**2. Genetic basis of G6PD**

66 Dehydrogenases

*falciparum, P. vivax, P. ovalae, P. malariae and P.knowlesi*—lead to this disease by infecting erythrocytes. Malaria can become a life-threatening condition when it is not treated. Each year, malaria leads to deaths of millions of people all around the world and a large percentage of deaths are seen in Sub-Saharan regions of Africa. As it can be easily seen, malaria and G6PD deficiency share the same geographic distribution. It was shown that G6PD enzyme has various genetic variants and polymorphic frequencies. Highly polymorphic frequencies, which are indicators of G6PD deficiency, are seen in endemic regions for malaria such as Asia, Africa, Central and South America, while in non-endemic regions, these rates decrease, suggesting the relationship between G6PD deficiency and malaria (Haworth*, et al.*, 1988, Organization, 2009, Sutherland*, et al.*, 2010). This relationship reveals two important results. One of them is that G6PD deficiency provides great protection from malaria infection, especially for falciparum infections (Motulsky, 1961, Siniscalco & Bernini, 1961, Ganczakowski*, et al.*, 1995). On the other hand, G6PD deficiency has been recently demonstrated to cause serious problems in fighting against malaria. Primaquine, which is the only drug currently, used in the treatment of Plasmodium infections leads to severe hemolysis in G6PD-deficient patients. This drug may even cause death in G6PD-deficient patients. When primaquine is administered to individuals with G6PD deficiency, its metabolites lead to more severe hemolysis by inducing oxyhemoglobin generation, GSH depletion and stimulation of the hexose monophosphate pathway (Beutler*,* 

Therefore, investigations on detection of G6PD deficiency have a vital importance for malaria patients before their treatment with primaquine. On the other hand, the methods that are used for diagnosing G6PD deficiency are unreliable. Even worse is that it is very difficult to distinguish heterozygously-deficient patients from healthy individuals (Peters & Noorden, 2009). All these data indicate that there is an urgent need to develop new methods for reliable detection of G6PD deficiency in order to prevent hemolysis in patients treated with primaquine. Current methods cannot determine primaquine sensitivity in patients with G6PD deficiency every time. However, in our previously researches, we developed a new method for the determination primaquine induced hemolysis *in vitro*. This method provides the determination of G6PD deficiency patients that are susceptible to primaquine independently from the variants of G6PD deficiency. In our studies, it was determined that this method demonstrated high sensitivity for detection of primaquine-induced hemolysis before treatment of malaria patients with primaquine. This chapter aims to represent the relationship between G6PD deficiency and malaria, and to demonstrate the method that has high sensitivity for detection of primaquine-induced hemolysis in patients with malaria whose G6PD enzyme is deficient

This chapter aims to represent the problems in treatment of malaria patients with G6PD deficiency by using primaquine, different methods for determination of G6PD deficiency and a new method to determine primaquine induced hemolysis before treatment of patients

*et al.*, 1955, Bolchoz*, et al.*, 2002, Beutler & Duparc, 2007).

before their treatment with primaquine.

with G6PD deficiency.

G6PD deficiency was identified in 1956 by Carson *et al.* (Alving*, et al.*, 1956), and its Xchromosomal inheritance was discerned in the 1950s by Childs *et al.* (Childs*, et al.*, 1958). G6PD was cloned and sequenced by Persico *et al.*(Persico*, et al.*, 1986, Persico*, et al.*, 1986) in 1986 and independently by Takizawa and Yoshida (Takizawa*, et al.*, 1986) G6PD (Misumi*, et al.*, 1982) is in the hexose monophosphate pathway, the only NADPH-generation process in mature erythrocytes, which lack the citric acid cycle. Deficiency of this enzyme in erythrocytes causes various forms of illnesses such as favism, anemia, chronic nonspherocytic hemolytic anemia, drug-sensitive hemolytic anemia, primaquine sensitivity and jaundice in newborns (Beutler*, et al.*, 1968).

By virtue of fact that G6PD is found in all cells, functional and structural studies have revealed properties of this housekeeping gene (Luzzatto, 2006). G6PD expression level is regulated by hormonal and nutritional factors in only a few tissues. G6PD expression is regulated in liver and adipose tissue, and its activity depends on the rate of fatty acid biosynthesis (Greene, 1993). The G6PD gene region is one of the first regions of the human genome to be completely sequenced (Chen*, et al.*, 1996). The gene encoding G6PD is located near the telomeric region of the distal arm of the X chromosome (Pai*, et al.*, 1980, Szabo*, et al.*, 1984, Patterson*, et al.*, 1987) (band Xq28) and a valuable X-linked genetic marker for determination of X chromosome inactivations (Migeon, 1983). G6PD has various polymorphism sites at the G6PD locus like the colorblindness, Xg blood group and the hemophilia A locus and has close linkage at the X chromosome (Boyer & Graham, 1965, Adam*, et al.*, 1967). G6PD is one of a group of genes including fragile X, (Oberle*, et al.*, 1987) color vision (Motulsky, 1988, Filosa*, et al.*, 1993) hemophilia A (Boyer & Graham, 1965) clasped-thumb mental retardation syndrome (MASA), (Macias*, et al.*, 1992) and dyskeratosis congenita (Arngrimsson*, et al.*, 1993) existing on the distal long arm of the X chromosome.

The X-linkage of the G6PD gene has important implications. This linkage is very stable and linkage with other group locuses is similar in all mammals (Luzzatto & Battistuzzi, 1985, Group, 1989, Luzzatto, 1989, Beutler, 1990). In mice, X-linkage of G6PD was shown by Epstein (Epstein, 1969). Epstein concluded that the G6PD gene is X-linked in the mouse; its synthesis occurs in the oocyte and is dosage-dependent. G6PD is a sex-linked and very polymorphic gene in populations in which males have only one allele (hemizygous) and females have two G6PD alleles. Thus, females can be either normal or deficient (homozygous), or intermediate (heterozygous) phenotypes, whereas males can be either normal or G6PD-deficient phenotype (Luzzatto, 2006). The frequency of the deficient phenotype is higher in males than females owing to males being hemizygous, in which one allele of the gene expresses the deficient phenotype; to arise in females, G6PD-deficiency needs two deficient alleles. However, hemizygous deficient males and homozygous express the same degree of enzyme deficiency level. Since deactivation of one X-chromosome in embryological development in heterozygous females have two populations of red cells (G6PD-normal and G6PD-deficient), with a wide range of total G6PD enzyme activity depending on the relative proportions. If one of the alleles contains deficiency, as a result of

random deactivation of X-chromosomes, about half of the cells will be normal and the other half will be deficient, although there is a wide range of variation around that average (Nance, 1964, Rinaldi*, et al.*, 1976). For this reason, total G6PD activity in heterozygous females can show variability between near-normal to near-deficient (Luzzatto & Battistuzzi, 1985, Segel, 2000). Deactivation of X-chromosome actualizes at random. Correspondingly, binomial distribution would be expected in deficiency level; the extent of this distribution depends on X-inactivation time in embryonic tissue and the number of cells in the embryo. Furthermore, random deactivation of one X-chromosome engenders genetic mosaics in heterozygous females (Luzzatto, 2006). As a result, G6PD mutations show the typical mendelian X-linked inheritance (Adam, 1961), severe G6PD deficiency is much more common in males than in females, and X-chromosome inactivation in heterozygous females for two different G6PD alleles indicate somatic cell mosaicism (Beutler*, et al.*, 1962, Gall*, et al.*, 1965).

Glucose-6-Phosphate Dehydrogenase

Deficiency and Malaria: A Method to Detect Primaquine-Induced Hemolysis *in vitro* 69

177 bp upstream of the translation-initiating codon (Martini*, et al.*, 1986). The G6PD activity and mRNA quantity differ between tissues (Nagel & Roth, 1989, Greene, 1993). S1 nuclease and primer extension analyses of mouse G6PD mRNA indicate that when the transcriptional start site regulated with lipogenesis in liver and adipose tissue, in kidney G6PD is expressed constitutively (Ho*, et al.*, 1988); this quantity potentially depends on oxidative stress, tissue specific differences and reductive biosynthesis reactions (Nagel & Roth, 1989, Greene, 1993). Some different mRNA forms of G6PD mRNA have been found, but their functions are completely unknown. The alternatively spliced form has been documented (Hirono & Beutler, 1988, Hirono & Beutler, 1989, Cappellini*, et al.*, 1993), but this mRNA frame contains 138 nucleotides (Mason*, et al.*, 1988, Persico*, et al.*, 1989, Bautista*, et al.*, 1992, Tang*, et al.*, 1994). Some researchers (Kanno*, et al.*, 1989) suggested that in reality, G6PD translation product made from two separate mRNAs as a result of study to be based on an artifact (Henikoff & Smith, 1989, Beutler*, et al.*, 1990, Mason*, et al.*, 1990, Yoshida & Kan, 1990).

Up to 450 G6PD variants have been identified depending on the enzyme kinetics, physicochemical characteristics, and other parameters (Luzzatto & Battistuzzi, 1985, Chen*, et al.*, 1991). Nearly 300 variants of these have been confirmed by the World Health Organization (1967). Point mutations and small deletions trigger defects in the enzyme structure. These structural defects cause altered activity, instability of the enzyme or

G6PD is a typical cytoplasmic, housekeeping enzyme and has been found in all cells from liver to kidney and organisms, from prokaryotes to yeasts, to protozoa, to plants and animals (Luzzatto & Battistuzzi, 1985, Antonenkov, 1989, Glader, 1999, Notaro*, et al.*, 2000). Inactive form of G6PD is a monomer with 515 amino acids and has a molecular weight of over 59 kDa (Rattazzi, 1968). The primary structure of the G6PD enzyme in humans has been determined from the sequence of full-length cDNA clones (Persico*, et al.*, 1986). Furthermore, the tertiary structure of the enzyme has been determined (Au*, et al.*, 1999). Dimer structure of the two subunits in the enzyme are symmetrically located across a

Activation of the enzyme requires NADP+ tightly binding to dimer or tetramer formation of enzyme. G6PD catalyses the first step of the oxidative pentose phosphate pathway and controls reaction velocity (Wrigley*, et al.*, 1972). In this first step, while G6PD catalyses the conversion of glucose 6-phosphate (G6P) to 6-phosphogluconolactone, at the same time it reduces NADP to NADPH (Au*, et al.*, 1999, Turner, 2000). Human G6PD has no activity with nicotinamide adenine dinucleotide (NAD) as coenzyme. Also, G6P is very specific for its substrate compared to other hexose phosphates (e.g., galactose 6-phosphate or mannose 6-phosphate) (Luzzatto & Battistuzzi, 1985, Glader, 1999). The G6P binding site is nearby lysine 205 in tertiary structure of the enzyme, and this amino acid has a critical role in electron transfer (Bautista*, et al.*, 1995). The NADP binding site is located nearby 38 to 43 amino acids; this region constitutes the Nterminus in tertiary structure encoded in exon 3. This site is important for stability of G6PD (Au*,* 

decreased affinity of G6PD for its substrates (Luzzatto, 2006).

**3. Structure of G6PD and enzymatic properties** 

complex interface of β-sheets (Au*, et al.*, 1999).

The total length of the gene is about 18.5 kb on the X chromosome (xq28) and contains 13 exons. Exon 13 is about 800 nucleotides long and contains the translation stop codon (Nagel & Roth, 1989, Greene, 1993). The protein-coding region is divided into 12 segments, ranging in size from 12 to 236 bp (Martini*, et al.*, 1986). Exon and intron numbers and the exon sizes and sequences are conserved in higher eukaryotes (Nagel & Roth, 1989, Greene, 1993). The first exon contains no coding sequence and intron 2 between exons 2 and 3 is extraordinarily long, extending for 9,857 bp. The function of this long intron is unknown; it may be important for transcription or processing because compressed versions of the G6PD gene still have this largest intron in some species (Chen*, et al.*, 1991, Mason*, et al.*, 1995).

The sequence of the whole G6PD gene is known (Chen*, et al.*, 1991). G6PD sequence analogy between humans and mice or rats is 87%. The analogy between the mouse and rat cDNA sequences is greater than humans with 93% similarity. Most of the sequence dissimilarity is in the 3´- UTR region, which has 600 nucleotides on average and contains a single polyA site (Nagel & Roth, 1989, Greene, 1993). G6PD gene promoter is embedded in a CpG island that starts about 680 nucleotides upstream of the transcription initiation site, extending about 1,050 nucleotides downstream of the initiation site, and ends at the start of the first intron (Chen*, et al.*, 1991). CpG island is conserved between some species (Martini*, et al.*, 1986, Toniolo*, et al.*, 1991), and has highly enriched guanine and cytosine residues, like characteristically in other housekeeping genes and this island appears to be preserved between humans and mice (Toniolo*, et al.*, 1991).

The promoter of the G6PD gene contains a TATA-like, TTAAAT sequence, and a great number of stimulatory protein 1 (Sp1) elements (Philippe*, et al.*, 1994, Rank*, et al.*, 1994, Franze*, et al.*, 1998, Hodge*, et al.*, 1998). These Sp1-binding sites are essential for promoter activity (Philippe*, et al.*, 1994). Deletion analysis has uncovered that the "essential" segment of the promoter is only about 150 bp (Ursini*, et al.*, 1990).

The transcribed region from the initiation site to the poly(A) addition site covers 15,860 bp. (Chen*, et al.*, 1991). The major 5'-end of mature G6PD mRNA in several cell lines is located 177 bp upstream of the translation-initiating codon (Martini*, et al.*, 1986). The G6PD activity and mRNA quantity differ between tissues (Nagel & Roth, 1989, Greene, 1993). S1 nuclease and primer extension analyses of mouse G6PD mRNA indicate that when the transcriptional start site regulated with lipogenesis in liver and adipose tissue, in kidney G6PD is expressed constitutively (Ho*, et al.*, 1988); this quantity potentially depends on oxidative stress, tissue specific differences and reductive biosynthesis reactions (Nagel & Roth, 1989, Greene, 1993). Some different mRNA forms of G6PD mRNA have been found, but their functions are completely unknown. The alternatively spliced form has been documented (Hirono & Beutler, 1988, Hirono & Beutler, 1989, Cappellini*, et al.*, 1993), but this mRNA frame contains 138 nucleotides (Mason*, et al.*, 1988, Persico*, et al.*, 1989, Bautista*, et al.*, 1992, Tang*, et al.*, 1994). Some researchers (Kanno*, et al.*, 1989) suggested that in reality, G6PD translation product made from two separate mRNAs as a result of study to be based on an artifact (Henikoff & Smith, 1989, Beutler*, et al.*, 1990, Mason*, et al.*, 1990, Yoshida & Kan, 1990).

Up to 450 G6PD variants have been identified depending on the enzyme kinetics, physicochemical characteristics, and other parameters (Luzzatto & Battistuzzi, 1985, Chen*, et al.*, 1991). Nearly 300 variants of these have been confirmed by the World Health Organization (1967). Point mutations and small deletions trigger defects in the enzyme structure. These structural defects cause altered activity, instability of the enzyme or decreased affinity of G6PD for its substrates (Luzzatto, 2006).
