse narrow interendothelial consists of nonspecific loss of membrane clefts or slits (approximately 0.5 to 1.0 m surface accompanied by transformation of wide). Normally, deformable human red the spicules into a bizarre and irregular cells are able to move through the slits, contour in six to eight days. This occurs in but any reduction in deformability owing vivo only in conjunction with a function­ to membrane changes, alteration in the ing spleen. cytoplasmic state, or reduction in the sur­ While the cholesterol laden regularly face area-to-volume ratio of the cells, spiculated or targeted red cells are osmot- makes this passage difficult and leads to ically resistant reflective of their in­ splenic sequestration and destruction.39 creased membrane area, acanthocytes Reduced deformability of spur cells from when formed are osmotically normal. This severe alcoholic cirrhosis cases has been is probably a result of “splenic condition­ demonstrated using viscometric and mic­ ing” with nonspecific membrane loss ropore filtration techniques.36 and/or perhaps much of the membrane Splenic sequestration, particularly as­ excess is contained in the peripheral sociated with the congestive splenomeg­ spikes of the acanthocytes, and is unavail­ aly of cirrhosis, has been shown to be the able for cell expansion under osmotic dominant factor in their reduced survival. threat.45 Following splenectomy, the Similarly, Cooper’s patients16 who cholesterol laden cells of such cases main­ showed a 55 percent increase in red cell tain their orderly spiculation, are osmoti­ membrane free cholesterol with essen­ cally resistant, and do not develop the tially no change in phospholipid have bizarre morphology of acanthocytes.16 shown decreased deformability of the red Several mechanisms have been cells before and after splenectomy, as in­ suggested for the massive accumulation of dicated by increased filtration time. It is membrane cholesterol associated with suggested that the increased microviscos­ severe liver disease. These include de­ ity of the spur cell membrane may explain creased lecithin-cholesterol acyltrans- why the passage of these cells is retarded ferase (LCAT) activity owing to increased both through filters of small pore size in surface active bile acids15 and the pres­ vitro and through the spleen in vivo and it, ence of an abnormal lipoprotein in therefore, appears to represent the mini­ obstructive liver disease.49 Interestingly, mal cell defect responsible for hemolysis in the rare familial disorder, LCAT defi­ in this disorder. Since the membrane mic­ ciency, there is an equally great free roviscosity is directly related to the cholesterol accumulation by the red cell disproportionate excess of membrane membranes, but these red cells remain cholesterol, the magnitude of the choles­ targeted and do not develop spicules. Ob­ terol balance would be anticipated to cor­ viously, cholesterol accumulation alone relate, in turn, with the degree of anemia. cannot fully explain acanthocyte forma­ This correlation was demonstrated in tion in severe liver disease. Cooper’s patients with spur cells.16 Deformability is a specific functional With increasing evidence that the characteristic of red cell membranes that membrane proteins play a dominant role allows the red cell to undergo extreme in the regulation of membrane properties changes in shape during circulation while lipids contribute only minimally, ACANTHOCYTOSIS 243

Mohandas et al39 have found it difficult to from 50 to 90 percent. The exact mecha­ rationalize the correlation between de­ nism of formation of acanthocytes is un­ creased lipid fluidity and reduced cellular known although the demonstrated deformability. On the other hand, Lux35 biochemical abnormalities have been lim­ cited evidence that the protein membrane ited to the cell membranes. The studies of skeleton might not be involved in the Ways et al in 1963 demonstrated red cell acanthocytes of ABL and spur cell anemia. and plasma lipid abnormalities in three He has noted that membrane skeletons cases of ABL.52 These showed essentially always retain the shape of the ghosts from normal values for total lipid, cholesterol which they were prepared. and phospholipid in the red cell mem­ F or example, the ghosts and skeletons of branes, but an increase in sphingomyelin sickled cells, hereditary and and a decrease in lecithin, (reversed hereditary pyropoikilocytes retain the ab­ sphingomyelin/lecithin [S/L] ratio). Total normal cellular shape of the parent cell. plasma phospholipids were less than 20 Thus, the membrane skeleton seems to be percent of normal values with a compara­ the major determinant of shape in normal ble reversed S/L ratio owing to a propor­ and some pathologic red cells. However, tional increase in sphingomyelin and a when ghosts and skeletons of acantho­ corresponding decrease in lecithin. cytes of ABL and spur cell anemia were Ways et al52 also noted a profound defi­ prepared by the Triton extraction proce­ ciency in linoleic acid in the acanthocytic dure, they reverted to the usual round membrane and plasma lipids. Linoleic shape. Thus it would appear that the ab­ acid normally accounts for 10 to 14 per­ normality in the lipid bilayer is the more cent of the esterified fatty acids in the red important or only change. cell membrane, whereas only 2 to 3 per­ cent were found in acanthocytes. This re­ duction is most notable in that linoleic Abetalipoprote inemia acid is esterified in lecithin. Acanthocytes were first demonstrated Attempting to find a uniform mecha­ in 1950, in association with ABL, and re­ nism, McBride and Jacob36 studied acan­ main one of the characteristic hallmarks of thocytes from human ABL, experimental that rare inherited disease.3 As of 1978, ABL of rats fed orotic acid, rabbits on high over 40 cases of ABL had been reported cholesterol diets and human spur cell and summarized,26 with the emergence of anemia. Elevated cholesterol and C/P the classical clinical presentation of ratios were found in all types of acantho­ malabsorption of fat, retinitis pigmentosa, cytes. While cholesterol normally flows severe neuropathy and acanthocytic red bidirectionally between red cell mem­ blood cells. The terms “abetalipopro- branes and plasma lipoproteins, plasma teinemia” and “ApoB-deficiency”30 re­ from these instances of acanthocytosis flect the unique feature of the disease, were uniformly inefficient in accepting namely, the absence of and apparent ina­ cholesterol flux from the affected erythro­ bility to synthesize Apolipoprotein B (Apo cytes. This inefficiency could be cor­ B). Consequently, these patients with no rected by addition of concentrated human trace of betalipoprotein in their plasma betalipoprotein to the ABL plasma. are unable to elaborate chylomicrons and McBride and Jacob36 also demonstrated very low density lipoproteins (VLDL) and increased rigidity and resistance to flow of exhibit marked hypocholesterolemia and the acanthocytes as manifested by in­ hypotriglyceridemia. creased viscosity in cone-plate vis- Acanthocytes constitute the majority of cosimeters and diminished filterability the red cells in patients with ABL, ranging through micropore filters. While acknow­ 244 BIEMER ledging differences such as the reversed that the autohemolysis test can be cor­ S/L ratio in ABL acanthocytes, it was con­ rected in vitro by adding vitamin E to cluded that a uniform mechanism could acanthocytes or in vivo by parenteral in­ be applied to the formation of acantho­ jections to ABL patients.29 cytes, namely, accumulation of, and even One of the earliest manifestations of more importantly, stagnation of choles­ Vitamin E deficiency is the production of terol on the red cell membranes. hemolysis by exposure of red cells to When studying ABL acanthocytes, small amounts of hydrogen peroxide (i.e., Cooper et al12,13 reported that membrane hydrogen peroxide hemolysis test). This cholesterol was within the high limits of test is strikingly positive in ABL patients, normal and phospolipids were within the but it can be corrected by addition of vit­ low limits of normal resulting in a slight amin E.20 Acanthocyte microviscosity is increase in C/P ratio. These values were not related to vitamin E levels,12 nor is more comparable to those of most previ­ acanthocyte morphology altered by its ous investigators.52 A marked increase in oral administration to ABL patients.5 sphingomyelin with a decrease in lecithin However, it has been impossible to de­ was also demonstrated, as well as poor de- termine which features of ABL are due to formability of the acanthocytes to the in­ genetic alterations and how many may be creased S/L ratio. Similar results were due to deficiency of absorption of essen­ noted in the microviscosity of liposomes tial substances in the diet such as vitamin of varying S/L mole ratios prepared from E. Muscular weakness and dystrophy, brain sphingomyelin and egg lecithin central nervous system lesions and with equimolar cholesterol. Cooper et ophthalmic changes, including retinal de­ al12,13 concluded that the increased S/L struction and increased hemolysis of red ratio was responsible for the decreased cells, are all features of experimental vit­ membrane fluidity of the ABL acantho­ amin E deficiency in animals. If, as cytes, whereas it was the increased suggested, the major role of vitamin E is as cholesterol and C/P ratio that was respon­ an antioxidant, it may be that the central sible for decreased membrane fluidity in nervous system and opthalmic lesions of spur cells associated with liver disease. ABL are also related to vitamin E defi­ ABL acanthocytes have an increased ciency.29 Reports of the benefit of long rate of autohemolysis which can be cor­ term vitam in E therapy in ABL hold prom ­ rected by addition of normal serum. Ways ise that it may delay the development or and Simon53 have shown that the protec­ progression of the neurological and reti­ tive factor(s) in serum are contained in low nal lesions.40 density lipoprotein (LDL) and high den­ Red cell survival in ABL is usually shor­ sity lipoprotein (HDL). Vitamin E, in its tened and patients have frequently shown role as a biological antioxidant, appears to mild anemia, decreased haptoglobin, re- be related to the tendency toward au­ ticulocytosis and bone marrow erythroid tohemolysis of acanthocytes. Kayden and hyperplasia. Acanthocytes have de­ Silber18 have shown that the major lipo­ creased membrane fluidity and increased protein for plasma transport of vitamin E is microviscosity owing to an increase in the the LDL 1.006 to 1.063 fraction, (be- S/L ratio. It seems likely that these talipoprotein). Lacking betalipoprotein, changes account for the premature de­ vitamin E levels in ABL patients are the struction of the ABL acanthocytes, par­ lowest recorded in all human diseases. ticularly older ones, by the spleen. While Those small amounts of detectable vita­ the hemolytic state is never so marked as min E in these patients are predominantly that associated with spur cell anemia, this carried by HDL. It has also been shown may be accounted for by the absence of ACANTHOCYTOSIS 245 congestive which is so tion states and has occurred both with ,characteristic of severe liver disease with and without acanthocytosis. Gracey et portal hypertension. Otherwise, it seems aj 24,25 cjte njne instances of malnutrition likely that the spleen behaves as has been and intestinal malabsorption in Australian previously demonstrated in spur cell aboriginal infants with prominent acan­ anemia, that is, conditioning the acantho­ thocytosis and HBL. The acanthocytosis cytes to transform them from a scalloped proved to be transient, disappearing when contour to a thorny shape and then finally general nutrition and serum betalipo- destroying them.12 proteins returned to normal. The possible role of vitamin E deficiency in the infants Hypobetalipoproteinemia was not investigated but warrants Familial hypobetalipoproteinemia consideration. (HBL) is an inherited disorder of lipid me­ tabolism which has been reported in ap­ Acanthocytosis with Neurological proximately 12 families.26 The heterozy­ Disease in the Absence of Plasma gous state seems to be characterized by re­ Lipid Abnormalities duction in the rate of synethesis of struc­ turally normal LDL33 and clinically is Acanthocytosis has been reported in a usually asymptomatic although some pa­ number of families in association with tients have had neurological problems.37 neurological disease but without discer­ Acanthocytes are either absent or in­ nible serum lipid abnormalities.2,8,18,19,21, frequently seen.7 Mars et al37 demon­ 32,34 The neurological abnormalities have strated “acanthocytes” in apparent been rather varied, suggesting similarities heterozygotes who had serum cholesterol to Huntington’s chorea, Charcot Marie- of less than 100 mg per dl after incubation Tooth Disease and the Gilles de la in tissue culture media and autologous Tourette syndrome rather than the long serum, but they were not noted spontane­ tract signs with ataxia usually associated ously on peripheral blood smears. In addi­ with ABL. Critchley et al19 described the tion, they reverted to normal shape upon distinctive clinical features as an extra- addition of hyperlipemic or hypercholes- pyramidal movement disorder with dys- terolemic serum, suggesting to some that tonic, choreiform and athetoid move­ they may not have been true acanthocytes. m ents, and a w ide variety o f oro-facial tics. Lipid analysis of red cell ghosts has shown The mode of inheritance has appeared to normal total cholesterol, phospholipid be both autosomal dominant and reces­ and S/L ratio.37 sive. The acanthocytes have had typical The homozygous state of HBL is clini­ morphology, but are quantitatively less cally and biochemically indistinguishable than those seen in ABL, having been re­ from classical ABL.6,7,17,26,42 Acanthocytes ported as up to 50 percent of the red cell are fully as numerous in homozygous population, but more often less than 20 HBL as in classical ABL. The red cell percent. All of the cases have had normal lipids from Salt’s42 patients were studied serum lipids. Studies ofthe red cell lipids by Shacklady et al44 who reported results have revealed normal cholesterol and that were indistinguishable from those of phospholipid with no change in the S/L classical ABL, i.e., increased cholesterol ratio.2,8,21 and sphingomyelin content and reduced Linoleic acid content was described as linoleic acid and increased oleic acid es- normal by Estes et al21 and decreased by terified to phosphatidylcholine. 25 percent by Bird et al.8 A number of Secondary or acquired HBL has been instances were reported where normal reported in a wide variety of malabsorp­ red cells were converted to acanthocytes 246 BIEMER when incubated in patient’s compatible designated McLeod. The report of Wimer serum, suggesting that some serum factor et al55 in 1977 drew attention to the fact may be responsible for the acantho­ that the red cells in the McLeod cytosis.2'21 phenotype showed prominent acan­ thocytosis and were associated with a Infantile Pyknocytosis—Vitamin E compensated hemolytic process. In study­ D eficiency ing the propositus’s peripheral blood over a period of several months, the proportion In 1959, Tuffy et al80 first described a of acanthocytes varied from eight to 85 brisk hemolytic anemia in infants in as­ percent with approximately 25 percent sociation with up to 50 percent distorted found with immediate fixation as used for and contracted erythrocytes which were scanning electron microscopy. Serum referrred to as pyknocytes. Comparison triglycerides, lipoproteins and cholesterol was drawn to the acanthocytes in ABL were performed ruling out the possibility from which their “pyknocytes” could not of ABL. There was evidence of a compen­ be differentiated on a morphologic basis. sated hemolytic state with moderate re- The etiology of the anemia was undeter­ ticulocytosis, decreased serum haptoglo­ mined, but it was transient and responded bin and slight splenomegaly with normal to transfusion therapy. Similarly in 1967, osmotic fragility and autohemolysis. The Oski and Barness41 reported a hemolytic propositus’ mother and two daughters anemia in 11 low birth weight infants at were mosiacs when tested in the Kell six to 11 weeks of age, which they attri­ blood group, and they also had up to 10 buted to vitamin E deficiency. These in­ percent acanthocytes. Their reticulocytes fants had “pyknocytes” that were similar were also slightly elevated. The work of to those of Tuffy et al,50 abnormal hy­ Marsh38 has demonstrated that the drogen peroxide hemolysis tests and low McLeod phenotype is inherited as an vitamin E serum levels. The hemolytic X-linked characteristic and their red cells state and acanthocytosis were corrected lack Kx, a precursor-like substance, that by administration of vitamin E. It was appears to be necessary for the proper suggested that inadequate serum vitamin biosynthesis of Kell antigens. Marsh also E, a known potent antioxidant, was as­ reported that normal phagocytic leuko­ sociated with peroxidation of red cell cytes have Kx antigen whereas the leuko­ lipids which bound free sulfhydryl groups cytes of male patients with chronic leading to decreased red cell life span. granulomatous disease (CGD) do not. This mechanism has been supported by Because of the Lyon phenomenon of X the comprehensive studies of vitamin E chromosome inactivation, female carriers deficient rats by Jacob and Lux.28 The of the McLeod phenotype and CGD have cases of acanthocytosis secondary to mosiac populations of Kx positive and malabsorption and hypobetalipopro- negative red cells, or Kx positive and teinemia reported by Gracey etal24,25 bear negative leukocytes, respectively. It is great similarity to those of infantile pyk­ proposed that, the common allele called nocytosis, suggesting that these infants X Jk orders synthesis of Kx by phagocytic may also have had vitamin E deficiency. leukocytes and red cells in healthy people, whereas inheritance of a variant McLeod Phenotype of the Kell Blood Xk allele leads to absence of Kx on leuko­ Group System cytes and C G D , absence of Kx on red cells and the McLeod syndrome or absence of In 1961, Allen et al1 reported a new Kell Kx on both leukocytes and red cells in blood group system phenotype which was patients with both diseases.38 ACANTHOCYTOSIS 247

The mosiacism of females proves that Miscellaneous Conditions the Kx is a property of the red cells them­ selves and is not acquired by passive ab­ Acanthocytes have also been reported sorption from plasma. This is further sup­ in a number of diverse clinical situations. ported by the inability to convert McLeod Brecher et al10 have noted two to 10 per­ acanthocytes to normal morphology by in­ cent acanthocytes in post-splenectomy cubation in normal plasma or convert patients. It is suggested that these cells normal cells to acanthocytes by incuba­ are probably continually made and tion in McLeod plasma.22 Galey et al22 in promptly removed by the spleen. Noting studying McLeod acanthocytes found that their numbers gradually increase for them to have normal membrane lipids, several weeks following splenectomy, it microviscosity and electrolyte transport, appears that they then enjoy a longer sur­ but osmotic water permeability was 30 vival until a stable plateau is reached. percent below normal. Two to five percent acanthocytes were Little is known of the biochemical na­ found in hypothyroidism, reticulum cell ture of the antigenic determinant of the sarcoma, psoriatic skin lesions and in one Kell system. Indirect evidence, however, infant with unexplained possibly congen­ suggests that the Kell antigen system may ital hemolytic anemia. Horton et al27 re­ be attributable to a polysaccharide, and it ported rare acanthocytes in 19 percent of has been suggested that the Kx antigen their patients with hypothyroidism. Their may be a marker on a structural protein to incidence could not be correlated with the which the Kell specific sugars are at­ level of serum cholesterol or thyroxine, tached.38 Galey et al also cite indirect evi­ but they felt that they tended to be as­ dence related to the demonstrated de­ sociated with the more advanced cases of creased osmotic water permeability and hypothy ro idi s m. its known association with altered sulf- For the most part, the acanthocytes dis­ hydryl bonds to suggest that the McLeod appeared following response to therapy. cell abnormality is due to a protein altera­ Wardrop and Hutchinson,51 while not tion. Obviously, confirmation must await specifically referring to them as acantho­ a detailed study of its membrane proteins. cytes, reported the association of similar In any event, McLeod acanthocytes, cells in hypothyroidism. Other cited which are morphologically indistinguish­ conditions in which acanthocytes have able from all other acanthocytes, are been found include , somewhat unique in that while most panhypopituitarism, carcinoid tumor others have some demonstrable lipid ab­ metastatic to the liver with hemolysis and normality, these, instead, have normal hypoplastic anemia.14 They have also lipids and a high probability of a primary been found in dogs in association with membrane protein alteration. Such an as­ splenic neoplasms consisting of heman­ sociation of an altered membrane struc­ giomas or hemangiosarcomas.23 tural protein with acanthocytosis would The common feature shared by all of the be comparable to that stomato-sphero- these conditions is that those acantho­ cytosis and hemolytic anemia shown to be cytes which were noted have formed a characteristic of the Rh null blood type.48 minor proportion of the total red cell As recently suggested by Schmidt,43 the population. In those instances where possibility of finding additional null membrane lipids have been measured, blood types in a wide variety of hemolytic they have been normal. Furthermore, it is states with altered red cell morphology, not entirely clear from the published re­ may be a fertile field for investigation in ports whether all of the cells have been the future. true acanthocytes. Clearly, the possibility 248 BIEMER of echinocytes, schizocytes and non­ 17. COTTRILL, C ., G l UECK, C. J., LEUfiA, V., ET AL: Familial homozygous hypobetajipoprotein- specific poikilocytes may not have been emia. M etabolism 23:779-791, 1974. completely excluded. 18. C r it c h l e y , E . M. R., B e t t s , J. J., N ic h o l s o n , J. T., ET AL: Acanthocytosis, normolipopro- teinaemia and multiple tics. Postgrad. Med. J. References 46:698-701, 1970. 19. C r it c h l e y , E . M. R., C l a r k , D. B., and 1. A l l e n , F. H., K r a b b e , S. M. R., and C o r c o ­ WlKLER, A.: Acanthocytosis and heurological r a n , P. A .: A new phenotype (McLeod) in the disorders without betalipoproteirtemia. Arch. Kell blood-group system. Vox. Sang. 6:555-560, Neurol. 18:134-140, 1968. 1961. 20. D o d g e , J. T., C o h e n , G., Ka y d e n , FI. J., e t a l : 2. Am in OFF, M. J.: Acanthocytosis and neurologi­ Peroxidative hemolysis of red blood cells from cal disease. Brain 95:949-960, 1972. patients with abetalipoproteinemia (acan­ 3. BASSEN, F. A. and Kornzweig, A. L.: M alfor­ thocytosis). J. Clin. Invest. 46:357-368, 1967. mation of the erythrocytes in a case of atypical 21. E s t e s , J. W., M o r l e y ., T. J., L e v in e , I. M ., e t retinitis pigmentosa. Blood 5:381-387, 1950. AL: A new hereditary acanthocytosis syndrome. 4. B e S S I S , M.: Living Blood Cells and Their Ul­ A m er. J. M e d . 42:868-881, 1967. trastructure. New York, Springer-Verlag, 1972. 22. G a l e y , W. R., E v a n s , A. P., Va n N i c e , P. S., e t 5. B i e m e r , J. J.: Unpublished observations, 1979. AL: Morphology and physiology of the McLeod 6. B ie m e r, J. J. and McCammon, R. E.: Abeta- erythrocyte. I. Scanning electron microscopy lipoproteinemia and hypobetalipoproteinemia and electrolyte and water transport properties. within the same family. Ann. Clin. Lab. Sci. Vox. Sang. 34:152-161, 1978. 3:308, 1973. 23. G e l b e r g , H. and St a c k h o u se , L- L.: Three 7. Bie m e r , J. J. and M c C a m m o n , R. E.: The ge­ cases of canine acanthocytosis associated with netic relationship of abetalipoproteinemia and splenic neoplasia. Vet. Med./Small Anim. Clin. hypobetalipoproteinemia: A report of the oc­ 72:1183-1184, 1977. currence of both diseases within the same fam­ 24. G r a c e y , M. and H il t o n , H. B.: Acanthocytes ily. J. Lab. Clin. Med. 85:556-565, 1975. and hypobetalipoproteinaemia. Lancet 1:679, 8. Bird, T. D., Cenderbaum, S., Valpey, R. W. 1973. ET AL: Familial degeneration of the basal gang­ 25. G r a c e y , M., W ils o n , R. G ., and P e t e r s e n , M.: lia with acanthocytosis: A clinical, neuropatho- Transient Acanthocytosis and hypobetalipo­ logical, and neurochemical study. Ann. Neurol. proteinaemia. Aust. New Zea. J. Med. 4:397- 3:253-258, 1978. 401, 1972. 9. B R E C H E R , G. and B e S S I S , M.: Present status of 26. H e r b e r t , P. N., G o t t o , A. M.; and F r e d ­ spiculed red cells and their relationship to the r ic k s o n , D. S.: Familial lipoprotein defi­ discocyte-echinocyte transformation: A critical ciency. The Metabolic Basis of Inherited Dis­ review. Blood 40:333-344, 1972. ease, 4th ed. Stanbury, J. B., Wyngaarden, J. B. 10. B r e c h e r , G., H a l e y , J. E., and W allerstein, and Fredrickson, D. S., eds. New York, R. O.: Spiculed erythrocytes after splenectomy McGraw-Hill, 1978, pp. 544-588. acanthocytes or non-specific poikilocytes? 27. H o r t o n , L., C o b u r n , R J., E n g l a n d , J. M., e t Nouv. Rev. Fr. Hematol. 12:751-754, 1972. AL: The haematology of hypothyroidism. Quart. 11. C o o p e r , R. A.: Abnormalities of the cell- J. Med. 45:101-123, 1976. membrane fluidity in the pathogenesis of dis­ 28. Ja c o b , H. S. and Lux, S. E.: Degradation of ease. New Eng. J. Med. 297:371-377, 1977. membrane phospholipids and' thiols in 12. C o o p e r , R. A., D u r o c h e r , J. R., and L e s l i e , peroxide hemolysis: Studies in vitamin E defi­ M. H.: Decreased fluidity of red cell membrane ciency. Blood 32:549-568, 1968. lipids in abetalipoproteinemia. J. Clin. Invest. 29. K a y d e n , H. J. and S i l b e r , R.: The role o f vita­ 60:115-121, 1977. m in E deficiency in the abnormal auto­ 13. Cooper, R. A., D urocher, J. R., and Leslie, hemolysis of acanthocytosis. TAns. Assoc. M. H.: Viscous membrane lipids in the Amer. Physicians. 78:334-342, 1965. pathogenesis of rigid red cells in abetalipo­ 30. Ko s t n e r , G. M.: ApoB-deficiericy (abetalipo- proteinemia. Clin. Res. 24:438A, 1976. proteinaemia): A model for studying the lipo­ 14. COOPER, R. A. and J a n d l , J. H.: Acanthocytosis. protein metabolism. Lipid Absorption: , 2nd ed. Williams, W. J., Beutler, Biochemical and Clinical Aspects. Rommel, K. E., Erslev, A. J., et al., eds. New York, and Bohmer, R. eds. Baltimore, University Park McGraw-Hill, 1977, pp. 461-465. Press, 1976, pp. 203-236. 15. C o o p e r , R. A. and J a n d l , J. H.: Bile salts and 31. LESSIN, L. S., Kl u g , P. P., and JENSEN, W. N.: cholesterol in the pathogenesis of target cells in Clinical implications of red cell shape. Adv. obstructive jaundice. J. Clin. Invest. 47:809- Intern. Med. 21:451-500, 1976. 822, 1968. 32. Le v in e , I. M., E s t e s , J. W., and L o o n e y , J. M.: 16. C o o p e r , R. A., Kim ball, D. B., and D u r o c h e r , Hereditary neurological disease with acan­ J. R.: Role of the spleen in membrane condition­ thocytosis. Arch. Neurol. 19:403-409, 1968. ing and hemolysis of spur cells in liver disease. 33. L e v y , R. I., L a n g e r , T., G o t t o , A. M., e t a l : New Eng. J. Med. 290:1279-1284, 1974. Familial hypobetalipoproteinemia, defect in ACANTHOCYTOSIS 249

lipoprotein synthesis. Clin. Res. 18:539, 1970. 44. Shacklady, M. M., D jardjouras, E. M., and 34. L o w e , D., I b b o t s o n , R. M., and Loizou, L. A.: L lo y d , J. K.: Red cell lipids in familial alpha- Benign normolipoproteinaemic acanthocytosis. lipoprotein deficiency (Tangier Disease). Lan­ Brit. Med. J. 2:99-100, 1977. cet 2:151-153, 1968. 35. Lux, S. E.: Spectrin-actin membrane skeleton 45. Shohet, S. B.: Hemolysis and changes in of normal and abnormal red blood cells. Semi­ erythrocyte membrane lipids. New Eng. J. nar Hema. 16:21-51, 1979. Med. 286:577-583, 638-644, 1972. 36. M cBride, J. A. and J a c o b , H. S.: Abnormal 46. Shohet, S. B. and N e s s , P. M.: Hemolytic kinetics of red cell membrane cholesterol in anemias. Failure of the red cell membrane. acanthocytes. Studies in genetic and experi­ Med. Clin. North Amer. 60:913-932, 1976. mental abetalipoproteinemia and in spur cell 47. Singer, S. J. and Nicolson, G. L.: The fluid anemia. Brit. J. Haematol. 18:383-397, 1970. mosiac model of the structure of cell mem­ 37. M a r s , H., L e w is , L. A., Robertson, A. L ., e t branes. Science 175:720-731, 1972. AL: Familial hypo-B-lipoproteinemia. A ge­ 48. STURGEON, P.: Hematological observations on netic disorder of lipid metabolism with nervous the anemia associated with blood type Rh null. system involvement. Amer. J. Med. 46:886- Blood 36:310-320, 1970. 900, 1969. 49. SwiTZER, S.: Plasma lipoproteins in liver dis­ 38. MARSH, W. L.: Chronic granulomatous disease, ease. I Immunologically distinct low-density Kx antigen and the Kell blood groups. Prog. lipoproteins with biliary obstruction. J. Clin. Clin. Biol. Res. 21:493-511, 1978. Invest. 46:1855-1866, 1967. 39. M ohandas, N., Phillips, W. M ., and B e s s is , 50. T u f f y , P., B r o w n , A. K., and Z u e l z e r , W. W.: M .: Red blood cell deformability and hemolytic Infantile Pyknocytosis. Amer. J. Dis. Child. anemias. Seminar Hemat. 16:95-114, 1979. 98:227-241, 1959. 40. M u l l e r , D. P. R., L lo y d , J. K., and B ir d , A. C.: 51. WARDROP, C. and Hutchison, H. E.: Red-cell Long-term management of abetalipopro- shape in hypothyroidism. Lancet 1 :1243, 1969. teinaemia. Possible role for vitamin E. Arch. 52. W a y s, P., R e e d , C. F., H a n a h a n , D. J., e t a l : Dis. Child. 52:209-214, 1977. Red-cell and plasma lipids in acanthocytosis. J. 41. O s k i, F. A. and BARNESS, L. A.: Vitamin E defi­ Clin. Invest. 42:1248-1260, 1963. ciency: A previously unrecognized cause of 53. W a y s, P. and S im on, E. R.: The role of serum in hemolytic anemia in the premature infant. J. acanthocyte autohemolysis and membrane Pediat. 70:211-220, 1967. lipid composition. J. Clin. Invest. 43:1322- 42. S a l t , H. B., W o l f f , O. H., L lo y d , J . K . , e t a l : 1328, 1964. On having no beta-Iipoprotein: A syndrome 54. WEED, R. I.: Membrane structure and its rela­ comprising a-beta-lipoproteinemia, acantho­ tion to haemolysis. Clin. Haematol. 4:3-28, cytosis, and steatorrhea. Lancet 2:325-329, 1975. 1960. 55. W im e r, B. M., M arsh, W. L., T a s w e l l , H. F., 43. S c h m id t, P. J.: Hereditary hemolytic anemias ET AL: Haematological changes associated with and the null blood types. Arch. Intern. Med. the McLeod phenotype of the Kell blood group 139:570-571, 1979. system. Brit. J. Haematol. 36:219-224, 1977.