The Journal of Immunology, 1998, 161: 5116-5119.
Copyright © 1998 by The American Association of Immunologists
Cutting Edge: Pig Islet Xenografts Are Susceptible to "Anti-Pig" But Not Gal
(1,3)Gal Antibody Plus Complement in Gal o/o Mice1
Ian F. C. McKenzie2,*,
Maria Koulmanda
,
Tom E. Mandel and
Mauro S. Sandrin*
*
Austin Research Institute, Austin and Repatriation Medical Centre, Heidelberg, Australia; and
Walter and Eliza Hall Institute, Royal Melbourne Hospital, Parkville, Australia
 |
Abstract
|
|---|
Hyperacute rejection due to Gal
(1,3)Gal (Gal) Ab plus
complement (C') is a major problem in xenografting vascularized organs
from pigs to primates, but the fate of neovascularized xeno islets is
unclear. Nonendocrine islet cells are Gal+, and there is a
large rise in Gal Abs after transplantation, but graft remnants persist
for some days in monkeys and humans. To define the role of Gal Ab
plus C' in porcine islet graft rejection, cultured porcine fetal islets
were grafted to mice lacking the (1,3)galactosyltransferase gene.
Anti-Gal Ab plus C' did not cause islet damage or rejection in mice
lacking the (1,3)galactosyltransferase gene, even when additional Ab
plus C' was given; in addition, hyperimmune mice (titer >1/20,000) did
not reject pig islets, showing that islets are resistant to Gal Ab plus
C'. However, islets can be destroyed by polyclonal mouse anti-pig
Abs. Thus, the focus of islet xenografting should not be on Gal Ab plus
C'.
|
Introduction
|
|---|
Amajor
problem in grafting vascularized pig organs to primates is hyperacute
rejection due to Gal(1, 3)Gal
(Gal)3 Abs. These Abs
react with the Gal carbohydrate present on the endothelial cells of the
pig organ, fix complement (C'), and cause hyperacute rejection (1).
However, the fate of neo-vascularized pancreatic islet xenografts in
the first few days after xenotransplantation is not so clear. Like
other tissues, the ducts and blood vessels of the pig pancreas are
Gal+, whereas the , ß, and 
endocrine cells (ECs)
are Gal- (2, 3). When such islets are cultured, they
express large amounts of Gal on ducts but not on ECs (4); also, a large
increase of Gal Abs occurs when pig islets are transplanted to humans
(5, 6). Thus, although the important ECs do not express Gal, the
transplant itself does; consequently, there is the potential for Ab
plus C'-mediated damage that may secondarily destroy the
Gal- ECs. However, the immediate fate of pig islets
transplanted to humans is unknown, for such grafts were not monitored
in the first hours or days during the time when Ab-mediated rejection
might occur; rather, these grafts were examined some weeks later, when
graft tissue was found in one recipient (5). It is not known whether
grafts are damaged by Gal Ab plus C' diffusing into the fetal pig
pancreas (FPP) within the first few days or whether islets are entirely
resistant to a Ab plus C'-mediated damage. Furthermore, it is possible
that the rise in Gal Ab after pig-to-human transplantation, which gives
a T cell-dependent 
60-fold rise in amount and an 100-fold
increase in the affinity of Gal Ab (5, 6), could cause delayed
rejection. To answer these questions and determine whether Gal Abs are
as important in islet transplantation as in the rejection of other
tissues, we examined Ab plus C'-mediated rejection of pig islets in Gal
o/o mice, that, like humans, lack a functional
(1, 3)galactosyltransferase gene and have naturally occurring
Gal Abs (7, 8).
We monitored the islet grafts for damage and cell infiltration at
frequent intervals over the first 120 h after transplantation,
used additional Gal Ab plus C' approximately as used in other systems
to cause xenograft rejection to ensure that these were not
limiting (9, 10), and used hyperimmune mice in which the amount
of Gal Ab present was in excess of that seen in patients after
transplantation. Finally, as the studies showed pig islets to be
resistant to Gal Ab plus C', we used a polyclonal mouse anti-pig Ab
that was able to destroy the islets in 3–5 days. Thus,
Gal- ECs are resistant to direct insult by Gal Ab plus C'
but can be destroyed by anti-pig Abs directed against other
determinants.
|
Materials and Methods
|
|---|
Mouse anti-pig serum (MAPS) was produced by immunizing
(BALB/c x CBA)F1 mice with the pig endothelial cell
line PIEC, emulsified in CFA, and administered repeatedly over 8 wk.
Normal human serum (NHS), which was obtained from healthy volunteers,
was used immediately or kept at 4°C for 
5 days. Normal rabbit serum
was used as a source of C' (RC'); rabbits were bled, blood was allowed
to clot at 4°C, and serum was collected and stored at -70°C. Gal
o/o mice were immunized with 200 µl of packed rabbit RBCs (which have
high levels of Gal (11)) administered i.p. weekly for 3 wk; Gal
Ab levels were measured by ELISA using plates coated with Gal (12).
Recipients were inbred young-adult Gal o/o mice (7, 10), which have a
mixed background of C57BL/6, 129, and DBA/2 strains; the other mice
used were C57BL/6 and SCID mice. Donor tissue was obtained from an
outbred Landrace pig (gestational age of 85 days). The FPP was
dissected and placed into organ culture in 37°C at a gas/medium
interface for 3 days in 90% O2/10% CO2 and
for 1 day in 90% air/10% CO2 (13). Grafts were then
transplanted under the left kidney capsule to seven groups of five mice
each: 1) Gal o/o controls; 2) 0.3 ml (i.p.) of MAPS on day 0; 3) 1 ml
(i.p.) of NHS on day 0; 4) 0.5 ml (i.p.) of RC' on day 0; 5) 1 ml
(i.p.) of NHS plus 0.5 ml (i.p.) of RC' on day 0; 6) hyperimmunized
mice (Gal Ab titer >1/20,000); and 7) hyperimmunized mice receiving
0.5 ml (i.p.) of RC' on day 0. Animals received the depleting
anti-CD4 mAb GK1.5 (14) (0.5 mg/i.p./mouse) on day 0 to
prevent acute cellular rejection during the observation. Grafts were
removed at 1, 2, 4, 6, 8, 10, 12, 24, 48, 72, 96, and 120 h
posttransplantation, fixed in Bouin’s solution, and processed.
Paraffin-embedded tissue that had been sectioned at 4–5 µm was
stained with hematoxylin and eosin for the assessment of infiltration
and with Gomori’s aldehyde fuchsin for the detection of ß-cells;
immunoperoxidase staining was also used for the detection of , ß,
and cells (15). At least nine coded sections from each tissue were
examined by light microscopy. Rejection was determined by the absence
of any residual identifiable graft and the presence of an infiltrate at
the graft site.
|
Results and Discussion
|
|---|
We have shown previously in T cell immunosuppressed mice that FPP
xenografts could survive for 28 days (16); similar survival occurs in
Gal o/o mice that have anti-Gal Abs. The grafts appeared to be
histologically normal in the inbred mice, but it was not clear how much
of the islet tissue had been rejected, as early monitoring had not been
performed. Any destruction could have been obscured by the extensive
regeneration that occurs after the transplantation of fetal islets.
Therefore, it was important that pig islet xenografts be examined in
Gal o/o mice in the presence of Gal Abs and, further, that excess Ab
plus C' be provided to ensure that neither were limiting. Also,
monitoring was performed frequently in the first 120 h to detect
any features of Ab-mediated rejection (Table I
).
Pig islet rejection Gal o/o mice examined at 5 days postgrafting
Cultured FPPs were transplanted to Gal o/o mice and monitored by
histology at 1, 2, 4, 6, 8, 10, 12, 24, 48, 72, 96, and 120 h.
Over this time, there was no cellular infiltration of granulocytes,
which are indicative of Ab-mediated events, or of mononuclear cells,
which are indicative of cell-mediated rejection (Fig. 1, A and B).
However, it was possible that pig islets were intrinsically resistant
to Ab-mediated mechanisms or that insufficient Ab and/or C' was
available in the Gal o/o mice to cause damage; these possibilities were
examined.
View larger version (140K):
[in this window]
[in a new window]
|
FIGURE 1. FPP grafts in Gal o/o mice were examined at different times (1–120 h);
those shown were examined at 96 h. A, An FPP at
x100 magnification; B, A normal FPP (x400)
with islets and the central necrotic area of the transplant;
kidney cells are also present (top left). The
same appearance was seen with additional Gal Ab plus C' in the
absence or presence of RC' or in the hyperimmune mice in the absence or
presence of RC'. C (x100) and D (x400)
represent FPP grafts after MAPS that show destruction of islet tissue
and granulocyte (+eosinophil) infiltration.
Immunoperoxidase staining for insulin showed numerous insulin-producing
cells in FPPs (E, x100; F x400) in
untreated Gal Ab in the absence or presence of C' or in
hyperimmune mice in the absence or presence of RC' (as well as
the same appearance); after MAPS, few insulin-producing cells remained:
G, x100; H x400.
|
|
Destruction of pig islets by polyclonal mouse anti-pig Abs
Polyclonal MAPS (0.3 ml) was given i.p. to mice (either inbred
C57BL/6 or Gal o/o), and pig xenografts were performed and monitored in
the same 12 time intervals as described above. Little change was
seen in the grafts until 48 h, when a cell infiltrate consisting
mostly of polymorphs occurred. The infiltrate subsequently
increased, and islet tissue was mostly destroyed over the next
48–72 h; i.e., by 96–120 h, the islets had been almost totally
destroyed, showing central necrosis and a heavy infiltrate of
granulocytes in the graft (Fig. 1
, C and D).
Immunoperoxidase staining for insulin and somatostatin showed a major
reduction or disappearance of these cells (Fig. 1, E–H).
Clearly, islets can be destroyed by MAPS in a manner similar to the
destruction of rat skin or heart grafts by mouse anti-rat Ab (9, 10). There was no noticeable difference in the Ab-mediated
destruction in inbred Gal+ or Gal o/o mice. It was of
interest that >24–48 h had to elapse before any noticeable infiltrate
with polymorphs was observed. Thus, islets are not intrinsically
resistant to destruction by Ab-mediated mechanisms; islet destruction
can occur if the target cells contain the Ag and there is sufficient Ab
plus C' present. Clearly, pig islets are susceptible to
Ab-mediated damage (rejection), although not by intrinsic Gal Ab plus
C'; this mechanism must be considered in pig islet graft rejection.
Fate of pig islet xenografts in Gal o/o mice given additional Gal
Ab plus C'
FPPs were grafted to Gal o/o mice; NHS containing a high titer of
Gal Ab (17, 18) was given to ensure that sufficient Ab was present.
When 1 ml of Gal polyclonal Ab was administered, no signs of cell
infiltrate or tissue damage were seen for 120 h (Fig. 1, A
and B). Furthermore, when additional RC' was provided using
fresh rabbit serum, there was also no evidence of rejection; RC' alone
had no effect. Thus, it appears that islets can be destroyed by the
appropriate Ab (see above), whereas Gal Abs, even with additional C',
have no such effect. However, it was possible that the Abs in human
serum were too low in amount and/or affinity to cause rejection;
therefore, hyperimmune mice were used.
Pig islet cell transplantation to hyperimmune Gal o/o mice
To examine whether larger amounts of Gal Ab could cause rejection,
Gal o/o mice were hyperimmunized and had Gal Ab titers of >1:20,000.
FPP transplants were performed and monitored over 5 days; no damage
occurred, but an occasional granulocyte was noted in the graft. When
additional RC' was provided, there was also no damage noted. These
studies are important, as Groth and colleagues have shown that after
pig islet xenotransplants to humans, even with heavy immunosuppression,
a large rise of Gal Ab occurred (5, 6), in excess of a 60-fold increase
in amount and affinity; to study Ab-mediated rejection, it is necessary
to examine grafts in the first few days posttransplant, and also
later, when a large rise in Ab occurs. The amount of Ab present in the
hyperimmune mice was in excess of that which is seen in humans after
islet transplantation; if this delayed rise in Gal Ab was likely to
cause rejection, it should have been seen in these mice. We can
conclude that pig islets are resistant to destruction by Gal Abs
(both to the amount of Gal Ab seen at first exposure and to the
secondary rise in Gal Abs that occurs after transplantation);
i.e., pig islets xenografts are entirely resistant to rejection by Gal
Ab plus C' but are not resistant to other non-Gal Ab-mediated
mechanisms.
Perusal of the existing data on islet xenotransplantation in the
presence of anti-Gal Abs does not give a clear indication of the
fate of such grafts during the first few days posttransplant. In
Groth’s patients, remnants of pig tissue were found in one patient;
however, whether the graft was mostly destroyed by Gal Ab plus C' was
not apparent (5). In Mandel’s studies in immunosuppressed cynomologous
monkeys, healthy islets were observed at 42 days; however, it was still
not apparent whether these were remnants left after Gal Ab plus C'
destruction and then regeneration (19). From the present studies, we
can conclude that Gal Ab plus C' was unlikely to have directly damaged
the graft in the first few days.
It is clearly important to determine what molecules are involved in the
MAPS-mediated damage and whether molecules these could be relevant to
human anti-pig reactions. In addition, as Ab-mediated rejection can
occur, transgenic pigs expressing human C'-regulatory molecules
(currently being evaluated in vascularized organ transplants) could be
useful in islet transplants, although preliminary studies (T.E.M.,
unpublished observations) indicated that islets from
decay-accelerating factor transgenic pigs fared no better than
nontransgenic islets.
We conclude that pig islets are not destroyed by Gal Abs plus C' but
are susceptible to other Abs. In addition, pig to human/primate
xenotransplantation graft rejection mechanisms must include: 1) non-Gal
Ab plus C'; 2) Ab (via Ab-dependent cell-mediated cytotoxicity)
involving FcR+ cells; 3) macrophages (without Ab); or 4) T
cells of different types. The observation that encapsulated xenografts
but not allografts are destroyed in vivo (20) would tend to exclude
cells and very large molecules but would allow for damage caused by
other means, including prostaglandins, cytokines,
superoxides, nitric oxide, and other substances. Other than excluding
Gal Abs, most mechanisms must considered important in the rejection of
pig xenografts in Gal o/o individuals
|
Acknowledgments
|
|---|
We thank Mandy Siatskas, Dr. Yu-Qin Li (Austin Research
Institute), and Stephen Mihijlovic (Walter and Eliza
Hall Institute) for technical assistance and Susan Collins for
secretarial support.
|
Footnotes
|
|---|
1 Grant support was obtained from the National Health and Medical Research Council of Australia and the Juvenile Diabetes Foundation International, USA. 
2 Address correspondence and reprint requests to Prof. Ian F. C. McKenzie, Austin Research Institute, Austin and Repatriation Medical Centre, Studley Road, Heidelberg Vic 3084, Australia. E-mail address:
3 Abbreviations used in this paper: Gal, Gal(1,3)Gal; C', complement; EC, endocrine cell; FPP, fetal pig pancreas; Gal o/o mice, mice lacking the (1,3)galactosyltransferase gene; MAPS, mouse anti-pig serum; NHS, normal human serum; RC', rabbit C'.
Received for publication August 5, 1998.
Accepted for publication September 2, 1998.
|
References
|
|---|
-
Sandrin, M. S., I. F. C. McKenzie. 1994. Gal(1,3)Gal, the major xenoantigen(s) recognized in pigs by human natural antibodies. Immunol. Rev. 141:169.[Medline]
-
McKenzie, I. F. C., M. Koulmanda, T. E. Mandel, P.-X. Xing, M. S. Sandrin. 1995. Pig-to-human xenotransplantation: the expression of Gal(1,3)Gal epitopes on pig islet cells. Xenotransplantation 2:1.
-
Koulmanda, M., I. F. C. McKenzie, M. S. Sandrin, T. E. Mandel. 1995. Fetal pig islet xenografts in NOD/Lt mice: the effects of peritransplant anti-CD4 monoclonal antibody and graft immunomodulation on graft survival and lack of expression of Gal(1,3)Gal on endocrine cells. Xenotransplantation 2:295.
-
McKenzie, I. F. C., M. Koulmanda, T. E. Mandel, M. S. Sandrin. 1995. The expression of Gal(1,3)Gal by porcine islet cells and its relevance to xenotransplantation. Xenotransplantation 2:139.
-
Groth, C. G., O. Korsgren, A. Tibell, J. Tollemar, E. Moller, J. Bolinder, J. Ostman, F. P. Reinholt, C. Hellerstrom, A. Andreson. 1994. Transplantation of porcine fetal pancreas to diabetic patients. Lancet 344:1402.[Medline]
-
Galili, U. A., A. Tibell, B. Samuelson, L. Rydberg, C. G. Groth. 1995. Increased anti-Gal activity in diabetic patients transplanted with fetal porcine islet cell clusters. Transplantation 59:1549.[Medline]
-
Thall, A. D., P. Maly, J. B. Lowe. 1995. Oocyte Gal(1,3)Gal epitopes implicated in sperm adhesion to the zona pellucida glycoprotein ZP3 are not required for fertilization in the mouse. J. Biol. Chem. 270:21437.[Abstract/Free Full Text]
-
Tearle, R. G., M. J. Tange, Z. L. Zannettino, M. Katerelos, T. A. Shinkel, B. J. W. Van Denderen, A. J. Lonie, I. Lyons, M. B. Nottle, T. Cox, et al 1996. The -1,3-galactosyltransferase knockout mouse: implications for xenotransplantation. Transplantation 61:13.[Medline]
-
Winn, H. J., C. A. Baldmus, S. V. Jooste, P. S. Russell. 1973. Acute destruction by humoral antibody of rat skin grafted to mice: the role of complement and polymorphonuclear granulocytes. J. Exp. Med. 137:893.[Abstract]
-
McKenzie, I. F. C., Y. Q. Li,
K. Patton, A. Thall, and M. S.
Sandrin. 1998. A murine model for antibody-mediated hyperacute
rejection due to anti-Gal(1,3)Gal antibodies in Gal o/o mice.
Transplantation, In press.
-
Galili, U., A. E. Rachmilewitz, A. Peleg, I. Flechner. 1984. A unique natural human IgG antibody with anti--galactosyl specificity. J. Exp. Med. 160:1519.[Abstract/Free Full Text]
-
Vaughan, H. A., K. R. Oldenburg, M. A. Gallop, J. D. Atkin, I. F. C. McKenzie, M. S. Sandrin. 1995. Recognition of an octapeptide sequence by multiple Gal(1,3)Gal-binding proteins. Xenotransplantation 3:18.
-
Mandel, T. E., M. Koulmanda. 1994. Islet isolation from the fetal pancreas. R. P. Lanza, and W. I. Chick, eds. In Pancreatic Islet Transplantation Vol. 1:107.-124. R.G. Landes, Austin, Texas.
-
Dialynas, D. P., Z. S. Quan, K.
A. Wall, A. Pierres, J. Quintans, M.
R. Loken, and F. W. Fitch. Characterization of
the murine T cell surface molecule, designated L3T4, identified by
monoclonal antibody GK1.5: similarity of L3T4 to the human Leu-3/T4
molecule. J. Immunol. 131:2445.
-
Stenberger, L. A.. 1979. Immunocytochemistry 2nd Ed. John Wiley, New York.
-
Mandel, T. E., M. Koulmanda. 1993. Effect of immunosuppression with anti-T cell mAbs on the survival of organ-cultured fetal pig xenografts in NOD mice. Transplant. Proc. 25:2928.[Medline]
-
McKenzie, I. F. C., H. A. Vaughan, M. S. Sandrin. 1994. How important are anti-Gal(1,3)Gal antibodies in pig-to-human xenotransplants?. Xeno 2:107.
-
Sandrin, M. S., H. A. Vaughan, P. L. Dabkowski, I. F. C. McKenzie. 1993. Anti-pig IgM antibodies in human serum react predominantly with Gal1–3Gal epitopes. Proc. Natl. Acad. Sci. USA 90:11391.[Abstract/Free Full Text]
-
Mandel, T. E., M. Koulmanda, E. Cozzi, P. Waterworth, M. Tolan, G. Langford, D. J. G. White. 1997. Transplantation of normal and DAF-transgenic fetal pig pancreas into cynomologous monkeys. Transplant. Proc. 29:940.[Medline]
-
Loudovaris, T., T. E. Mandel, B. Charlton. 1996. CD4+ T cell-mediated destruction of xenografts within cell-impermeable membranes in the absence of CD8+ T cells and B cells. Transplantation 61:1678.[Medline]
Top
Abstract
Introduction
