The alpha-1,3-galactosyltransferase knockout mouse. Implications for xenotransplantation

Organ xenografts in discordant combinations such as pig-to-man undergo hyperacute rejection due to the presence of naturally occurring human anti-pig xenoantibodies. The galactose alpha(1,3)-galactose epitope on glycolipids and glycoproteins is the major porcine xenoantigen recognized by these xenoantibodies. This epitope is formed by alpha(1,3)-galactosyltransferase, which is present in all mammals except man, apes, and Old World monkeys. We have generated mice lacking this major xenoantigen by inactivating the alpha(1,3)-galactosyltransferase gene. These mice are viable and have normal organs but develop cataracts. Substantially less xenoantibody from human serum binds to cells and tissues of these mice compared with normal mice. Similarly, there is less activation of human complement on cells from mice lacking the galactose alpha(1,3)-galactose epitope. These mice confirm the importance of the galactose alpha(1,3)-galactose epitope in human xenoreactivity and the logic of continuing efforts to generate pigs that lack this epitope as a source of donor organs.

Renal xenografts from triple-transgenic pigs are not hyperacutely rejected but cause coagulopathy in non-immunosuppressed baboons

BACKGROUND

The genetic modification of pigs is a powerful strategy that may ultimately enable successful xenotransplantation of porcine organs into humans.

METHODS

Transgenic pigs were produced by microinjection of gene constructs for human complement regulatory proteins CD55 and CD59 and the enzyme alpha1,2-fucosyltransferase (H-transferase, HT), which reduces expression of the major xenoepitope galactose-alpha1,3-galactose (alphaGal). Kidneys from CD55/HT and CD55/CD59/HT transgenic pigs were transplanted into nephrectomised, nonimmunosuppressed adult baboons.

RESULTS

In several lines of transgenic pigs, CD55 and CD59 were expressed strongly in all tissues examined, whereas HT expression was relatively weak and did not significantly reduce alphaGal. Control nontransgenic kidneys (n=4) grafted into baboons were hyperacutely rejected within 1 hr. In contrast, kidneys from CD55/HT pigs (n=2) were rejected after 30 hr, although kidneys from CD55/CD59/HT pigs (n=6) maintained function for up to 5 days. In the latter grafts, infiltration by macrophages, T cells, and B cells was observed at days 3 and 5 posttransplantation. The recipients developed thrombocytopenia and abnormalities in coagulation, manifested in increased clotting times and an elevation in the plasma level of the fibrin degradation product D-dimer, within 2 days of transplantation. Treatment with low molecular weight heparin prevented profound thrombocytopenia but not the other aspects of coagulopathy.

CONCLUSIONS

Strong expression of CD55 and CD59 completely protected porcine kidneys from hyperacute rejection and allowed a detailed analysis of xenograft rejection in the absence of immunosuppression. Coagulopathy appears to be a common feature of pig-to-baboon renal transplantation and represents yet another major barrier to its clinical application.

The human ICAM-2 promoter is endothelial cell-specific in vitro and in vivo and contains critical Sp1 and GATA binding sites

The expression of intercellular adhesion molecule 2 (ICAM-2) in adult tissues is restricted to vascular endothelial cells and megakaryocytes. We have previously shown that the endothelial-specific in vivo activity of the human ICAM-2 promoter is contained within a small (0.33-kilobase (kbp)) 5'-flanking region of the gene. Here we describe the in vitro characterization of this region. The ICAM-2 promoter is TATA-less, and transcription in endothelial cells initiates at four sites. Reporter gene expression directed by the promoter was 125-fold greater than vector alone in bovine aortic endothelial cells but less than 2-fold vector alone in non-endothelial (COS) cells, confirming that specificity in vivo was paralleled in vitro. The addition of 2.7 kbp of 5'-flanking region to the 0.33-kbp fragment had no effect on promoter activity or specificity. The mutation of an Sp1 motif centered at base pair -194 or an eight-base pair palindrome at -268 each reduced promoter activity by 70%. Mutation of GATA motifs at -145 and -53 reduced promoter activity by 78 and 61%, respectively. Specific binding of bovine aortic endothelial cells nuclear proteins to the Sp1 and GATA sites was demonstrated by gel shift analysis. Promoter activity in COS cells was transactivated 3-4-fold by overexpression of GATA-2. The results presented here suggest that transcription from the ICAM-2 promoter in endothelial cells is regulated by the interplay of several positive-acting factors and provide the basis for further analysis of endothelial-specific gene expression.

Changes in cell surface glycosylation in alpha1,3-galactosyltransferase knockout and alpha1,2-fucosyltransferase transgenic mice

BACKGROUND

Inactivation of the alpha1,3-galactosyltransferase (GalT) gene by homologous recombination (knockout [KO] mice) and competition for the enzyme's N-acetyllactosamine substrate by transgenically expressed alpha1,2-fucosyltransferase (H-transferase) are two genetic approaches to elimination of the Gal alpha1,3Gal (alphaGal) epitope, which is the major xenoantigen in pigs against which humans have preformed antibodies. Such genetic manipulations often have unpredictable results.

METHODS

A panel of 19 selected lectins was used to characterize the changes in cell surface glycosylation in GalT KO and H-transferase transgenic mice, compared with nontransgenic littermate controls.

RESULTS

GalT KO mice showed complete elimination of the alphaGal epitope, as reported previously. Surprisingly, however, this was associated with only a modest increase in N-acetyllactosamine residues and had little other effect on the pattern of lectin binding. In contrast, the pattern of lectin binding to H-transferase transgenic mouse cells was more profoundly disturbed and indicated, in addition to the expected expression of H substance and suppression of the alphaGal epitope, that there was a marked reduction in alpha2,3-sialylation and exposure of the normally cryptic antigens, sialylated Tn and Forssman antigens. Similar changes in lectin reactivity with porcine aortic endothelial cells were induced by neuraminidase treatment.

CONCLUSIONS

Lectins were able to bind underlying carbohydrate structures (sialylated Tn and Forssman antigens) that are normally cryptic antigens on H-transferase transgenic mouse spleen and cardiac endothelial cells, probably as a consequence of the reduction in the electronegativity of the cell surface due to reduced sialylation. As humans have preformed anti-Tn and anti-Forssman antibodies, it is possible that these structures may become targets of the xenograft rejection process, including hyperacute rejection.

Targeting gene expression to endothelial cells in transgenic mice using the human intercellular adhesion molecule 2 promoter

Genetic engineering of donor animals in xenotransplantation research has been directed largely toward obtaining expression of various immunoregulatory molecules on vascular endothelium, the initial target of recipient antibody and complement. However, specific high-level expression of transgenes throughout the vascular tree in adult animals has proved difficult to achieve, perhaps because of the inherent heterogeneity of endothelium. Using the promoter of the gene for intercellular adhesion molecule 2 (ICAM-2), which is constitutively expressed on all vascular endothelium, we have developed a system for endothelial cell gene targeting in vivo. A 334-basepair fragment from the 5' flanking region of the human ICAM-2 gene was used to drive the expression of human CD59 in transgenic mice. Strong and uniform expression of CD59 was observed on the endothelial cells of all blood vessels in the heart, kidney, lung, liver, and pancreas in the three lines of mice examined. Little or no expression was seen in other cell types, with the exception of neutrophils and monocytes. These results suggest that this small promoter region contains most of the signals necessary to endow it with endothelial cell specificity, making it a potentially valuable tool in areas ranging from xenotransplantation to gene therapy.

Anti-Gal antibody-mediated allograft rejection in alpha1,3-galactosyltransferase gene knockout mice: a model of delayed xenograft rejection

BACKGROUND

The key role of anti-galactose alpha1,3-galactose (anti-alphaGal) xenoantibodies in initiating hyperacute xenograft rejection has been clearly demonstrated using a variety of in vitro and in vivo approaches. However, the role of anti-alphaGal antibodies in mediating post-hyperacute rejection mechanisms, such as antibody-dependent cellular cytoxicity, remains to be determined, primarily because of the lack of a small animal model with which to study this phenomena.

METHODS

Hearts from wild-type mice were transplanted heterotopically into alpha1,3-galactosyltransferase knockout (Gal KO) mice, which like humans develop antibodies to the disaccharide galactose alpha1,3-galactose (Gal). At the time of rejection, hearts were examined histologically to determine the mechanism of rejection.

RESULTS

Hearts from wild-type mice transplanted into high-titer anti-alphaGal recipients were rejected in 8-13 days. Histological examination demonstrated a cellular infiltrate consisting of macrophages (80-90%), natural killer cells (5-10%), and T cells (1-5%). In contrast, wild-type hearts transplanted into low anti-Gal titer recipients demonstrated prolonged (>90 day) survival. However, a significant proportion (30-40%) of these underwent a minor rejection episode between 10 and 13 days, but then recovered ("accommodated").

CONCLUSIONS

The results of this study suggest that the Gal KO mouse is a useful small animal vascularized allograft model, in which the role of anti-alphaGal antibody in graft rejection can be studied in isolation from other rejection mechanisms. The titer of anti-alphaGal antibody was found to be the critical determinant of rejection. The histopathological features of rejection in this model are very similar to other models of delayed xenograft rejection, in both the timing and composition of the cellular infiltrate. The Gal KO mouse therefore provides a new rodent model, which will aid in the identification of the distinct components involved in the pathogenesis of delayed xenograft rejection.

Protective effects of recombinant human antithrombin III in pig-to-primate renal xenotransplantation

Delayed rejection of pig kidney xenografts by primates is associated with vascular injury that may be accompanied by a form of consumptive coagulopathy in recipients. Using a life-supporting pig-to-baboon renal xenotransplantation model, we have tested the hypothesis that treatment with recombinant human antithrombin III would prevent or at least delay the onset of rejection and coagulopathy. Non-immunosuppressed baboons were transplanted with transgenic pig kidneys expressing the human complement regulators CD55 and CD59. Recipients were treated with an intravenous infusion of antithrombin III eight hourly (250 units per kg body weight), with or without low molecular weight heparin. Antithrombin-treated recipients had preservation of normal renal function for 4-5 days, which was twice as long as untreated animals, and developed neither thrombocytopenia nor significant coagulopathy during this period. Thus, recombinant antithrombin III may be a useful therapeutic agent to ameliorate both early graft damage and the development of systemic coagulation disorders in pig-to-human xenotransplantation.

Angiotensinogen is secreted by pure rat neuronal cell cultures

Previous studies are divided between those which support a neuroglial (astrocyte) source for brain angiotensinogen and those which indicate that both astrocytes and neurones synthesize the precursor of angiotensin II. In this study, separate cultures of astrocytes and neuronal cells were prepared and established as being essentially pure by appropriate immunocytochemical cell markers. Angiotensinogen production by these cultures, as measured by a direct radioimmunoassay, was 20.74 +/- 3.62 ng angiotensinogen/10(6) cells/24 h (mean +/- S.D., n = 8) for astrocytes and 4.39 +/- 0.94 ng/10(6) cells/24 h (mean +/- S.D., n = 29) for neurones. Angiotensinogen secretion from both cell types was unaffected by treatments which stimulate the regulatory secretory pathway by modulating intracellular cAMP levels. In contrast, it was reduced from 23.20 +/- 2.14 to 8.14 +/- 1.31 ng/10(6) cells/24 h (S.E.M., n = 7) in astrocyte cultures by the constitutive pathway inhibitor, monensin. Angiotensinogen secreted by astrocytes and neurones was compared to pure angiotensinogen and that in plasma and cerebrospinal fluid (CSF) by cation-exchange mono S column chromatography. Pure angiotensinogen eluted as two separate peaks corresponding to the major forms of plasma angiotensinogen, whereas angiotensinogen in CSF and culture media coeluted with a third minor form of plasma angiotensinogen. It was concluded that neuronal cells as well as astrocytes secrete angiotensinogen which is distinctly different from plasma angiotensinogen.

Knock out of alpha1,3-galactosyltransferase or expression of alpha1,2-fucosyltransferase further protects CD55- and CD59-expressing mouse hearts in an ex vivo model of xenograft rejection

BACKGROUND

Organs from transgenic animals with high-level endothelial expression of the human complement regulatory factors CD55 and CD59 are significantly protected from human complement-mediated injury. Elimination or reduction of the major xenoepitope alphaGal, achieved by knocking out the alpha1,3-galactosyltransferase gene (Gal KO) or expressing human alpha1,2-fucosyltransferase (H transferase or HTF), also affords protection, although to a lesser degree. In this study, we examined whether the protection provided by strong CD55 and CD59 expression can be augmented by the Gal KO or HTF modifications.

METHODS

Hearts from four groups of mice (wild type, CD55/CD59, CD55/CD59/HTF, and CD55/CD59/Gal KO) were perfused ex vivo with 40% human plasma. Mean heart work for each group was compared over a 60-min period.

RESULTS

Wild-type hearts ceased to function effectively within 15 min of plasma addition. CD55/CD59 hearts displayed prolonged survival and maintained approximately 10% maximum work at the end of perfusion. Introduction of Gal KO or HTF onto the CD55/CD59 background resulted in a further prolongation, with work maintained at 20-30% of the maximum level.

CONCLUSIONS

We used an ex vivo model to demonstrate that eliminating alphaGal expression further prolongs the function of mouse hearts expressing high levels of CD55 and CD59. In addition, we showed that reducing alphaGal by expressing HTF is equally as effective in prolonging CD55/CD59 heart function as knocking out Gal transferase, thus providing a feasible strategy for translating these advances to the pig.

Immunocytochemical localization of angiotensinogen in the fetal and neonatal rat brain

The aim of this study was to define the temporal appearance and regional distribution of angiotensinogen in the fetal and neonatal rat brain. This was done by immunocytochemical localization of angiotensinogen in brains from embryonic day 16 to postnatal day 12. Immunostaining was first observed on embryonic day 18, and persisted to postnatal day 2, in the choroid plexus and ependymal cells lining the third ventricle. This initial expression of angiotensinogen at embryonic day 18 was followed at postnatal day 20 by a rapid progression of angiotensinogen staining appearing in astrocytes in the paraventricular nucleus, medial preoptic area, ventromedial and arcuate hypothalamic nuclei; these areas showed the highest astrocyte staining intensity in the brain. This was followed sequentially by staining in areas of the thalamus, midbrain, forebrain and brainstem. In general, neuroglial staining was higher in regions proximal to the cerebral ventricles and cerebral aqueduct. Neuronal angiotensinogen was observed at day postnatal day 0 and later. The most consistent immunopositive areas were in the forebrain and thalamus; in particular, the hippocampus, anterior and posterior cingulate cortex, basal and lateral amygdala, the caudate-putamen, globus pallidus, lateral septum, medial habenular nuclei and lateral thalamic nuclei. Most of the immunopositive cells in the hypothalamus and brainstem were astrocytes, while those in the cortex were almost exclusively neurons. Staining in thalamic regions was both neuronal and neuroglial. From the intensity of staining and cell density, it was determined that a rapid increase in angiotensinogen occurs between embryonic day 20 and postnatal day 0, followed by further, smaller increases postnatally. In conclusion, this study has shown that angiotensinogen, the protein from which angiotensin II is generated, is present in the rat fetal brain. The timing of its appearance supports the establishment of a renin-angiotensin system by late gestation. Its predominance in fetal hypothalamic nuclei and in thalamic, cerebellar and cortical neurons suggests major roles in prenatal fluid and electrolyte balance, in sensorimotor development and in brain maturation.

Expression of functional decay-accelerating factor (CD55) in transgenic mice protects against human complement-mediated attack

Transgenic mice expressing human CD55 were generated by microinjection of a CD55-minigene under the control of the mouse H2K(b) (MHC class I) promoter. Offspring were tested for transgene integration by PCR analysis, and for CD55 expression on peripheral blood leukocytes (PBLs) by flow cytometry. Expression levels of 15 founders ranged from 30 to 80% of that on human neutrophils. Immunohistochemical analysis of kidney, heart, liver, and lung tissue demonstrated staining for CD55 on endothelial surfaces as well as general diffuse staining throughout the tissues. The capacity of the transgenically expressed CD55 to prevent human C3 deposition on the surface of mouse splenocytes was assessed by flow cytometry. Cells from hemizygous mice incubated with 10% fresh human serum as a source of natural antibody and complement bound approximately 65% less C3 than control littermates. No further protection was seen using cells from homozygous littermates, and the protective effect was abrogated by prior incubation with an OFFi-CD55 monoclonal antibody. Similarly, transgenic mice were afforded significant protection from human serum-mediated lysis, determined using an LDH release assay. Hearts perfused with human plasma showed no increase in survival time in a modified Langendorff perfusion system, however deposition of human C3c was greatly reduced in transgenic hearts.

High-level co-expression of complement regulators on vascular endothelium in transgenic mice: CD55 and CD59 provide greater protection from human complement-mediated injury than CD59 alone

High-level endothelial expression of the human complement regulatory factor CD59 has been shown to protect transgenic mouse hearts from human complement-mediated injury in an ex vivo perfusion model. In this study we examine whether co-expression of CD55 provides additional protection. CD55/CD59 double-transgenic mice were generated by co-injection of CD55 and CD59 expression constructs driven by the human intercellular adhesion molecule 2 (ICAM-2) promoter. A line was established from one mouse that exhibited strong expression of CD55 and CD59 on vascular endothelium in the heart and other transplantable organs. An ex vivo perfusion model was used to compare hearts from these CD55/CD59 mice with hearts from a previously established line, which expressed CD59 at a similar level to the double transgenic line. CD59 hearts displayed prolonged survival compared to wild-type hearts during perfusion with 40% human plasma and maintained approximately 20% maximum work after 60 min. CD55/CD59 hearts were further protected, with work maintained at 35% of the maximum level after 60 min. The data demonstrate that high-level endothelial co-expression of CD55 and CD59 provides greater protection from human complement-mediated injury in this model than expression of CD59 alone.

High-level endothelial expression of human CD59 prolongs heart function in an ex vivo model of xenograft rejection

BACKGROUND

Hyperacute rejection of discordant xenografts is dependent on activation of the complement system of the recipient. Transgenic expression of recipient complement regulatory factors in donor tissue has proved to be a promising approach to dealing with hyperacute rejection, although the relationship between the level of complement regulatory factor expression and the degree of protection is not well established. Here, we examine this relationship using CD59 transgenic mouse hearts in an ex vivo model of xenograft rejection.

METHODS

The level of expression of CD59 in two lines of transgenic mice, in which CD59 is expressed under the control of either the murine H2Kb (MHC class I) promoter (line CA-17) or the endothelium-specific human intercellular adhesion molecule-2 promoter (line 237-7), was compared by immunohistochemistry and flow cytometry. Hearts from both groups and wild-type controls were perfused ex vivo with human plasma, and mean heart work for each group was compared over a 60-min period.

RESULTS

CD59 expression on cardiac endothelial cells isolated from homozygous CA-17 mice was 25- to 30-fold lower than that on cardiac endothelial cells from heterozygous 237-7 mice. CA-17 hearts perfused with 6% human plasma exhibited a reduction in deposition of the membrane attack complex, but not a prolongation of function, compared with nontransgenic mouse hearts. In contrast, 237-7 hearts showed significantly prolonged function during perfusion with 20% plasma.

CONCLUSIONS

High-level endothelial-specific expression of CD59 was effective in prolonging the function of mouse hearts perfused with 20% human plasma, whereas low-level, broader expression did not provide protection from 6% plasma.

Angiotensinogen secretion by single rat pituitary cells: detection by a reverse haemolytic plaque assay and cell identification by immunocytochemistry

A reverse haemolytic plaque assay (RHPA) for angiotensinogen was developed in rat hepatoma H4 cells and applied to investigate the possible secretion of angiotensinogen from rat pituitary cells in primary culture. Over a 24-hour incubation period in Cunningham chambers plaques with a mean area of 2,800 +/- 430 and 590 +/- 220 microns2/plaque (SD, n = 6) formed around all viable H4 cells and 2.8 +/- 0.59% of viable pituitary cells respectively. As a positive control PRL secretion from lactotrophs was routinely checked by the RHPA and shown to form plaques with a mean area of 4,050 +/- 1,850 microns2/plaque after a 4-hour incubation. By comparing plaque size in H4 cells with angiotensinogen release in cell culture, as quantified by radioimmunoassay, the secretion rate of angiotensinogen from pituitary cells was calculated as 22 +/- 8 ng/10(6) cells/24 h. Plaque-forming cells consisted of two morphologically distinct populations; 78% being small cells (less than 6 microns diameter) containing little cytoplasm and 22% were large (greater than 9 microns diameter) cells with an abundant cytoplasm. Immunocytochemical staining of pituitary cells after formation of plaques with anti-angiotensinogen, anti-LH and anti-PRL antiserum showed that the large plaque-forming cells were gonadotrophs and none were lactotrophs. All plaque-forming cells stained for angiotensinogen but only 44% of the viable cells which stained for angiotensinogen actually formed plaques. The possibility that cellular angiotensinogen was imported from extracellular sources was investigated by incubation of pituitary cells with pure 125I-angiotensinogen for periods up to 24 h. No uptake of the radiolabelled protein was found.(ABSTRACT TRUNCATED AT 250 WORDS)

Expression and functional analysis of glycosyl-phosphatidyl inositol-linked CD46 in transgenic mice

BACKGROUND

Complement activation plays a pivotal role in hyperacute xenograft rejection. In humans, activation of complement is regulated by a number of cell surface regulatory proteins. Membrane cofactor protein (CD46) is one such regulator that protects cells by acting as a cofactor for the factor I-mediated cleavage of C3b and C4b. Transgenic animals expressing human CD46 may provide organs that are resistant to complement attack. However, attempts to generate mice expressing human CD46 using cDNA-based constructs have been largely unsuccessful.

METHODS

Transgenic mice expressing a glycosylphosphatidyl inositol (GPI)-linked form of CD46 were generated by microinjection of a hybrid CD46/CD55 cDNA under the control of the human intercellular adhesion molecule-2 promoter. Expression of CD46-GPI on the vascular endothelium was determined by immunohistochemistry. The ability of CD46-GPI to protect mouse tissues from human complement attack was determined using an ex vivo isolated perfused heart model.

RESULTS

Three founder animals expressing CD46-GPI were identified. Histological analysis showed strong and uniform expression of CD46-GPI on the vascular endothelium of all organs examined. Ex vivo perfusion of transgenic mouse hearts with human plasma showed a reduction in C3c deposition and a slightly prolonged function compared with controls.

CONCLUSIONS

High-level expression of CD46-GPI was achieved in transgenic mice by using a modified cDNA-based construct. The CD46-GPI was functional, providing some protection from complement-mediated damage in the ex vivo model, and may be useful in xenotransplantation if expressed in combination with CD55 and CD59.

Anti-xenograft immune responses in alpha 1,3-galactosyltransferase knock-out mice

Although originally generated to test the effect of eliminating the alpha-Gal epitope on HAR, it is becoming increasingly clear that GalT KO mice offer a convenient and inexpensive model to investigate many aspects of the anti-xenorgraft immune response. Clearly, not all aspects of anti-xenograft rejection responses are identical in mice and primates, which should be kept in mind when interpreting results of GalT KO mouse studies. However, with this and other mouse models it is possible to test a large number of variables, which is impractical for both logistical and financial reasons with primates. Furthermore the short gestation time and large litter size of mice means that genetic strategies targeting different aspects of the anti-xenograft immune response can be combined and subsequently tested to identify the optimal combination of genetic and therapeutic approaches to achieve long term xenograft survival. In this regard the GalT KO mouse has been and will continue to be a valuable small animal model for the study of all facets of xenograft rejection involving anti-Gal antibodies.