Is there a need for an alternative source of red blood cells for clinical transfusion and will gene-edited pigs fulfill that need?

Is there a need for an alternative source of red blood cells for clinical transfusion?

The transfusion of human (allogeneic) red blood cells (RBCs) is a cornerstone of modern healthcare. Indeed, RBCs are the most transplanted product in the world from human-to-human with over 120 million transfusions occurring worldwide annually¹. Some patients require transfusions throughout life and may receive greater than 1000 units of packed RBCs during their lifetime². However, many of these patients become sensitized against human RBCs, rendering treatment increasingly difficult.

Despite their vital role, allogeneic RBCs face limitations. As an example, the COVID-19 pandemic caused the cancellation of thousands of blood drives in the USA, resulting in a million fewer blood product donations. Following the pandemic, the demand for blood increased, resulting in a critical shortage of blood³.

The challenges of maintaining an adequate supply of human blood

First, the supply of RBCs is dependent on having a large healthy donor base with a complex technocratic system that can ensure that these donations are produced and stored safely and efficiently for their potential recipients⁴. Second, there is a noticeable trend of a declining healthy donor base among the major blood operators worldwide. This decline is evident in the USA where blood donations are dwindling across all ethnic and racial groups. This is particularly occurring among the fastest-growing population groups. Contributing to this trend are demographic shifts, such as an aging population, which increases the proportion of individuals unsuitable for donation and who may themselves require more blood transfusions. Other factors include the overall time it takes for donors to make donations and finding a suitable location to donate.

Are there alternative sources of RBCs?

Two sources are currently being investigated – (i) autologous or more likely allogeneic human stem cells² and (ii) gene-edited animals. Sourcing RBCs from stem cells in the very large numbers that would be required would entail the ‘industrial’ production of billions of cells, which provides an immensely difficult and expensive challenge². We suggest that sourcing RBCs from genetically-engineered pigs would be simpler. In addition, hemoglobin from the marine worm, Arenicola marina (M101), has been demonstrated to be an oxygen carrier^5, but its ability to remove carbon dioxide and regulate pH are still uncertain.

The alternative source, namely gene-edited animals, could offer a simpler yet promising solution to many of these issues. This is especially true in the modern era of organ transplantation where the xenotransplantation of pig organs into humans has already become a reality⁶ and is likely to expand rapidly during the next few years. Gene-edited pigs could provide a consistent and infection-free supply of RBCs of blood type O (see below), reducing dependence on donations from human donors.

Will gene-edited pigs fulfill the need for an alternative source of red blood cells for clinical transfusion?

Pigs with multiple gene-edits are now available whose organs are largely protected from the human immune response. The clinical transfusion of their RBCs is becoming feasible.

Advantages of pRBCs for clinical transfusion

Apart from the obvious advantage that pigs can now be extensively gene-edited to protect their cells and tissues from the human immune response, the characteristics of pig (p) RBCs are similar in many respects to those of human RBCs⁷. However, the oxygen/carbon dioxide transport efficiency of pRBCs has yet to be fully proven.

As there has been concern that the transfer of porcine endogenous retroviruses (PERVs) with a pig organ graft may be problematic to a human recipient⁸, it is important to note that pRBCs do not contain PERVs because they have no nucleus (DNA). However, this is even less of a problem today since gene-edited pigs are now available in which PERVs have been completely inactivated⁹.

Another concern relates to the transfer of exogenous pig microorganisms with the pRBCs, e.g., pig cytomegalovirus (CMV), that proved a factor in the loss of the graft in the first patient to receive a pig heart transplant¹⁰. However, potential organ-source pigs are now monitored intensively for the presence of microorganisms¹¹. To avoid the possibility of transfer of a pathogenic organism, the pigs are bred and housed in a stringent biosecure environment, which is clearly demanding and expensive¹².

Recent progress in pRBC xenotransfusion

Early studies demonstrated that RBCs from wild-type (WT, i.e., genetically-unmodified) pigs survived for only minutes after transfusion into NHPs¹³ (Table 1). To achieve more success, it was necessary to prevent the destruction of pRBCs by human (or NHP) natural (preformed) antibodies to pigs. There are three known glycan xenoantigens expressed on pig cells, including on pRBCs (Table 2). The transfusion of RBCs from pigs with knockout of a single xenoantigen (α1,3-galactosyltransferase gene-knockout [GTKO] pigs) showed little or no increase in survival¹³ (Table 1).

Table 1:

Summary of the results of previous in vivo pRBC xenotransfusion studies

Reference (Year)	Pig	NHP	Therapy	Survival
Eckermann22 (2004)	WT	Baboon	None	5min
			α-galactosidase	2h
			CVF	24 h
			α-galactosidase/CVF	72h
			α-galactosidase/CVF/ML	72h
Dor23 (2004)	WT	Baboon	Soluble Gal conjugate/CVF	15min
	GTKO	Baboon	None	5min
Tan24 (2006)	WT	Rhesus	α-galactosidase/SPA+PEG+IS	40h
Yamamoto17 (2021)	TKO	Capuchin	None	5–7d

CVF = cobra venom factor; ML = medronate liposomes; SPA+PEG = succinimide proprionate-linked methoxypolyethylene glycol.

Table 2:

Known carbohydrate xenoantigens expressed on pig cells.

Carbohydrate (Abbreviation)	Responsible enzyme	Gene-knockout pig
1.Galactose-α1,3-galactose (Gal)	α1,3-galactosyltransferase	GTKO
2.N-glycolylneuraminic acid (Neu5Gc).	CMAH	CMAH-KO
3.Sda	β−1,4N-acetylgalactosaminyltransferase.	β4GalNT2-KO

CMAH = Cytidine monophosphate-N-acetylneuraminic acid hydroxylase (CMAH).

Enzymatic treatment of pRBCs with α-galactosidase (which removes Gal expression) extended survival only marginally¹³ (Table 1). However, when treatment with α-galactosidase was combined with the covalent attachment of cyanuric acid-linked methoxypolyethylene glycol, in vitro Immunoglobulin (Ig) M binding to the pRBCs was reduced by 61%¹⁴. Similar therapy extended pRBC survival to 40 hours when the NHP also received immunosuppressive therapy¹³.

In recent years, thanks to the development of the simpler and more rapid gene-editing technology by CRISPR-Cas9, expression of all three of the known glycan xenoantigens on pig cells (against which humans have natural antibodies) can be deleted, resulting in ‘triple-knockout’ [TKO] pigs¹⁵ (Table 2), greatly reducing human antibody binding to pRBCs¹⁶ (Figure 1). The transfusion of RBCs from TKO pigs into NHPs resulted in RBC survival for up to 7 days¹⁷ (Figure 2). Direct phagocytosis (without opsonization) of pRBCs by human (or NHP) PBMC-derived macrophages was limited⁷. Loss of pRBCs was largely associated with the production of elicited anti-pig antibody and its binding to the pRBCs, resulting in complement activation and phagocytosis^{7, 17}. Therefore, although TKO pRBCs are probably protected from humoral immunity, they remain susceptible to innate and adaptive cellular responses However, the immune response to pRBCs differs from that to a pig organ transplant because RBCs do not express the major histocompatibility complex (MHC)¹⁷ that can play a major role in rejection.

Figure 1: Correlation of human serum antibodies to WT and TKO pRBCs with age.

Relative geometric mean (rGM) binding and age correlation of human serum IgM (A,C) and IgG (B,D) antibodies to WT (top) and TKO (bottom) pRBCs. The dotted lines indicate no IgM or IgG binding. (Note the great difference in the scale on the Y axis between the top and bottom figures.) (Reproduced with permission from Li Q, et al. Ann Thorac Surg. 2020;109:1268–1273)

Figure 2: Comparison of survival after allotransfusion (black bars) and xenotransfusion (red bars) in New World (capuchin) moneys.

One hour after xenotransfusion, the mean survival of TKO pRBCs was 66%, which was significantly lower than that of monkey RBCs after allotransfusion (87%) (p<0.01). One day and three days after xenotransfusion, survival of TKO RBCs had fallen to 54% and 48%, respectively, which were similar to survival after allotransfusion (55% and 44%, respectively). Mean survival of TKO pRBCs five days after xenotransfusion (2%) was significantly lower than that of allogeneic monkey RBCs after allotransfusion (43%) (p<0.05). (ns=not significant, *p<0.05, **p<0.01). (Reproduced with permission from Yamamoto T et al, Transfusion 2021;61:3104–3118.)

The protection provided by TKO might be increased further if the pRBCs expressed one or more human complement-regulatory proteins, such as CD55, and/or CD59¹⁸. CD55 mainly inhibits C3 convertase, whereas CD59 inhibits the membrane attack complex (MAC) downstream of C3. In view of opsonization by C3b/iC3b, we therefore suggest that a high level of expression of hCD55 would be more effective. Expression of hCD55 in pRBCs has proved elusive, in part perhaps because RBCs do not have a nucleus.

Potential future progress in pRBC xenotransfusion

Testing human serum complement-dependent cytotoxicity (CDC) of TKO pRBCs that express high levels of hCD55 (or another human complement-regulatory protein) is our next priority¹³. An alternative approach would be the administration of a complement inhibitor immediately before transfusing the pRBCs. This would be ethically justified in patients in whom a transfusion of pRBCs would be life-saving, e.g., in major trauma, and would be clinically feasible and should extend pRBC survival, thus providing a longer period of time in which to obtain ABO-compatible human blood.

The fact that sensitization to pig xenoantigens develops within a few days of a pRBC transfusion might well preclude subsequent pRBC transfusions (though not human blood transfusions)¹⁶. Therefore, we propose the administration of one of the Fc-modified anti-CD154 monoclonal antibodies (mAb) now available that would prevent sensitization by blockade of the CD40/CD154 T cell co-stimulation pathway¹⁹ and thus enable further transfusions of pRBCs to be administered. Anti-CD154mAb can block not only (i) the T cell-antigen-presenting cell interaction, but also T cell-B cell interactions²⁰, resulting in inhibition of class switching and de novo donor-specific antibody production²¹. However, if B cells independently (i.e., in the absence of T cell help) mature and produce anti-pRBC antibody, the addition of an immunosuppressant that deletes B cells or blocks B cell maturity could be beneficial.

This would be necessary if pRBCs were being transfused into patients with sickle cell disease who had become highly sensitized to human blood and would therefore require further transfusions of pRBCs¹⁷.

Comment

Therapy with one or more of the above agents will likely prove necessary until gene-editing provides us with pRBCs that are innately resistant to the human immune response. With future advances in gene-editing techniques, this is likely to become possible, resulting in their prolonged survival. With additional therapies, such as CD40/CD154 co-stimulation pathway blockade (with or without B cell inhibition), it is likely that pRBC xenotransfusion will become successful in life-threatening emergencies when human ABO-compatible blood is not available. Eventually, if the problem of sensitization can be prevented, we predict that pRBCs may become the standard for all blood transfusions.

Abbreviations

Gal

galactose-α1,3-galactose

GTKO

α1,3-galactosyltransferase gene-knockout

p

pig

RBC

red blood cell