rich3/ztr-tp/ztr-tp/ztr01408/ztr1947-08z xppws Sⴝ1 6/3/08 7:19 Art: TP200392 Input-ma
Quantitative In Vivo Islet Potency Assay in
Normoglycemic Nude Mice Correlates With Primary
Graft Function After Clinical Transplantation
Robert Caiazzo,1,2 Valery Gmyr,1,3 Bertrand Kremer,1 Thomas Hubert,1 Benoit Soudan,4
Bruno Lukowiak,1,3 Brigitte Vandewalle,1 Marie-Christine Vantyghem,5 Francois Pattou,1,2,3,6
and Julie Kerr-Conte1,3
Reliable assays are critically needed to monitor graft potency in islet transplantation (IT). We tested a quantitative in
vivo islet potency assay (QIVIPA) based on human C-peptide (hCP) measurement in normoglycemic nude mice after
IT under the kidney capsule. QIVIPA was initially tested by transplanting incremental doses of human islets. hCP levels
in mice were correlated with the number of several transplanted islet equivalents (r2⫽0.6, P⬍0.01). We subsequently
evaluated QIVIPA in eight islet preparations transplanted in type 1 diabetic patients. Conversely to standard criteria
including islet mass, viability, purity, adenosine triphosphate content, or glucose stimulated insulin secretion, hCP in
mice receiving 1% of the final islet product was correlated to primary graft function (hCP increase) after IT (r2⫽0.85,
P⬍0.01). QIVIPA appears as a reliable test to monitor islet graft potency, applicable to validate new methods to produce
primary islets or other human insulin secreting cells.
Keywords: Islet transplantation, Human, Mouse, Potency assay.
(Transplantation 2008;86: 000–000)
Fn1
AQ: 3
espite significant efforts to standardize human islet isolation procedures, primary islet graft function remains
variable (1). Reliable islet potency assays are therefore critically needed and numerous methods have been proposed to
monitor qualitative or quantitative characteristics of clinical
islet preparations (2). None of the currently used tests could
however be consistently correlated with primary graft function after clinical islet transplantation (IT). We describe here
a simple and reproducible quantitative in vivo islet potency
assay (QIVIPA) based on human C-peptide (hCP) measurement in normoglycemic immunodeficient mice after the
D
This work was supported by grants from French Ministry of Health (Programme Hospitalier de Recherche Clinique 2001), European Community (Fonds Européen de Développement Régional), Conseil Régional
Nord Pas de Calais (IFR 114, Institut de Médecine Prédictive et de Recherche Thérapeutique), Association de Langue Française pour l’Etude
du Diabète et des Maladies Métaboliques (DIACELL), and Juvenile Diabetes Research Foundation (6-2005-1178 ECIT Islet for Research program). RC was the recipient of stipends from Centre Hospitalier Régional
et Universitaire de Lille (“année recherche”) and Institut Georges Lopez,
Saint Didier au Mont d’Or, France.
1
INSERM U859, Thérapie Cellulaire du Diabète, Faculté de Médecine, Lille,
France.
2
Chirurgie Générale et endocrinienne, Centre Hospitalier Universitaire de
Lille, Lille, France.
3
Institut de Médecine Prédictive et de Recherche Thérapeutique, Biothérapies, Université de Lille 2, Lille, France.
4
Biochimie, Centre Hospitalier Universitaire de Lille, Lille, France.
5
Endocrinologie et Métabolismes, Centre Hospitalier Universitaire de Lille,
Lille, France.
6
Address correspondence to: Francois Pattou, Chirurgie Générale et endocrinienne, CHRU de Lille, Inserm UNIT-M 859, Faculté de Médecine,
Centre Hospitalier et Universitaire de Lille, 1 Place de Verdun, F59045
Lille, France.
E-mail: [email protected]
Received 3 September 2007. Revision requested 20 September 2007.
Accepted 22 April 2008.
Copyright © 2008 by Lippincott Williams & Wilkins
ISSN 0041-1337/08/8602-1
DOI: 10.1097/TP.0b013e31817ef846
Transplantation • Volume 86, Number 2, July 27, 2008
transplantation under the kidney capsule of a fixed fraction of
the final islet product. Initially tested with incremental doses
of human islets, the value of this model was confirmed in the
clinical setting.
Human islets were isolated at the Lille University Hospital, using a modified version of the automated Ricordi’s
method (3). According to French regulations, four islet preparations of adequate quality, but insufficient in number of
islets for transplantation, were initially used to develop the
QIVIPA model. After 24 hr culture, increasing doses of purified human islets (500, 1000, 2000, 3000 islet equivalent
[IEQ]) were transplanted under the kidney capsule of normoglycemic immunodeficient mice (Swiss nu/nu mice,
Charles River Laboratories France, Arbresle, France, n⫽25)
as previously described (4). Human C-peptide was measured
after an overnight fast in mouse tail blood, before transplantation and biweekly for 6 weeks thereafter, using a specific
radioimmunoassay (C-peptide IRMA kit, Immunotech, Prague, Czech Republic). With less than 1% of cross reactivity
between human and mouse C-peptides, this assay differentiates the secretion originating from transplanted human islets
from the endogenous murine C-peptide secretion that persists in these nondiabetic mice.
As expected, hCP was undetectable (⬍0.2 ng/mL) in
plasma of native nu/nu mice and readily measurable in fasting mice as soon as 2 weeks after transplantation (Fig. 1a).
After 6 weeks, immunostaining demonstrated the persistence
of insulin positive cells in the abundant and well-demarcated
graft in the explanted kidney (Fig. 1b). Undetectable levels of
hCP measured in five mice 1 week after nephrectomy further
confirmed the absence of endogenous hCP secretion. Noteworthy, increasing numbers of transplanted islets lead to
higher hCP levels in a dose dependant manner. As depicted in
Figure 1(c), hCP levels measured throughout the study duration (area under the curve) were significantly correlated with
the transplanted number of IEQ (r⫽0.6, P⬍0.005). Notewor-
1
AQ: 1
F1
rich3/ztr-tp/ztr-tp/ztr01408/ztr1947-08z xppws Sⴝ1 6/3/08 7:19 Art: TP200392 Input-ma
2
Transplantation • Volume 86, Number 2, July 27, 2008
C
O
L
O
R
FIGURE 1. Outcome of QIVIPA after transplantation of incremental numbers of human islet in nu/nu mice: (a) Biweekly
hCP levels in mice after transplantation; (b) Persistence of insulin positive cells under the kidney capsule, 6 weeks after
transplantation; (c) Area under the curve of hCP levels during 6 weeks.
thy, hCP levels measured at day 15 were already predictive of
the overall results (r⫽0.9, P⬍0.0001).
To confirm the clinical pertinence of QIVIPA, we subsequently designed a validation study using eight consecutive
islet preparations transplanted in four patients with brittle
type 1 diabetes enrolled in a phase 1 to 2 study of IT with the
Edmonton Protocol (clinicalTrial.gov NCT00446264). A
standardized sample (1%) of each preparation was taken
from the final clinical product and transplanted under the
kidney capsule of a nude mouse. With only one exception
because of a technical failure (capsule rupture), all mice were
successfully transplanted and remained available for hCP
measurement at day 15. In parallel, each islet preparation was
routinely evaluated as previously described (5), using standard in vitro quantitative (islet and beta cell number, islet
mass, and purity) and qualitative (viability, adenosine
triphosphate [ATP] and insulin content, and glucose stimulated insulin secretion) methods.
By design, all transplanted preparations met product
release criteria of at least 250,000 IE (and/or 4000 IE/kg) with
a purity above 30% and a viability of at least 90%, and exogenous insulin could be withdrawn within 3 weeks after last
infusion in the four patients. The primary function of each
preparation, as reflected by the increase in basal C-peptide
level (mean⫾SD: 0.9⫾0.5 ng/mL) 1 week after the infusion
(2), seemed however strikingly variable (Fig. 2). These results
suggest that even in experienced centers using “optimal” islet
preparations isolated with state of the art techniques from
selected donors, islet graft potency may vary considerably.
We also confirmed that all in vitro criteria currently
used to qualify islet preparations for transplantation, such as
islet mass, viability, purity, ATP content, or glucose stimulated insulin secretion are only poorly correlated with clinical
graft outcome (2). This study eventually failed to confirm the
recently suggested correlation of primary graft function with
the crude number of islets (6) or of beta cells (7). Other recently described methods such as beta-cell viability assessment by laser scanning cytometry (8), ADP/ATP ratio (9), or
oxygen consumption rate (10) were not tested. These more
sophisticated techniques do not however quantify islet specific function within the final cell product and have not yet
been correlated with clinical outcome.
Conversely, we found here that QIVIPA results were
significantly correlated with clinical primary graft function
(Table 1). As shown in Figure 2, hCP levels measured in non
diabetic nude mice, 2 weeks after transplantation of a standardized sample of the final product seemed more reliable
than the number of IEQs generally reported for characterizing the potency of a given islet preparation. First described
two decades ago (4), in vivo testing of human islet function in
F2
T1
rich3/ztr-tp/ztr-tp/ztr01408/ztr1947-08z xppws Sⴝ1 6/3/08 7:19 Art: TP200392 Input-ma
Brief Reports
© 2008 Lippincott Williams & Wilkins
Islet mass (IEQ/Kg)
8000
7000
p = 0.31
r2 = 0.17
6000
5000
a
4000
0.0
0.2
0.4
0.6
0.8
1.0
hCp in mice (ng/mL)
C-peptide increase in man (ng/mL)
3
p = 0.003
r2 = 0.85
2
1
b
0
0.0
0.2
0.4
0.6
0.8
1.0
C-peptide increase in man (ng/mL)
FIGURE 2. Correlation between primary graft function
in type 1 diabetic recipients (hCP increase at 1 week after
islet infusion) and (a) islet mass (number of transplanted
islet equivalents per kg) or (b) hCP levels in nu/nu mice
transplanted with 1% of the final islet product.
TABLE 1. Correlation between graft characteristics
and C-peptide increase in man
Criteria
Islet mass (IEQ/kg)
Islet number (n/kg)
Beta cell number (106 cells/kg)
Islet viability (%)
Islet purity (%)
ATP content (pmoles/40IE)
insulin secretion stimulation index
hCP in mice (ng/mL)
MeanⴞSE
r2
P
5366⫾1127
5661⫾282
2.9⫾0.5
93⫾1
51⫾6
129.9⫾26.5
2.3⫾0.6
1.3⫾0.5
0.17
0.17
0.09
0.40
0.13
0.33
0.07
0.85
0.31
0.30
0.50
0.12
0.38
0.18
0.56
0.003a
a
Statistically significant.
hCP, human C-peptide; IEQ, islet equivalent.
AQ: 2
immunocompromised mice has been extensively used for experimental studies. Hering et al. (11) also recently used this
approach to assess the potency of six clinically transplanted
islet preparations. Based on the delay of reversal of STZinduced diabetes in nod-SCID mice after transplantation of
incremental islet numbers, their assay remained however
only semiquantitative. To overcome the lability and morbid-
3
ity of STZ-induced diabetes (12), we chose here to use normoglycemic mice as previously proposed by Gaber et al. (13).
Initially undetectable in mice, fasting hCP increased within 2
weeks in a dose-dependent manner after transplantation of
incremental human islet numbers from the same islet preparations. The hCP levels measured during 6 weeks confirmed
the stable function of islets transplanted in this normoglycemic environment, mimicking the period of tight glycemic
control obtained by continuous insulin administration in the
immediate period after transplantation. When transplanted
in a non immune environment, endocrine tissues such as
islets (14) and parathyroid glands (15) become fully revascularized within 7 to 14 days. To limit the duration of the test,
we therefore chose hCP levels at day 15 as the primary endpoint for the clinical validation of QIVIPA. Sampling 1% of
the final product appeared as the best compromise between
an islet mass sufficient (i.e., at least 2000 IEQ) to ensure
largely detectable fasting hCP levels in the mouse (i.e., above
1 ng/mL), and a limited tissue volume (i.e., under 100 ␮L) to
limit the risk of technical failure. To sensitize in vivo islet
testing, Gaber (13) measured stimulated hCP levels after intraperitoneal glucose injection. This model could however be
biased by inter individual variations of in vivo glucose absorption and utilization and remained qualitative. As illustrated by our 12% failure rate, IT under the mouse kidney
capsule remains a technical challenge and requires dedicated
and well trained staff. If necessary, the fraction of the final islet
product devoted to evaluation could be reasonably increased
to transplant 2 or 3 mice per preparation and further increase
the consistency of QIVIPA. Quantitative in vivo islet potency
assay intraassay variability could also be reduced without prolonging its duration by taking into account the mean of several daily measures around day 15.
In summary, we described here a new quantitative bioassay in normoglycemic immuno-incompetent mice that
seems significantly correlated with the primary graft function
after intraportal infusion in type 1 diabetic patients. Despite
the retrospective nature of bioassays, QIVIPA could be of
significant help to monitor graft potency in IT clinical trials.
Using normalized islet numbers from the same islet preparations as illustrated in the first part of this study, this analytical
model could also facilitate the preclinical evaluation of new
procedures aiming to increase islet potency. Finally, QIVIPA
could prove precious to compare human insulin secreting
cells derived from alternative sources with human primary
islets (16).
ACKNOWLEDGMENTS
The authors are grateful to Isanga Aluka, Sandrine Belaı̈ch, Nathalie Delalleau, Rimed Ezzouaoui, Grace Lee, Nina
Mikkola, Ericka Moerman, and Violeta Raverdy, and to the
coordination teams of the Agence de la Biomédecine.
REFERENCES
1.
2.
Shapiro AM, Ricordi C, Hering BJ, et al. International trial of the Edmonton protocol for islet transplantation. N Engl J Med 2006; 355:
1318.
Bertuzzi F, Ricordi C. Prediction of clinical outcome in islet allotransplantation. Diabetes Care 2007; 30: 410.
rich3/ztr-tp/ztr-tp/ztr01408/ztr1947-08z xppws Sⴝ1 6/3/08 7:19 Art: TP200392 Input-ma
4
3.
4.
5.
6.
7.
8.
9.
Transplantation • Volume 86, Number 2, July 27, 2008
Hubert T, Gmyr V, Arnalsteen L, et al. Influence of preservation
solution on human islet isolation outcome. Transplantation 2007;
83: 270.
Ricordi C, Scharp DW, Lacy PE. Reversal of diabetes in nude mice after
transplantation of fresh and 7-day-cultured (24 degrees C) human
pancreatic islets. Transplantation 1988; 45: 994.
Vandewalle B, Douillard C, Kerr Conte J, et al. Human pancreatic islet
quality control: Easy assessment of metabolic functions. Exp Clin Endocrinol Diabetes 1999; 107: 214.
Lehmann R, Zuellig RA, Kugelmeier P, et al. Superiority of small islets
in human islet transplantation. Diabetes 2007; 56: 594.
Keymeulen B, Gillard P, Mathieu C, et al. Correlation between beta cell
mass and glycemic control in type 1 diabetic recipients of islet cell graft.
Proc Natl Acad Sci USA 2006; 103: 17444.
Ichii H, Inverardi L, Pileggi A, et al. A novel method for the assessment
of cellular composition and beta-cell viability in human islet preparations. Am J Transplant 2005; 5: 1635.
Goto M, Holgersson J, Kumagai-Braesch M, et al. The ADP/ATP ratio:
A novel predictive assay for quality assessment of isolated pancreatic
islets. Am J Transplant 2006; 6: 2483.
10.
11.
12.
13.
14.
15.
16.
Papas KK, Colton CK, Nelson RA, et al. Human islet oxygen consumption rate and DNA measurements predict diabetes reversal in nude
mice. Am J Transplant 2007; 7: 707.
Hering BJ, Kandaswamy R, Harmon JV, et al. Transplantation of cultured islets from two-layer preserved pancreases in type 1 diabetes with
anti-CD3 antibody. Am J Transplant 2004; 4: 390.
Warnock GL, Kneteman NM, Rajotte RV. Effect of diabetes on the
function of transplanted human islets of Langerhans. Transplant Proc
1990; 22: 804.
Gaber AO, Fraga D, Kotb M, et al. Human islet graft function in NODSCID mice predicts clinical response in islet transplant recipients.
Transplant Proc 2004; 36: 1108.
Menger MD, Vajkoczy P, Leiderer R, et al. Influence of experimental
hyperglycemia on microvascular blood perfusion of pancreatic islet
isografts. J Clin Invest 1992; 90: 1361.
Prigouris S, Lorentziadis M, Stylogiannis S, et al. Experimental autotransplantation of the parathyroid gland. Br J Surg 1996; 83: 410.
Narushima M, Kobayashi N, Okitsu T, et al. A human beta-cell line for
transplantation therapy to control type 1 diabetes. Nat Biotechnol 2005;
23: 1274.
JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 1 OUTPUT: Tue Jun 3 07:19:17 2008
/rich3/ztr⫺tp/ztr⫺tp/ztr01408/ztr1947⫺08z
AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES
1—The name Ricordi mentioned in the text does not match with author names given for Ref 3.
Kindly correct the text or the reference in the list whichever is incorrect.
2—Kindly spell out STZ and SCID.
3—Kindly provide the academic degree for the corresponding author.
1