JOURNAL TRANSCRIPT

---

Vet Pathol 31518-527 (1994)

Equine Glucose-6-phosphate Dehydrogenase Deficiency S. L. STOCKHAM, J. W. HARVEY, AND D. A. KINDEN Department of Veterinary Pathology, University of Missouri-Columbia, Columbia, MO (SLS, DAK); and Department of Physiological Sciences, University of Florida, Gainesville, FL (JWH)

Abstract. Glucose-6-phosphate dehydrogenase (G6PD) deficiency is a well-characterized X-linked inherited disorder in humans but has not been reported in horses. We describe a persistent hemolytic anemia and hyperbilirubinemia due to a severe G6PD deficiency in an American Saddlebred colt. Other abnormalities in the colt’s erythrocytes as compared with those of healthy horses ( n = 22-35) included increased activities of hexokinase and pyruvate kinase, decreased concentrations of reduced glutathione and reduced nicotinamide adenine dinucleotide phosphate (NADP), and increased concentration of oxidized NADP. Morphologic abnormalities included eccentrocytosis, pyknocytosis, anisocytosis, macrocytosis, and increased number of HowellJolly bodies. Scanning and transmission electron microscopic examinations revealed that eccentrocytes had contracted to spherical regions and thin collapsed regions. Eccentrocytes were more electron dense than were normal erythrocytes when examined by transmission electron microscopy. When exposed to acetylphenylhydrazine, erythrocytes from the G6PD-deficient colt produced more and smaller Heinz bodies than did erythrocytes from normal horses. Abnormalities in the colt’s dam included presence of eccentrocytes and pyknocytes; her average erythrocyte G6PD activity was slightly below the range of reference values. Key words: Abnormal erythrocytes; congenital hemolytic anemia; glucosephosphate dehydrogenase deficiency; horse diseases; reference values.

Glucose-6-phosphate dehydrogenase (G6PD) is the first and the rate-limiting enzyme of the hexose monophosphate shunt, which produces reduced nicotinamide adenine dinucleotide phosphate (NADPH). In erythrocytes, NADPH is used to keep glutathione in a reduced state and catalase in an active form. Both reduced glutathione (GSH) and catalase protect erythrocytes from oxidative insults. As determined by biochemical methods, there are over 400 variants of G6PD described in humans, and there are more known mutants of G6PD than of any other enzyme of h ~ m a n s . ~Sequencing .~J~ of DNA in recent years has shown that some of these described variants have the same mutation and thus are not really different.5z6Also, not all G6PD variants are characterized by deficient enzyme activity. A World Health Organization (WHO) classification groups the human variants into five classes: class 1 -nonspherocytic hemolytic anemia; class 2-severe enzyme deficiency (< 10% of normal); class 3 -moderate deficiency (1 060% of normal); class 4-very mild or no deficiency (60-100% of normal); class 5-increased enzyme activity (>twice Class 3 deficiencies of G6PD (using WHO criteria) have been recognized in a dog,50a mutant strain of m i ~ e , ’and ~ . highly ~ ~ inbred rats.53The G6PD-deficient dog was one of 3,300 dogs screened for the defect, and there were no clinical abnormalities associated with

the deficiency. The mutant mouse strain was produced from male mice treated with ethylnitrosourea. Increased splenic weight was reported in hemizygous males but other hematologic abnormalities were not found. The manifestations of G6PD deficiency in rats were not determined because the strain was lost to an infectious disease. We identified an American Saddlebred colt with a severe erythrocyte deficiency in G6PD activity (class 1). This is the first report of a G6PD deficiency that produced a persistent hemolytic anemia in an animal other than humans. A 6-month-old male American Saddlebred horse had an anemia of unknown origin. The dam’s colostrum was withheld from this colt at birth because the mare’s colt in the previous year had a lethal neonatal hemolytic disorder thought to be neonatal isoerythrolysis. This colt’s hematocrit was 0.32 at birth and decreased to 0.20 by the third day. The colt responded to a whole blood transfusion and was allowed to nurse until weaning. When the colt was examined at 6 months of age, it had three edematous legs, distal leg dermatitis involving all four legs, mild fever, and macrocytic anemia (Table 1: 12/16/91). The dermatitis was thought to be related to inadequate husbandry. Treatment included cleaning affected legs and oral trimethoprim/sulfamethoxazole. The colt was transported to a training

518

Equine G6PD Deficiency

Vet Pathol 315, 1994

519

Table 1. Selected hematologic and bilirubin data from a G6PD-deficient colt. Analyte*

Hct (literdliter) Hgb (glliter) RBC ( x 1O12/liter) MCV (fl) MCHC (g/dl) Bili (wmol/liter) Bc (pmol/liter) Bu (wmol/liter) 6 Bili (Kmol/liter)

Sample Datest I2/16/9 1

0.11 32.0 2.16 49 30.4 83.8 ND ND ND

12/23/9 1

1/2/92

0.16

0.24 82.0 5.38 44 34.7 71.8 0.0 65.0 6.8

57.0

3.21 49 36.2 ND ND ND ND

Reference Limits+

0.29-0.41 108.0-154.0 7.9-1 1.6 32-39 3340 1.7- 12.0 < 12.0 < 12.0 < 12.0

* Hct = hematocrit; Hgb = hemoglobin; RBC = red blood cell; MCV = mean corpuscular volume; MCHC = mean corpuscular hemoglobin concentration; Bili = total bilirubin; Bc = conjugated bilirubin; Bu = unconjugated bilirubin; 6 Bili = delta bilirubin. t Assays on 12116191 completed by Roche Biomedical Laboratories; other assays completed by the Clinical Pathology Laboratory, College of Veterinary Medicine, University of Missouri-Columbia. ND = not done. Hematologic values for 6-month-old horses.*’ Bilirubin reference values from Kodak.

+

stable with a recommendation to continue treatments but also to pursue the anemia’s cause. A veterinarian for the new stable submitted blood for analysis. Laboratory analyses showed the presence of a macrocytic normochromic anemia (Table 1: 12/23/9 1) with anisocytosis, eccentrocytosis, and pyknocytosis (Fig. 1); leukogram results and platelet concentration were within reference limits. Heinz bodies were not found when fresh blood was stained with new methylene blue (NMB). There was no known exposure to oxidants other than the oral sulfamethoxazole; treatment with the antibiotic compound was stopped. During the next 10 days, the dermatologic disorder responded well to mild daily cleaning. Abdominal ultrasonography demonstrated a mildly enlarged spleen. A screening test for G6PD activity indicated the colt’s blood was G6PD deficient, whereas blood samples from two healthy horses produced fluorescence indicating G6PD activity. Quantitative G6PD assays were used to confirm the diagnosis of G6PD deficiency causing hemolytic anemia with eccentrocytosis and pyknocytosis.

analyzed each time that samples from the G6PD-deficient colt and his dam were assayed. Heparinized blood was used for the acetylphenylhydrazine challenge test. Serum samples for serum biochemical assays were analyzed the day of collection. Complete blood cell counts An impedance cell counter (Coulter S-Plus 4, Coulter Electronics, Hialeah, FL; modified for optimal counting of domestic animal erythrocytes) was used for cell counting, measurement of hemoglobin concentration, and determination of erythrocyte indices; calculated hematocrit values were confirmed by centrifuged microhematocrit determinations. Manual leukocyte differential counts and microscopic examination of blood cells were completed on stained blood films (Hema-tek@ stain, Miles, Diagnostics Division, Elkhart, IN).

Materials and Methods Blood samples Samples for complete blood cell counts and routine Heinz body staining were collected with tripotassium ethylenediaminetetraacetate (EDTA) anticoagulant and analyzed the same day. Samples for quantitative erythrocyte enzyme assays and nicotinamide adenine dinucleotide phosphate (NADP) assays were collected with EDTA, kept near 4 C, and shipped the same day via overnight carrier to the University of Florida for analysis. Blood extracts for adenosine triphosphate (ATP), 2,3-diphosphoglycerate (2,3-DPG), and GSH assays were prepared from EDTA blood rapidly after sample collection, immediately frozen (- 70 C), and shipped with dry ice via overnight carrier. Blood samples from norma1 control horses were collected, processed, shipped, and

Fig. 1. Blood; G6PD-deficient horse. Erythrocyte abnormalities include presence of eccentrocytes (e), pyknocytes (p), macrocytes (m), and anisocytosis. Pyknocytes have decreased cell diameters and increased staining intensity as compared with normal horse erythrocytes. Hema-tek@stain. Bar = 2.8 wm.

Stockham, Harvey, and Kinden

520

Heinz bodies and reticulocytes Near equal volumes of fresh EDTA blood and NMB stain (new methylene blue “N”, Ricca Chemical Co., Arlington, TX) were mixed. After 10 minutes, smears were made, allowed to air dry, and examined microscopically for Heinz bodies and reticulocytes.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) Erythrocytes in EDTA blood were fixed with 2% paraformaldehyde-2% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.3) for 2 hours at 4 C. After primary fixation, specimens were washed thoroughly with buffer and postfixed with aqueous 1% osmium tetroxide for 2 hours at 4 C. For SEM samples, erythrocytes were placed on poly-L-lysine-coated glass cover slips, dehydrated in a graded ethanol series, and critical point dried. Specimens were then mounted on aluminum stubs with copper foil tape and sputter coated with gold. For TEM samples, erythrocytes were suspended in 2% water agar, which was cut into 1 mm cubes. Subsequently, specimens were stained en bloc with aqueous 1% uranyl acetate for 1 hour, dehydrated in a graded ethanol series, and embedded in a mixture of Epon-Araldite epoxy resin. Ultrathin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate.

Serum chemical assays Serum concentrations of glucose, urea, creatinine, sodium, potassium, chloride, total protein, albumin, calcium, inorganic phosphorus, bilirubin (total, conjugated, unconjugated, and delta), and total carbon dioxide were measured with commercial assays in an automated chemistry unit (Kodaka Ektachem 500, Eastman Kodak, Rochester, NY). Serum activities of aspartate transaminase, alkaline phosphatase, y-glutamyl transferase, and creatine kinase were determined with kinetic assays from the same manufacturer. @

G6PD assays A fluorescent spot test (G6PD Deficiency Screen, Sigma Chemical, St. Louis, MO) was initially used to pursue the G6PD deficiency in the colt. Control samples included a commercial human control solution (G6PD Control Normal; Sigma Chemical) and blood from two healthy adult horses. Quantitative G6PD activity was measured spectrophotometrically (DU-4 Spectrophotometer, Beckman Instruments, Fullerton, CA) at 37 C in hemolysates prepared following the removal of leukocytes by filtration through microcrystalline cellulose and alpha-cellulose columns.*The G6PD activity was measured by both the WHO and the Glock and McLean methods.2 In the WHO method, some of the measured G6PD activity is due to 6-phosphogluconate dehydrogenase (6PGD) activity. With the Glock and McLean method, the apparent G6PD activity is corrected by subtracting the contribution of 6PGD activity.

Vet Pathol 3 1 5 , 1994

also measured at 37 C in hemolysates prepared following the removal of leukocytes.2 Erythrocyte 2,3-DPG and ATP concentrations were measured spectrophotometrically using trichloroacetic acid-precipitated extracts (Kits 35-UV and 366UV, Sigma Chemical). Total NADP,55NADPH,55and GSH44 were extracted, and concentrations were measured spectrophotometrically using established methods. Methemoglobin concentrations were determined in distilled water hemolysates. 30

Acetylphenylhydrazine challenge (Heinz body) test The sensitivity of erythrocytes to form Heinz bodies after exposure to an oxidant was assessed via an in vitro acetylphenylhydrazine challenge test.’ Heparinized blood samples from the G6PD-deficient colt, his dam, and two nonrelated horses were evaluated. The number of Heinz bodies in each of 200 erythrocytes was determined by microscopic examination of erythrocytes prior to and after exposure to acetylphenylhydrazine. Erythrocytes were stained with methyl violet. Any cell that contained Heinz bodies that were too small and numerous to be counted reliably was classified as a stippled cell.

Results Complete blood cell counts

The G6PD-deficient colt’s anemia was most severe (hematocrit = 0.11 literdliter) when the colt had a distal leg dermatitis. As the dermatologic condition improved, the colt’s hematocrit values increased and then fluctuated between 0.19 and 0.26 literslliter (X = 0.22 literdliter) during the next 15 months while the colt was clinically healthy. Eccentrocytosis, pyknocytosis, anisocytosis, macrocytosis, and increased number of Howell-Jolly bodies persisted throughout the evaluation. Heinz bodies and reticulocytes were not found in NMB-stained smears. The increased staining density of eccentrocytes was similar to the density of pyknocytes in both Romanowsky- and NMB-stained blood films. Rare erythrocytes contained rectangular hemoglobin crystals. Abnormalities were not found in the colt’s leukograms or platelet concentrations. Complete blood counts for the colt’s dam were consistently within normal limits, except for rare eccentrocytes and pyknocytes. Electron microscopy

When examined by SEM, eccentrocytes had two major features: a contracted to spheroid region with either a rugose or smooth surface and a thin collapsed region frequently with or without projections or fragments (Figs. 2-4). Pyknocytes were identified as contracted spherocytic cells with projections (Fig. 5) in contrast to a normal equine discocyte (Fig. 6). Echinocytic Other erythrocyte assays change was present in many cells and was considered Erythrocyte hexokinase (HK), phosphofructokinase (PFK), to be artifact. Via TEM, cross sections of eccentrocytes pyruvate kinase (PK), catalase, and 6PGD activities were showed projections that corresponded with the thin,

Vet Pathol 315, 1994

Equine G6PD Deficiency

collapsed regions seen via SEM (Fig. 7). Compared with normal erythrocytes, eccentrocytes were uniformly more electron dense, and they consistently had a slightly darker concentric line about 0.1 pm in width and 0.2 pm from the cell margin. No other abnormalities were detected. Serum chemical assays

521

Acetylphenylhydrazine challenge (Heinz body) test

After exposure to acetylphenylhydrazine, nearly all erythrocytes from the G6PD-deficient colt contained more than 20 tiny Heinz bodies, and many were classified as stippled cells (Figs. 8, 10). In the normal horses, erythrocytes most commonly had < 10 distinct Heinz bodies, and stippled cells were very uncommon (Figs. 9, 10). Results for the colt’s dam were very similar to those of normal horses, except about 7% of her erythrocytes were stippled cells. Heinz bodies were not found in erythrocytes from the G6PD-deficient colt, his dam, or the two normal horses before exposure to acetylphenylhydrazine.

The colt had a persistent hyperbilirubinemia throughout the evaluation; most of the serum bilirubin was unconjugated (Table 1). During the first 20 months of evaluation, the bilirubin concentrations ranged as follows: total = 49.6-1 26.5 pmol/liter; conjugated = 0.0-10.3 pmol/liter, unconjugated = 49.6-126.5 pmoU Discussion liter; 6 = 0.0-13.7 pmol/liter. Serum alkaline phosGlucose-6-phosphate dehydrogenase catalyzes the phatase activity was about three to four times the upper limit value found in healthy adult horses, but values first reaction in the hexose monophosphate (or penwere considered normal for a young growing horse. tose) shunt, which produces NADPH, the major source Serum concentrations and activities of the other com- of reducing equivalents used to protect erythrocytes mon measured analytes were all within normal limits from oxidative injury. Low levels of hydrogen peroxide (H202)are produced by normal cellular events, and throughout the evaluation. higher levels may be generated by exogenously administered oxidative compounds. 13,38,45 Hydrogen perG6PD activity oxide is catabolized by either catalase or the glutaBlood of the G6PD-deficient colt did not fluoresce thione reductase/peroxidase pathway, and NADPH is in the fluorescent spot test, whereas blood of two healthy believed to be essential in maintaining the activity of horses and a normal human control solution did. As both of these NADPH is utilized to reduce determined by the WHO method, the colt’s average oxidized glutathione to GSH, the substrate for the gluG6PD activity was 0.3 IU/g Hgb ( n = 5), whereas tathione peroxidase reaction, and it is bound to catavalues in normal horses ranged from 15.2 to 27.3 IU/g lase, preventing and reversing the accumulation of Hgb ( n = 35). When the contribution of 6PGD activity compound 11, an inactive form of catalase that is genwas subtracted (Glock and McLean assay), however, erated when catalase is exposed to H,02.35 The inthe colt’s G6PD activity was consistently below de- creased sensitivity of G6PD-deficient erythrocytes to tection limits (