Haemolytic disease of fetus and newborn caused by ABO antibodies in a cisAB offspring

Available online at www.sciencedirect.com

Transfusion and Apheresis Science 39 (2008) 123–128 www.elsevier.com/locate/transci

Haemolytic disease of fetus and newborn caused by ABO antibodies in a cisAB offspring Zhi-Hui Deng a,*, Axel Seltsam b, You-Wan Ye c, Qiong Yu a, Qian Li a, Yu-Qing Su a, Yan-Lian Liang a, Hao Zang a a

ShenZhen Institute of Transfusion Medicine, ShenZhen Blood Center, ShenZhen, GuangDong, PR China b Institute for Transfusion Medicine, Hannover Medical School, Hannover, Germany c Department of Transfusion Medicine, ShenZhen Longgang Center Hospital, ShenZhen, GuangDong, PR China

Abstract ABO hemolytic disease of fetus and newborn (ABO-HDFN) occurs almost exclusively in infants of blood group A or B who are born to group O mothers because IgG anti-A or -B occurs more commonly in group O than in group A or B individuals. We report a case of clinically significant ABO-HDFN where the mother was blood group O with elevated IgG anti-A and anti-B titers and delivered a child with an A2B phenotype. This unusual ABO constellation between mother and infant was based on the inheritance of a rare ABO allele encoding for a glycosyltransferase capable of synthesizing both A and B antigens. Because both anti-A and anti-B antibodies may have been involved in hemolysis in this case, it may be relevant to consider the cisAB phenomenon when monitoring ABO-incompatible pregnancies and births. Ó 2008 Published by Elsevier Ltd. Keywords: Hemolytic disease of fetus and newborn; ABO blood group; cis-AB allele

1. Introduction Hemolytic disease of fetus and newborn (HDFN) is a clinical condition in which neonatal red cells are destroyed by maternal IgG alloantibodies against infants red cells antigens inherited from paternal. ABO-HDFN is one of the most important causes of neonatal hyperbilirubinemia. Depending on the ethnic group, laboratory methods and diagnostic definitions used, incidence and severity degree of *

Corresponding author. Tel.: +86 755 8313 7826; fax: +86 755 8322 1000. E-mail address: [email protected] (Z.-H. Deng). 1473-0502/$ - see front matter Ó 2008 Published by Elsevier Ltd. doi:10.1016/j.transci.2008.07.007

ABO-HDFN were disparate. Published literatures report that 0.33–2.2% of all neonates have some manifestation of ABO incompatibility while 15% of live births are at risk as type A/B offspring of mothers with type O blood [1,2]. In the USA, 6.9% of all births are of infants with maternal-fetal ABO incompatibility [3]. In Chinese, the incidence of HDFN due to ABO incompatibility antibodies ranges from 0.7% to 4% [4]. The severity of ABO-HDFN varies from no to severe clinical symptoms. In general, hemolysis caused by ABO incompatibility is minimal and the clinical course is relatively benign, characterized by high level serum bilirubin, moderately severe anemia, the appearance of jaundice, reticulocytosis and sphe-

124

Z.-H. Deng et al. / Transfusion and Apheresis Science 39 (2008) 123–128

rocytosis. Hydrops caused by ABO-HDFN is extremely rare, and very occasionally exchange transfusion for the prevention of kernicterus is indicated. The mild forms of ABO-HDFN do normally need no therapy, while cases with moderate severity are often treated with phototherapy. Exchange transfusion with or without phototherapy and intravenous gamma globulin (IVGG) therapy can become necessary in severely compromised or asphyxiated infants. Here we report an unusual case of significant ABO-HDFN in a neonate with a cisAB phenotype due to both immunoglobulin G (IgG) anti-A and anti-B antibodies presenting with severe anaemia. 2. Materials and methods 2.1. Patient report A 34-year-old Chinese woman delivered a fullterm male baby by natural spontaneous vaginal delivery. The woman had four pregnancies with two abortions in her history partnered by the same man. Her first child had been born without anemia. The woman typed as blood group O Rh(D) positive, and had no blood transfusions in her history. The baby was referred to our hospital 40 min after delivery, because of a newborn inhalant syndrome and jaundice with a total bilirubin level of 13.2 mg/dl (Lab’s normal reference: <6 mg/dl, 24 h after delivery) and reticulocyte count of 2.8%. For unknown reasons, the baby left on day 3 and returned with a total bilirubin of 24.1 mg/dl (Lab’s normal reference: <12 mg/dl, between 3 days and 7 days after delivery) to the hospital on the same day. On day 4, the total bilirubin peaked at 37.5 mg/dl and indirect bilirubin was 35.6 mg/dl. The infant’s blood hemoglobin was 8.2 g/dl (Lab’s normal reference ranges from 17 to 20 g/dl). Laboratory examination revealed serologic evidence for ABO incompatibility between mother and child. Her first child’s blood group was O Rh(D) positive. Her husband’s blood group was AB Rh(D) positive. The baby was typed blood group AB; anti-A and anti-B antibodies were detectable in the infant’s serum and could be eluted from his red blood cells (RBCs). The direct antiglobulin test (DAT) performed on the infant’s RBCs was negative. The infant showed neither hepatosplenomegaly nor neurologic abnormalities. His erythrocyte glucose6-phosphate dehydrogenase (G6PD) activity was normal and sickle cell test yielded negative results. He was ruled out thalassaemia and hemoglobinopathy performed by Kleinhaner-Batke. After photo-

therapy, high-dose intravenous gamma globulin (IVGG) therapy and treatment with recombinant human erythropoietin (rhEPO) the infant was discharged with a total bilirubin of 6.6 mg/dl and a hemoglobin level of 10.7 g/dl on day 15. His physical growth and developmental processes at 9 months of age revealed no abnormalities of neurodevelopmental maturation. 2.2. Serology Blood samples of the father and other family members were collected with informed consent. Serologic ABO grouping of RBCs was carried out using standard hemagglutination techniques. Saliva sample of the father were collected to assayed blood group A or B substance. The following commercial anti-sera and lectins were used: murine monoclonal anti-A, anti-B, anti-AB reagents (Bioscot, UK), human anti-A, anti-B serum (ShuBao, Chengdu, China), anti-H from Ulex europaeus, and anti-A1 from Dolichos biflourus (Dominion Biologicals, Canada). Murine monoclonal anti-IgG (Immucor, Houston, US) was used to perform the direct antiglobulin. All reagents were used according to the manufacturer’s instructions. The isoagglutinin titers were determined by testing serial two-fold dilutions of the serum against A1 or B RBCs collected from randomly selected adults. The reciprocal of the highest dilution of serum that presents a ‘‘1+” reactions with A or B RBCs is referred to as the titer. IgG anti-A or antiB antibody was examined respectively in AHG phase after IgM anti-A and anti-B had been inactivated by 2-mercaptoethanol. 2.3. ABO genotyping A commercial PCR-SSP-based kit (G&T Biotech, Rockville, MD) was used for ABO genotyping. This kit is designed to detect the alleles A101, B101, O01 and A102 which are the most common alleles in Asian populations. The amplification products were separated electrophoretically and visualized on a 4% agarose gel prestained with ethidium bromide (0.2 lg/ ml). The coding sequences and flanking intronic regions of exons 6 and 7 of the ABO gene were analyzed by direct sequencing as previously described [5,6]. The region from exon 6 to 7 was further analyzed by cloning and sequencing to determine the cis/trans linkage of polymorphisms in this region.

Z.-H. Deng et al. / Transfusion and Apheresis Science 39 (2008) 123–128

A 2.1-kb fragment spanning a region from intron 5 to exon 7 was amplified using the primers 50 -CTGGAAGGGTGGTCAGAGGA-30 (sense; intron 5, nt 428–447) and 50 -GTTACTCACAACAGGACGGAC-30 (antisense; 30 -UTR, nt 91–111). The amplification was carried out in a volume of 50 lL containing 2 GC buffer I/II, 100 lM each dNTP, 0.1 lM each of two primers, 500 ng of genomic DNA, and 1 U of LA Taq polymerase (Takara, Dalian, China). The PCR conditions was performed: 1 cycle of 95 °C for 10 min; 30 cycles of 94 °C for 30 s and 60 °C for 30 s and 72 °C for 150 s; 72 °C for 10 min. The gel-purified PCR product was cloned into the PCR II vector with TOPO Cloning kit (Invitrogen, Groningen, Netherlands). Specific sequencing reaction was performed with the following five forward primers: F1, 50 -GGC GGC CGT GTG CCA GA-30 ; F2, 50 -TTG TCC TCC CAG AGG GTA GA-30 ; F3, 50 -CAA CCG CAG ACA CAT ACT TGA-30 ; F4, 50 -CAG GAC GGG CCT CCT GCA30 ; F5, 50 -CCA GTC CCA GGC CTA CAT-30 . 2.4. Paternity testing To clarify the parental relationships between the propositus and his parents, as well as the propositus’ father and his grandparents, the Li family-a total of three generations with eight members- were subjected to paternity testing. DNA samples were amplified using the ABI AmpFlSTR Identifier PCR amplification kit (Applied Biosystems) according to the manufacturer’s instructions. This multiplex fluorescent amplification kit consists of fifteen autosomal

125

short-tandem repeat (STR) loci including D5S818, D13S317, D7S820, D8S1179, D21S11, D18S51, D3S1358, vWA, FGA, TH01, D16S539, D2S1338, D19S433, TPOX, CSF1PO plus amelogenin locus. The PCR products were analyzed with an ABI Prism 3100 DNA sequencer and STR-allele-call were preformed using ABI Genotyper 3.7 software. Paternity index (PI) for each STR was calculated according to Essen–Mo¨ller’s formula (I = X/Y) using STR allele frequencies of south Chinese Han population data. 2.5. Classification and nomenclature of ABO alleles The ABO alleles were named according to the nomenclature used in the Blood Group Antigen Gene Mutation Database (http://www.ncbi.nlm.nih.gov/ projects/mhc/xslcgi.fcgi?cmd=bgmut/home). 3. Results The ABO phenotypes of all family members are shown in Table 1 and Fig. 1. The family members were healthy and had no history of hematologic diseases. The propositus, his father and grandfather were blood group A2B characterized by (1) a strong agglutination of RBCs with monoclonal or polyclonal anti-A, anti-B, and anti-A, B reagents; (2) strong agglutination of RBCs with lectin anti-H; (3) no agglutination of RBCs with anti-A1. The propositus’ A2B blood group was unexpected as his mother had an O phenotype. Reverse typing of his father and his grandfather samples showed no anti-A or anti-B activity, while the propositus’

Table 1 Serologic characteristics of the propositus and his family members Sample number Anti-Aa Anti-Ab Anti-A1 Anti-Ba Anti-Bb Anti-A,B Anti-H A1 cell B cell O cell Phenotype a b c

Liuc

Li family 1

2

3

4

5

6

7

8 (propositus)

4+ 4+ – 4+ 4+ 4+ 4+ – – – A2B

4+ 4+ 4+ – – 4+ 4+ – 3+ – A

4+ 4+ 3+ – – 4+ 4+ – 3+ – A

4+ 4+ – 4+ 4+ 4+ 4+ – – – A2B

– – – – – – 4+ 4+ 3+ – O

– – – – – – 4+ 4+ 3+ – O

– – – – – – 4+ 4+ 3+ – O

2+ 2+ – 4+ 4+ 4+ 4+ 2+ + – A2B

Murine monoclonal reagents (Bioscot, UK). Human polyclonal antisera (ShuBao, Chengdu, China). Serological test result of a Chinese individual with the same ABO gene molecule background previously reported [7].

4+ – – 4+ 4+ 4+ 4+ 2+ – – B(A)

126

Z.-H. Deng et al. / Transfusion and Apheresis Science 39 (2008) 123–128

2

1

A 2B cis-AB05/O02

3

A O02/A102

A A102/O01

4

A2B cis-AB05/O01

7

O O01/O02

6

5

O O02/O02

O O02/O01

8

A2B cis-AB05/O02

Fig. 1. Pedigree drawing of the Chinese Li family. The genotypes and associated phenotypes are indicated below each symbol (square for male and circle for female). The arrow indicates the propositus (newborn with signs of hemolytic disease of the newborn). The family members with the shaded symbols are heterozygous for the bispecific allele ABO  cis-AB05 (=ABO  B(A)06).

serum contained weakly reactive anti-A and anti-B antibodies. The mother’s saline isoagglutinin titers were 128 for anti-A and 512 for anti-B, while the IgG isoagglutinin titers were 64 and 512 for anti-A and Anti-B, respectively. Blood group A and B substance were found in the father’s saliva sample. All family members except for the propositus, his father and his grandfather were homozygous for common ABO alleles (Fig. 1) and their ABO genotypes were all in accordance with the associated ABO phenotypes. However, the propositus, his father and his grandfather were heterozygous for common deletional ABO  O alleles and an unusual ABO allele listed as ABO  cis-AB05 or ABO  B(A)06 in the blood group antigen gene mutation database [7,8]. This allele differs from the common ABO  B101 allele in that it has an 803C > G mutation predicting the amino acid exchange Ala268Gly. No exclusion results of parentage were observed among the three generations. The relative chance of paternity (RCP) values for the uncle, father, aunt and the propositus’ grandparents (individuals nos. 1 and 2 in Fig. 1) as well as for the propositus, his brother and their parents (individuals nos. 4 and 5 in Fig. 1) were all >99.99%. The STR genotyping results and statistical index indicated that the demonstrated paternity exist between the offsprings and parents.

4. Discussion To the best of our knowledge, this is the first detailed report of ABO-HDFN in an infant of cisAB blood type born to a blood group O mother. This case of ABO-HDFN is unusual in its degree of hyperbilirubinemia and in the fact that anti-A and anti-B antibodies were simultaneously involved in the maternal-fetal ABO incompatibility. The baby had a continually raising hyperbilirubinemia and severe anemia. Because informed consent couldn’t be obtained from the parents for exchange transfusions or simple transfusions, the baby was treated with phototherapy, IVIG and rhEPO. HDFN caused by ABO antibodies usually occurs in blood group A or B babies of group O mothers, yet IgG anti-A or-B is more often responsible than anti-A,B for HDFN and tends to cause more severe symptoms [9]. Since only IgG antibodies cross the placenta, those pregnant women with high levels of IgG anti-A,B, anti-A, or anti-B with an ABOincompatible fetus will be the ones to give birth to an infant with ABO-HDFN. The combination of a blood group O in the mother with a cisAB infant may have favoured the occurrence of HDFN in our case for two reasons: First, with the simultaneous presence of A and B determinants on the infant’s RBCs both the mater-

Z.-H. Deng et al. / Transfusion and Apheresis Science 39 (2008) 123–128

nal anti-A and anti-B isoagglutinins found their corresponding antigens. This may have very likely resulted in an higher sensitization of the infant’s RBC compared to constellations of maternal-fetal incompatibility in which only A or B antigens are present on the infant’s cells. The negative DAT in the infant may not be contrary to this assumption, as it is well established that ABO hemolytic disease can be associated with a negative or weakly reactive DAT in the presence of severe hemolysis [10,11]. Second, the relatively high IgG anti-A and anti-B titers detected in the blood group O mother could have arisen from exposure to both A and B RBCs during a previous pregnancy with a cisAB fetus. Blood group O persons exposed to A or B antigens by pregnancies or accidentally by ABO-incompatible transfusions are generally believed to produce higher amounts of IgG isoagglutinins than groups A and B individuals [12]. The only other report of a cisAB child born to a blood group O mother does not mention HDFN [13]. Unfortunately, this report was only presented in abstract form and does not provide sufficient serologic and clinical information to allow for comparison with the data of our study. Race continues to be regarded as one of the factors determining the level of ABO antibodies. From a number of studies it can be concluded that clinical ABO-HDFN is somewhat more common in Blacks than in Whites [14–16] and anti-B shows greater haemolytic activity than anti-A [17]. In Chinese, ABOHDFN is reportedly more severe than in white populations [18]. Among 1653 Chinese pregnant women typed as blood group O, 14% had IgG anti-A or anti-B titers of 64, 10% had titers of 128, 6% titers of 256 and 3% titers of 512 or more [19]. In Chinese clinical protocols, an IgG isoagglutinin titer of 64 or more is the criterion for monitoring newborns for the occurrence of HDFN. In addition, Voak and Bowley found that the majority of the sera from mothers delivering babies with HDFN due to anti-A and anti-B, respectively, contain IgG antibodies with a titer greater than 256 [20]. According to these data, our case of maternal-infant ABO incompatibility with maternal IgG anti-A and anti-B titers of 64 and 512, respectively, had to be considered as high risk. Although there is no strong correlation between the titer and the severity of HDFN, the lack of other causes in our case strongly suggests that the infant’s hemolysis was caused by the maternal ABO antibodies. There are basically two possibilities that a mother of blood group O can deliver a baby of group AB: chimerism and mosacism or the inheritance of a variant

127

ABO allele encoding for a bispecific glycosyltransferase [12,21]. By performing an extensive family study, we could clearly demonstrate that in our case a rare ABO allele encoding for a glycosyltransfersase capable of synthesizing both A and B antigens had been inherited from the father. This allele differs from ABO  B101 in that it lacks an ABO  B-specific mutation in exon 7; it is listed as ABO  cis-AB05 and ABO  B(A)06 in the blood group antigen gene mutation database. All family members heterozygous for this rare allele were blood group A2B. This is in contrast to our previous finding of this allele in an AwB phenotype [7]. Interestingly, another group reported the same rare allele to be associated with a cisAB phenotype [8]. It appears that the AB phenotypes generated by this rare allele can vary between individuals, a phenomenon that has been addressed in a recent review [22] .At least in part, the differential expression of the cisAB phenotype may be explainable by allelic interaction (competition or enhancement) of the co-inherited ABO alleles [23]. However, the fact that the ABO  cisAB05 [or ABO  B(A)06] allele was associated with an A2B phenotype in this paper but with an AwB phenotype in a recent paper is difficult to explain by allelic interaction as the rare allele was accompanied in either case by a deletional ABO  O allele (ABO  O01 or ABO  O02). Indeed, it cannot be excluded that our strategy to sequence exons 6 and 7 only led us to miss additional, phenotypically relevant mutations in the region from exons 1–5 or in regulatory sequences of the ABO  cisAB allele. In conclusion, the etiology of ABO-HDFN is complex as many factors can affect the severity of hemolysis, including the quantity and quality of maternal anti-A/B that cross the placenta as well as the degree and type of ABO antigen expression on the infant’s RBCs. It remains unclear whether and to which extend the unusual ABO phenotype of the infant contributed to the severity of the hemolytic disease in our case. However, since the cisAB phenotype is relatively common in some populations, such as the Japanese and Korean ones, this report may sensitize clinicians and immunohematologists to consider the cisAB phenomenon when monitoring pregnancies and birth with maternal-fetal ABO-incompatibility. Acknowledgement This paper was supported by the Research Fund of GuangDong Health Department, China (Project A2007589).

128

Z.-H. Deng et al. / Transfusion and Apheresis Science 39 (2008) 123–128

References [1] Dufour DR, Monoghan WP. ABO hemolytic disease of the newborn: a retrospective analysis of 254 cases. Am J Clin Pathol 1980;73:369–73. [2] Desjardins L, Blajchman MA, Chintu C, Gent M, Zipursky A. The spectrum of ABO hemolytic disease of the newborn infant. J Pediatr 1979;95:447–9. [3] Ozolek JA, Watchko JF, Mimouni F. Prevalence and lack of clinical significance of blood group incompatibility in mothers with blood type A or B. J Pediatr 1994;125:87–91. [4] Ye YW, Wang YS. National clinical laboratory work criterion. NanJin: Southeast University Publisher; 1997. [5] Olsson ML, Chester MA. Polymorphisms at the ABO locus in subgroup A individuals. Transfusion 1996;36:309–13. [6] Deng ZH, Yu Q, Wu GG, et al. Molecular genetic analysis for Ax phenotype of the ABO blood group system in Chinese. Vox Sang 2005;89:251–6. [7] Deng ZH, Yu Q, Liang YL, et al. Characterization of a novel B(A) allele with BBBA type at the ABO blood group. J Hum Genet 2006;51:732–6. [8] Blumenfeld OO, Patnaik SK. Allelic genes of blood group antigens: a source of human mutations and cSNPs documented in the blood group antigen gene mutation database. Hum Mutat 2004;23:8–16. [9] Issitt PD, Anstee DJ. Applied blood group serology. Miami: Montgomery Scientific Publications; 1998. [10] Sherer DM, Abramowicz JS, Ryan RM, et al. Severe fetal hydrops resulting from ABO incompatibility. Obstet Gynecol 1991;78:897–9. [11] Stiller RJ, Herzlinger R, Siegel S, et al. Fetal ascites associated with ABO incompatibility: case report and review of the literature. Am J Obstet Gynecol 1996;175:1371–2.

[12] Daniels G. Human blood groups. Oxford, United Kingdom: Blackwell Scientific Publications; 2002. [13] Vege S, Nickel C, Westhoff CM, et al. A new cis-AB allele encoding 268Gly on a B transferase gene. Transfusion 2005;45(Suppl.):348. [14] Bucher KA, Patterson Jr AM, Elston RC, et al. Racial difference in incidence of ABO hemolytic disease. Am J Public Health 1976;66:854–8. [15] Kirkman Jr HN. Further evidence for a racial difference in frequency of ABO hemolytic disease. J Pediatr 1977;90: 717–21. [16] Okafor LA, Enebe S. Anti-A and anti-B haemolysins, dangerous universal blood donors and the risk of ABO antagonism in a Nigerian community. Trop Geogr Med 1985;37:270–2. [17] Adewuyi JO, Gwanzura C. Racial difference between white, black Zimbabweans in the haemolytic activity of A B, O antibodies. Afr J Med Med Sci 2001;30:71–4. [18] Lin M. ABO HDN is more severe in Taiwan than in white populations. Vox Sang 1995;68:136. [19] Hu SJ, Sun H, Xu LZ, et al. Whether fetus would be ominous in blood group incompatibility if IgG antibody’s efficiency more than 1:512. Chin J Perinat Med 2003;6:82–4. [20] Voak D, Bowley CC. A detailed serologic study on the prediction and diagnosis of ABO heamolytic disease of the newborn (ABO HDN). Vox Sang 1996;17:321–48. [21] Cho D, Lee JS, Yazer MH, et al. Chimerism and mosaicism are important causes of ABO phenotype and genotype discrepancies. Immunohematol 2006;22:183–7. [22] Yazer MH, Olsson ML, Palcic MM. The cis-AB blood group phenotype: fundamental lessons in glycobiology. Transfus Med Rev 2006;2:207–17. [23] Cho D, Kee SJ, Shin JH, et al. Unusual phenotype of cisAB. Vox Sang 2003;84:336–7.