Transgenic Knockout Mice with Exclusively Human Sickle Hemoglobin and Sickle Cell Disease

See allHide authors and affiliations

Science  31 Oct 1997:
Vol. 278, Issue 5339, pp. 876-878
DOI: 10.1126/science.278.5339.876


To create mice expressing exclusively human sickle hemoglobin (HbS), transgenic mice expressing human α-, γ-, and βS-globin were generated and bred with knockout mice that had deletions of the murine α- and β-globin genes. These sickle cell mice have the major features (irreversibly sickled red cells, anemia, multiorgan pathology) found in humans with sickle cell disease and, as such, represent a useful in vivo system to accelerate the development of improved therapies for this common genetic disease.

A single base pair change in codon 6 of the β-globin gene causes sickle cell anemia in individuals who are homozygous for the mutation (1). Sickle hemoglobin [HbS (α2βS 2)] undergoes polymerization upon deoxygenation, thereby distorting erythrocytes into a variety of sickled shapes, damaging the erythrocyte membrane, and ultimately causing anemia, ischemia, infarction, and progressive organ dysfunction. Despite the impressive body of knowledge that has accumulated (2), many aspects of sickle cell disease are still poorly understood and treatment options remain limited. Because of the inhibitory effects of mouse α- and β-globin on sickling, transgenic mice expressing various sickle hemoglobins (HbS, HbSAD, HbS-Antilles) develop almost none of the clinical manifestations of sickle cell disease (3). Some sickle cell disease pathology has been reported in transgenic mice bred to produce higher concentrations of the “supersickling” hemoglobins (HbSAD and HbS-Antilles) (4); however, these animals still lack important features that are commonly found in humans with sickle cell disease (5). To overcome these limitations, we have created mice that no longer express mouse α- and β-globin; instead, they express exclusively human α- and βS-globin.

Three fragments of human DNA were coinjected into fertilized mouse eggs to generate transgenic founders expressing human α- and βS-globin (6). Because γ-globin has antisickling properties, we included the Gγ- andAγ-globin genes to decrease the likelihood that erythrocytes would sickle during gestation and cause fetal death. In the particular transgenic line that was generated [Tg(Hu-miniLCRα1GγAγδβS )],Gγ- and Aγ-globin are expressed during the embryonic and fetal stages of development and not in adult mice (Fig.1A) (7). Through successive rounds of breeding with knockout mice heterozygous for deletions of the murine α- and β-globin genes, Hba0//+ Hbb0 //+ (8, 9), mice homozygous for the α- and β-globin deletions and containing the sickle transgene were generated—Tg(Hu-miniLCRα1GγAγδβS )Hba0//Hba0Hbb0//Hbb0 , hereafter called sickle cell mice (10). Many of these mice turned purple and died a few hours after birth; their death was apparently a result of hypoxia brought about by respiratory distress. Because γ-globin concentrations are relatively low [range, 4 to 26% (γ/γ+βS)] in newborn sickle cell mice (Fig. 1B) compared with newborn humans, it is likely that these deaths are caused by the sickling of erythrocytes during the critical period just after birth when the lungs must begin the task of supplying oxygen. Sickle cell mice that survived this early critical period were able to reach adulthood (many are now more than 7 months old) with normal appearance, activity, and fertility (11). Erythrocytes in adult sickle cell mice contain exclusively human α- and βS-globin (Fig. 1B). There is an excess of α-globin chain synthesis (α/βS, 1.26 ± 0.02; n = 5) (12), which indicates that these sickle cell mice are slightly β-thalassemic.

Figure 1

Globin chains in transgenic and sickle cell mice. (A) HPLC profiles showing globin-chain composition of erythrocytes from wild-type (+//+) and transgenic 12.5-day gestation fetuses and from adult transgenic mice. (B) Globin-chain composition of erythrocytes from newborn and adult sickle cell mice.

Common findings in humans with sickle cell anemia include irreversibly sickled cells (ISCs), anemia, and increased rigidity of erythrocytes. Sickle cell mice are anemic, with average hematocrits only 65% of normal, and have markedly elevated reticulocyte counts (Table 1). ISCs, indicative of repeated cycles of in vivo sickling and unsickling, were observed at a frequency of 5 to 10% in oxygenated sickle cell mouse blood (Fig.2A). Upon deoxygenation in vitro (13), classically sickled cells formed at high frequency. Erythrocytes from sickle cell mice have significantly decreased osmotic fragility and increased dynamic rigidity (Fig. 2B) as measured by osmotic gradient ektacytometry (14). The various hematologic and erythrocytic perturbations that exist in these sickle cell mice closely parallel those observed in humans with sickle cell anemia (2).

Figure 2

Morphology and cellular characteristics of erythrocytes from adult sickle cell mice. (A) Oxygenated sickle cell mouse blood showing ISCs (elongated cells). (B) Osmotic deformability profiles of erythrocytes from wild-type (stippled curve) and sickle cell (solid curve) mice.

Table 1

Hematologic values for adult sickle cell and wild-type mice. Hct, hematocrit; MCH, mean corpuscular hemoglobin content; MCV, mean corpuscular volume; MCHC, mean corpuscular hemoglobin concentration; HDW, standard deviation of hemoglobin concentration histogram distribution width. MCH, MCV, MCHC, and HDW were determined with an automated hematology analyzer (H*3 System, Bayer Diagnostics, Tarrytown, New York). Animals were 3 to 7 months of age. Values are shown as mean ± standard error of the mean (n = 7 in each group). Means for each parameter were significantly different for sickle cell and wild-type mice (P < 0.006, t test).

View this table:

The final and most serious manifestation of sickle cell disease in humans is damage to multiple organs. In sickle cell mice, kidney and heart weights increased 2-fold, and spleen weight increased 13-fold compared with wild-type controls (15). Long-term increases in cardiac output and splenic erythropoiesis, both in response to the chronic anemia that exists in these mice, are likely to be responsible for the observed increases in heart and spleen weights. Histologic analysis (16) revealed tissue damage in multiple organs (Fig. 3): kidney (fibrosis, atrophy, infarcts, cysts; Fig. 3B), lung (vascular congestion; Fig. 3D), liver (multifocal ischemic infarcts; Fig. 3, E and F), and spleen (congested sinusoidal channels). Increased iron deposits were found in liver (Kupffer cells) and kidney (tubular epithelium). The extent and nature of the congestion, atrophy, fibrosis, and infarct found in organs of these sickle cell mice is very similar to what has been reported in humans with sickle cell disease (17).

Figure 3

Sickle cell mouse organ histopathology. (A) Wild-type kidney (cortex). (B) Sickle cell kidney (cortical microinfarct and cysts). (C) Wild-type lung parenchyma. (D) Sickle cell lung parenchyma (congested capillary bed). (E) Sickle cell liver (microinfarct surrounded by healthy tissue). (F) Higher magnification showing coagulative necrosis with loss of nuclear detail in hepatocytes. (G) Sickled erythrocytes in hepatic vascular channels. Scale bars = 15 μm; stained with hematoxylin and eosin.

We have extensively reengineered the murine globin system to create mice that express exclusively human sickle hemoglobin and that faithfully recapitulate the major genetic, hematologic, and histopathologic features of humans with sickle cell anemia. In contrast to the limited studies that can be performed in humans, these animals provide an opportunity for rapidly exploring an expanded range of inquiry in an in vivo setting. As such, these sickle cell mice are likely to play an important role in furthering our understanding of the pathophysiology of sickle cell disease and in developing improved therapies for treating the more than 100,000 individuals born each year with this genetic disease.

Note added in proof: With a similar approach, we have also created mice that express exclusively normal human hemoglobin (HbA).

  • * To whom correspondence should be addressed. E-mail: c_paszty{at}


View Abstract

Stay Connected to Science

Navigate This Article