Imaging �-Cell Death With a Near-Infrared ProbeZdravka Medarova,

1Susan Bonner-Weir,

2Myra Lipes,

2and Anna Moore

1

Evidence exists for an essential role of �-cell apoptosisin the pathology of type 1 and type 2 diabetes. Currentmethods for diabetes-associated apoptosis detection,however, suffer the drawbacks of relying on in situ–based strategies. In this study, we attempted to mea-sure, both in vitro and ex vivo, levels of �-cell apoptosisin diabetic mice using Cy5.5-labeled annexin V. We usedstreptozotocin-treated BALB/c mice and NOD mice ofdifferent ages as models of type 1 diabetes and db/dbmice as a model of type 2 diabetes. With annexin VCy5.5, we established differences in levels of apoptosisbetween diabetic and control animals. Intravenouslyadministered annexin V Cy5.5 accumulated in pancreataof diabetic mice but not in nondiabetic controls. Fur-thermore, its localization was specific to apoptoticevents within diabetic islets; its selectivity was sup-ported by transferase-mediated dUTP nick-end labelingstaining. Because annexin V defines an early marker ofapoptosis and the developed probe is suitable for in vivoadministration, it may provide a promising tool forreal-time identification in intact animals of the earlieststages of diabetes-associated �-cell death and for trac-ing the events that characterize the pathology of thedisease. Diabetes 54:1780–1788, 2005

In recent years, evidence has been accumulating thatapoptosis is a central event contributing to type 1and type 2 diabetes. In fact, the process of �-celldeath appears to constitute a major link between the

seemingly distinct disease pathways. Whereas type 1 dia-betes is characterized by the specific autoimmune recog-nition and destruction of insulin-producing �-cells, thesignals mediating �-cell destruction in type 2 diabetes areless clear but may be metabolic.

Type 1 diabetes–associated autoimmunity appears to beprimarily a T-cell–mediated process. Current theories im-plicate the process of neonatal �-cell remodeling, which is

associated with a wave of apoptosis, as a potential sourceof initiating autoantigen (1,2). Extensive evidence of �-celldeath in type 1 diabetes has been gathered using a varietyof model systems, including the NOD mouse (3) and themultiple low-dose streptozotocin (MLDS) model (1,4,5).Transgenic mouse models have been particularly useful inunraveling the complex molecular interactions accompa-nying �-cell death in type 1 diabetes (6,7). Evidence existsfor direct killing of �-cells by T-cells through the perforin/granzyme pathway (8,9) and Fas/FasL interaction (10,11),as well as for �-cell destruction through soluble mediators,the most important being interleukin-1 (12), tumor necro-sis factor-� (13), and interferon-� (14,15).

Insulin resistance, the increasing need for higher levelsof insulin in order to have glucose uptake in target tissues,is a hallmark of type 2 diabetes (16). Type 2 diabetesdevelops when there is inadequate compensation for insu-lin resistance (16). The main compensatory mechanismsto insulin resistance are enhanced function of the �-cellsand an increase in �-cell mass, resulting from a shift in thebalance between �-cell renewal (proliferation and neogen-esis) and �-cell apoptosis. In animal models, increased�-cell apoptosis has been associated with glucotoxicity(17–19), lipotoxicity (20–23), formation of amyloid depos-its (24–26), and/or genetic factors associated with defectsin insulin signal transduction (27). Recent evidence fromautopsied human pancreata suggests an increased fre-quency of �-cell apoptosis in cases of diabetes comparedwith nondiabetic control subjects (28).

Metabolic/environmental factors influence the need forinsulin secretion continually throughout the life of anorganism. The need to respond to ever-changing energydemands requires insulin secretion by the �-cell to be atightly regulated process. In type 1 and type 2 diabetes, thedelicate balance of cell renewal and cell loss is disrupted.Therefore, the ability to detect the earliest stages of �-cellapoptosis and to quantify �-cell death in vivo would beinstrumental in understanding the time course of the diseaseand in designing new potential therapeutic strategies.

To date, the evidence of diabetes-associated �-cell apo-ptosis has been almost exclusively based on histologicaldetection, namely microscopic identification of morpho-logical changes associated with apoptosis, transferase-mediated dUTP nick-end labeling (TUNEL), and activecaspase-3 detection. The first method relies on recognitionof events such as membrane blebbing, cell shrinkage, andnuclear condensation, whereas TUNEL identifies DNAfragmentation and can be positive for either apoptosis ornecrosis. A recent study explored the possibility for apo-ptosis detection in a cyclophosphamide-accelerated NODmodel of diabetes using caspase-3 labeling (29), but virtu-ally no �-cell death was identified even at the height of�-cell loss, presumably because of the rapid clearing of

From the 1MGH/MIT/HMS Athinoula A. Martinos Center for BiomedicalImaging, Department of Radiology, Massachusetts General Hospital/HarvardMedical School, Boston, Massachusetts; and the 2Joslin Diabetes Center,Harvard Medical School, Boston, Massachusetts.

Address correspondence and reprint requests to Anna Moore, PhD, MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging, Departmentof Radiology, Massachusetts General Hospital/Harvard Medical School, Room2301, Building 149, 13th St., Charlestown, MA 02129. E-mail: amoore@helix.mgh.harvard.edu.

Received for publication 13 September 2004 and accepted in revised form22 February 2005.

7-AAD, 7-amino-actinomycin; 99mTc, technetium-99m; FACS, fluorescence-activated cell sorting; FITC, fluorescein isothiocyanate; H&E, hematoxylin andeosin; MLDS, multiple low-dose streptozotocin; NIRF, near-infrared fluores-cence; SHDS, single high-dose streptozotocin; STZ, streptozotocin; TUNEL,transferase-mediated dUTP nick-end labeling.

© 2005 by the American Diabetes Association.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked “advertisement” in accordance

with 18 U.S.C. Section 1734 solely to indicate this fact.

1780 DIABETES, VOL. 54, JUNE 2005

dying �-cells. In addition, the cyclophosphamide-treated13-week-old NOD mouse model reflects a rather advancedstage of the disease. Many attempts to identify apoptoticevents associated with diabetes have been hampered bythe rapid clearing of dying �-cells as well as a long timecourse of the disease. Therefore, novel approaches for thedetection of early apoptosis would clearly contribute toour study of the progression of diabetes.

One of the earliest markers of apoptosis is the external-ization of phosphatidylserine, an aminophospholipid nor-mally found exclusively on the cytoplasmic leaflet of theplasma membrane. The externalization of phosphatidyl-serine makes it available for recognition by exogenousannexin V, a naturally occurring calcium-dependent 35-kDa protein that binds with high avidity (Kd � 1�10 mol/l)to membrane-associated aminophospholipids (30,31). An-nexin V labeled with a fluorescent tag is routinely used forhistological and cell-sorting studies to identify apoptoticcells (32). In vivo, annexin has been conjugated to tech-netium-99m (99mTc) for scintigraphy (33–36), superpara-magnetic iron oxide nanoparticles for apoptosis detectionby magnetic resonance imaging (37), and the Cy5.5 fluo-rescent dye for near-infrared optical imaging (38,39).

The success of imaging apoptosis using labeled annexinV and the fact that phosphatidylserine externalization is anearly apoptotic event prompted us to investigate thepotential of a Cy5.5-conjugated annexin V probe for thedetection of �-cell death in diabetes. In this study, wedemonstrate the successful localization of annexin V Cy5.5to apoptotic cells in diabetic islets using several modelsystems of type 1 and type 2 diabetes. Our results establishthe feasibility of using fluorescently labeled annexin V fordetection of the earliest stages of diabetes-associated�-cell death.

RESEARCH DESIGN AND METHODS

In this study we used different animal models of diabetes. The MLDS model oftype 1 diabetes involves intraperitoneal injections of streptozotocin (STZ) into6-week-old BALB/c mice at a concentration of 40 mg � kg body wt�1 � day�1 for5 consecutive days (1). The single high-dose STZ (SHDS) model involves anintraperitoneal injection of 300 mg STZ/kg body wt into 10-week-old BALB/cmice (Charles River Laboratories, Wilmington, MA). Apoptosis in both modelswas evaluated 48 h after the last STZ injection. In addition, we investigatedtype 1 diabetes–associated apoptosis in NOD mice (The Jackson Laboratory,Bar Harbor, ME) at 3, 6, 8, and 12 weeks of age. Type 2 diabetes–associatedcell death was investigated using 8-week-old BKS.Cg-m�/�Leprdb/J (db/db)mice; age-matched background strain (C57BLKS/J) mice were used as con-trols. On this background, the db/db mutation results in diabetes. Bloodglucose was measured by tail snip in fed state at 10:00 A.M. daily usingGlucometer Elite XL (Bayer, Elkhart, IN).Islet isolation and flow cytometric analysis of isolated primary pancre-

atic �-cells. Pancreatic islet isolation was performed as described byMontana et al. (40) using STZ-treated and control BALB/c mice, NOD mice ofthe specified ages, and 8-week-old db/db mice and controls. The capacity ofannexin V Cy5.5 to differentiate between apoptotic and nonapoptotic �-cellswas evaluated by fluorescence-activated cell sorting (FACS). Briefly, isolatedislets of Langerhans were dispersed by trypsin-DNase digestion for 15 min at37°C. After cell count adjustment to 1 � 105 cells per reaction, islet-derivedcells were coincubated at 4°C for 30 min with a 1:50 dilution of insulin-specificpolyclonal rabbit antiserum (MP Biomedicals, Aurora, OH), followed by afluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (VectorLaboratories, Burlingame, CA) and a 1:20 dilution of annexin V Cy5.5 (BDPharmingen, San Diego, CA). An additional control designed to distinguishdying lymphocytes from dying �-cells used flow cytometry of islet cells from8-week-old NOD mice, labeled with annexin V Cy5.5 and a hamster FITC-conjugated anti-mouse CD3 monoclonal antibody (Cedarlane Laboratories,Hornby, Ontario, Canada). Separate incubation to assess cell viability wasperformed with the vital dye 7-amino-actinomycin (7-AAD; BD Pharmingen,

San Diego, CA). Samples were analyzed using the Cell Quest software package(Becton Dickinson, San Diego, CA).Ex vivo optical imaging. For ex vivo near-infrared fluorescence (NIRF)imaging, mice were anesthetized by intraperitoneal injection of a ketamine-xylazine mixture (80 mg/kg/12 mg/kg; Parke-Davis, Morris Plains, NJ [ket-amine]; Miles, Shawnee Mission, KS [xylazine]). Annexin V was obtained fromBiovision (Mountain View, CA). Annexin V Cy5.5 (synthesized and tested asdescribed in 38) was injected intravenously via the tail vein (50 �g/animal). Sixhours postinjection, animals were killed and pancreata were removed andimaged ex vivo using a whole-body imaging system (Imaging StationIS2000MM; Kodak Scientific Imaging System, New Haven, CT) equipped witha band-pass filter set and near-infrared filter set (Omega Optical, Brattelboro,VT) connected to a Macintosh G4 computer equipped with Kodak ID3.6.3network version imaging software. Image reconstruction was performed usingIDL Virtual Machine 6.0 software with Image 2D version.

To image apoptosis in purified islets, mice were treated with MLDS andSHDS as described above. Pancreatic islets were isolated as described (40)and incubated in a 6-well plate with 5 �g/ml annexin V Cy5.5 for 2 h at 37°C.After incubation, islets were washed and the plate was placed in the imagingsystem. Imaging in the Cy5.5 channel was performed as described above.Histology and fluorescence microscopy. Immediately after the imagingsessions, pancreata were snap-frozen in liquid nitrogen and cut into serialsections. Islets were first identified using hematoxylin and eosin (H&E)staining. One section adjacent to the H&E-stained section was processed withTUNEL (Apoptag apoptosis detection kit; Chemicon International, Temecula,CA). The other section adjacent to the H&E-stained section was used foridentification of annexin V Cy5.5 accumulation. Both sections were counter-stained with the nuclear stain DAPI. Sections were then subjected tocorrelative fluorescence microscopy in the FITC (TUNEL staining) channelfor apoptosis assessment and the Cy5.5 (annexin V Cy5.5 staining) channel forprobe localization using a Nikon Eclipse 50i fluorescence microscopeequipped with an appropriate filter set (Chroma Technology, Rockingham,VT). Images were acquired using a charged-coupled device camera withnear-infrared sensitivity (SPOT 7.4 Slider RTKE; Diagnostic Instruments,Sterling Heights, MI) and analyzed using SPOT 4.0 advanced version software(Diagnostic Instruments, Sterling Heights, MI). Statistical analysis of apopto-tic rates was based on a total of 1,000 cells. Results were analyzed by Student’st test and represented as means SE.

RESULTS

We tested the capacity of a Cy5.5-conjugated annexin Vprobe to detect apoptotic �-cells in the pancreata ofdiabetic mice. We used four different mouse models: theMLDS model of type 1 diabetes, characterized by �-cellapoptosis; the SHDS model, associated with high levels ofcell death (4); the NOD mouse model, in which symptomsdevelop in an age-dependent manner; and the BKS.Cg-m�/�Leprdb/J (db/db) model of type 2 diabetes, distin-guished by loss of pancreatic �-cells.MLDS and SHDS models of type 1 diabetes. Thespecificity of annexin V Cy5.5 for apoptotic �-cells wasevaluated by flow cytometry of dispersed islet cells withdual channels for annexin V Cy5.5 (FL4) and anti-insulinFITC (FL1). Analysis was performed after gating on cellsnegative for the vital dye 7-AAD. As expected, a smallfraction of cells derived from age-matched untreatedBALB/c mice stained for annexin V. In agreement withprevious reports, the level of apoptosis in islet cells fromMLDS-treated mice was more than twofold higher thancontrols (41,42), and cell death in isolated islet cells fromSHDS-treated mice approached 40% compared with 0.75%in controls (4) (Fig. 1A). It is important to note that theapoptotic rates reported by this method may be skewedbecause of the severe impairment of islet yield as a resultof diabetes induction. This is particularly true for theSHDS model, in which our analysis was based on acomparatively low number of cells.

The differential staining was confirmed by fluorescenceimaging of STZ-treated versus nontreated isolated intact

Z. MEDAROVA AND ASSOCIATES

DIABETES, VOL. 54, JUNE 2005 1781

islets (Fig. 1B). Nontreated islets produced no signal in theCy5.5 channel with a few signal-positive foci in islets fromMLDS-treated mice. A larger fraction of islets from SHDS-treated mice stained for annexin V.

To investigate the in vivo accumulation of annexin VCy5.5, pancreata of MLDS- and SHDS-treated mice wereremoved 6 h after intravenous administration of the probeand imaged ex vivo. There were distinct foci of NIRFsignal due to probe binding in the pancreata of STZ-induced diabetic mice. By comparison, the signal associ-ated with pancreata from nontreated animals was at ornear background levels (Fig. 2A). Selective accumulationof our probe in islet regions containing apoptotic cells wasconfirmed by TUNEL assay on frozen sections: consider-ing that the two methods define different stages of apopto-sis, regions of annexin localization within islets correlatedspatially with apoptotic events detected by TUNEL assay.As expected, in the MLDS model, only a few cells withinislets bound annexin V and stained for TUNEL (2.3 0.2and 1.6 0.2%, respectively), but in the SHDS model, asignificant proportion of cells accumulated the probe andwere TUNEL positive (26.3 0.8 and 17.5 0.63%,respectively, P 0.05; Fig. 2B and Table 1).NOD model of type 1 diabetes. NOD mice ranging in agefrom 3 to 12 weeks were used to evaluate apoptosis atdifferent stages during the progression of type 1 diabetes.Average blood glucose levels in NOD mice of all ages werein the normal range and indicated that the animals were

pre-diabetic (results not shown). Flow cytometry of isletcells isolated from 3-, 6-, 8-, and 12-week-old mice showedan age-dependent progressive increase in the percentageof annexin V� cells (ranging from 0.67% at 3 weeks to3.45% at 12 weeks of age) (Fig. 3A). The majority ofannexin V� cells also stained for insulin, suggesting thatthe predominant number of apoptotic events represent�-cells. Because infiltrating lymphocytes can be a contrib-uting factor to the total population of apoptotic cells, weanalyzed cells from 8-week-old NOD mice double stainedwith annexin V Cy5.5 and an anti-CD3 antibody. Only 1% ofCD3� cells stained for annexin V, whereas 3.34% of thepopulation fell in the annexin V Cy5.5�, CD3� quadrant.Because this fraction is approximately the same as thatdetermined to be apoptotic and insulin positive, we con-clude that the majority of apoptotic cells are �-cells (Fig.3B). We need to emphasize that our estimate of thepercentage of apoptotic cells within each group is onlyrelative and attempts to clarify a trend associated with age.The trend of age-dependent increase in apoptosis wasapparent in each of three independent experiments.

The accumulation of in vivo–delivered annexin V Cy5.5appeared to be selective, because muscle tissue from thesame animals produced no signal in ex vivo NIRF images(Fig. 4A). Pancreata from 8- and 12-week-old mice haddistinct foci of NIRF signal, consistent with the hypothesisthat staining was islet specific. Histological analysis ofpancreatic tissue from 3-, 6-, 8-, and 12-week-old mice

FIG. 1. In vitro apoptosis detection using annexin V Cy5.5 in MLDS- and SHDS-treated BALB/c mice. A: Flow cytometry of cells derived from pancreaticislets of MLDS- and SHDS-treated mice and nontreated controls. A larger fraction of �-cells (FL1�, insulin) derived from diabetic mice showed a shiftin fluorescence in the FL4 (annexin V Cy5.5) channel compared with control mice. B: Fluorescence imaging of islets isolated from diabetic and controlmice and incubated in vitro with annexin V Cy5.5. Islets from diabetic mice showed higher fluorescence in the Cy5.5 channel.

IMAGING �-CELL DEATH

1782 DIABETES, VOL. 54, JUNE 2005

confirmed the differential accumulation of annexin V Cy5.5(Fig. 4B). The fraction of nuclei (DAPI) associated withfluorescence in the near-infrared range (annexin V Cy5.5)clearly increased with age. No annexin V binding wasobserved in sections derived from 3- and 6-week-old mice,whereas in 8- and 12-week-old animals, 2.6 0.34 and3.2 0.2% of nuclei were associated with probe staining(Table 1). TUNEL analysis of the consecutive slice pro-duced a signal associated with apoptotic events, whichcould be due to dying �-cells as well as infiltrating lympho-cytes. Apoptotic rates, as determined by TUNEL, were 0.6 0.1, 1.2 0.2, and 2.3 0.3% for 6-, 8-, and 12-week-old NODmice, respectively, and reflected a trend similar to the oneobtained with annexin V staining (Table 1).db/db model of type 2 diabetes. For this model, we usedthe BKS.Cg-m�/�Leprdb/J strain, in which the db/db muta-tion is placed on the highly diabetes-prone C57BLKS/J

genetic background. The average blood glucose level inthe 8-week-old db/db group was 317 36.9 mg/dl (n � 6)compared with 151 6.5 mg/dl in the controls (n � 3)(Fig. 5D). Flow cytometric analysis of islet cells isolatedfrom these diabetic db/db and control mice demonstrateddifferential binding of the annexin V Cy5.5 probe (Fig. 5A).Significantly more (7.24%) of insulin-positive, 7-AAD–neg-ative cells isolated from db/db mice had a signal in the FL4(Cy5.5) channel compared with only 0.59% cells derivedfrom control mice.

In vivo–administered annexin V Cy5.5 specifically accu-mulated in the pancreata of db/db mice, as evidenced bythe significantly higher NIRF signal yielded by the tissuefrom db/db compared with control C57BLKS/J tissue (Fig.5B). Histological examination of pancreata from diabeticmice showed Cy5.5� foci due to accumulation of the probe(Fig. 5C), with no Cy5.5 signal in sections derived fromcontrol pancreata. TUNEL staining was consistent withthe Cy5.5 staining. Relative apoptotic rates as determinedby annexin V Cy5.5 and TUNEL were comparable (2.3 0.3 and 2.0 0.12%, respectively) (Table 1). The higherlevels of apoptosis detected in db/db versus control micecorrelated with differences in blood glucose levels be-tween the two groups.

DISCUSSION

Diabetes is a major health care issue that is reachingepidemic proportions. Currently, the clinical diagnosis ofdiabetes is based on measurement of glucose levels.However, abnormal blood glucose levels are a relativelylate marker of the disease. Being able to identify earlyevents in the progression to diabetes as diagnostic mark-ers could enhance the potential for successful therapeuticintervention and lead to better understanding of the natu-ral history and the pathology of the disease.

In our study, an annexin V Cy5.5 probe was assessed forits potential to detect diabetes-associated �-cell death inmouse models of type 1 and type 2 diabetes, both in vitroand ex vivo, as a first step in noninvasive imaging of dying�-cells. However, our current concept of an expected lowfrequency of apoptosis over a fairly long time span in-creases the difficulty in measuring apoptosis in mostmodels of diabetes.

FIG. 2. Ex vivo apoptosis detection using annexin V Cy5.5 in MLDS- andSHDS-treated BALB/c mice. A: Ex vivo NIRF imaging of pancreataderived from MLDS- and SHDS-treated mice. Pancreata derived fromnontreated animals were used as controls. Strongly fluorescing fociwere visible in treated pancreata and not in controls. Results arerepresentative of three animals. B: Fluorescence microscopy of adja-cent pancreatic tissue sections derived from MLDS- and SHDS-treatedmice. Pancreatic islets were identified on H&E-stained sections. Oneadjacent section was observed under the Cy5.5 filter to evaluateannexin V accumulation. The other adjacent section was subjected toTUNEL staining (FITC channel). There was a good correlation betweenTUNEL and annexin V Cy5.5 staining. Arrows point to cells stainingwith annexin V Cy5.5 and with TUNEL. Islets from SHDS-treated micedemonstrated greater staining in both channels compared with MLDS-treated mice.

TABLE 1Apoptotic rates in pancreatic tissue sections from mouse modelsof diabetes, as determined by annexin V Cy5.5 and TUNELstaining

Annexin VCy5.5 TUNEL

Type 1 diabetesSTZ treated

MLDS 2.3 0.2 1.6 0.2SHDS 26.3 0.8 17.5 0.6

NOD3-week-old NOD 0 06-week-old NOD 0 0.6 0.18-week-old NOD 2.6 0.3 1.2 0.212-week-old NOD 3.2 0.2 2.3 0.3

Type 2 diabetesdb/db 2.3 0.3 2.0 0.1

Data are means SE (%).

Z. MEDAROVA AND ASSOCIATES

DIABETES, VOL. 54, JUNE 2005 1783

The time points used for assessment of apoptosis in theMLDS and NOD models of type 1 diabetes were early inthe progression to diabetes, as reported for both models:in the MLDS model immediately after 5 days of STZtreatment (1,41) and in 3- to 12-week-old NOD mice (3).Whereas occasional intraislet lymphocytes were detectedat 3–4 weeks of age (43–47), extensive lymphocytic infil-tration is usually seen by 8 weeks of age. As expected fromprevious studies, our analysis using this new probe foundelevated �-cell death in pancreata of STZ-induced diabeticmice (1,4,41,48) and increasing levels of apoptosis in NODmice between 3 and 12 weeks of age. This latter finding isconsistent with the observed differential priming of diabe-togenic CD8� T-cells in the pancreatic lymph nodes ofNOD mice between 4 and 8 weeks of age (5) and parallelsthe trend described by O’Brien et al. (3). In addition, weidentified increased apoptosis in pancreata of diabeticdb/db mice (49). Remarkably, in the latter model, apopto-sis was detected in both endocrine and exocrine tissue.This observation suggests the contribution from lympho-cytic cell death to the total pool of apoptotic events, apossibility that, to our knowledge, has not been addressedsystematically. The detection of immunological markersand inflammatory mediators in type 2 diabetes (50–52),however, has engendered speculations of immune cellinvolvement (53,54). Apoptosis itself can induce an im-mune response (53,55), and it is certainly feasible thatdepending on the magnitude and duration of the initialpathology, secondary phenomena, including insulitis, mayappear with different timing and severity. Generally, levelsof apoptosis, as determined by FACS of isolated primarypancreatic �-cells, were higher than the ones reported byex vivo annexin V Cy5.5 staining. This is possibly due tothe somewhat damaging isolation procedure, which haslikely resulted in overestimation of the percentage of deador dying cells by FACS.

We validated the specificity of in vivo–administeredannexin V Cy5.5 for apoptotic cells. The feasibility of usinglabeled annexin V as an in vivo probe is well supported byprevious research. In humans, 99mTc-labeled annexin V hasbeen used to detect apoptosis in the hearts of patients withacute myocardial infarction (34) as well as during acutecardiac rejection (56) and in an intracardiac tumor (57).More recently, 99mTc-labeled annexin V was used in aquantitative tumor apoptosis imaging study using single-photon emission computed tomography in patients withhead and neck carcinoma (35,36). Biodistribution anddosimetry measurements of various forms of 99mTc-la-beled annexin V in humans point to the safety of this tracerand its suitability for use in a clinical setting (58–60). Inrodents, 99mTc-labeled annexin V was tested as a marker ofapoptosis in a murine model of immune arthritis (61) andautoimmune myocarditis (62) as well as of chemotherapy-induced apoptosis in breast cancer (63) and hepatomaxenograft models (64). Annexin V labeled with the fluoro-phore Cy5.5 (Cy) has been used as a probe for imaging oftumor apoptosis using NIRF in nude mice (38,39). Consis-tent with our observations, no toxicity was observed atdoses up to 2.5 mg/kg, which is the highest reported dose(38). In addition, the small size of annexin V (35 kDa)allows for its rapid renal clearance (39).

We propose that this probe has potential as an in vivoimaging tool because of its accumulation in pancreata ofdiabetic mice after delivery through the circulation. Wewere able to observe clusters of dying cells in excisedpancreata in different diabetic animal models after intra-venous injection of the probe. This study was limited to exvivo imaging, although advances in optical imaging tech-nology that will be essential for progress toward noninva-sive in vivo measurement are anticipated. Issues such astissue scattering, limited depth penetration, low spatialresolution, and inadequate capacity for quantitative esti-

FIG. 3. In vitro apoptosis detection using annexin V Cy5.5 in pre-diabetic NOD mice. A: Flow cytometry of cells derived from pancreatic islets of3-, 6-, 8-, and 12-week-old NOD mice revealed a trend toward an increase in the detected levels of �-cell (FL1 channel, insulin) apoptosis (FL4channel, annexin V Cy5.5) between 3 and 12 weeks of age. B: Cells from 8-week-old NOD mice were stained with annexin V Cy5.5 and an anti-CD3antibody to evaluate the contribution from lymphocytes to the apoptotic cell population. The proportion of CD3� (FL1) cells that stained withannexin V Cy5.5 (FL4) was 0.94 or 22% of all apoptotic cells.

IMAGING �-CELL DEATH

1784 DIABETES, VOL. 54, JUNE 2005

mation of probe accumulation need to be addressed inorder to use in vivo optical imaging to detect and charac-terize discrete events such as isolated clusters of apoptoticcells within an internal organ.

Despite these shortcomings, optical imaging has severalattractive features such as the use of nonionizing low-energy radiation, high sensitivity with the possibility ofdetecting micrometer-sized objects, continuous data ac-quisition, and the development of potentially cost-effectiveequipment. Fast and relatively easy imaging proceduresmake this modality attractive for potential clinical use.Recently, several technical advances in developing highlysensitive detection devices have led to the biological useof cooled charged-coupled device cameras capable of

imaging very low levels of light emitted from internal bodyorgans of rodents. At the near-infrared region between 700and 900 nm, absorption by intrinsic photoactive biomol-ecules is low and allows light to penetrate into the tissueto a depth sufficient for applications in small animals (59).Therefore, imaging in the near-infrared region has theadvantage of minimizing tissue autofluorescence and dra-matically improving the target-to-background ratio (60).Fluorescence imaging can be carried out at differentresolutions and depth penetrations. For example, someprogress in resolving the above-mentioned problems hasalready been made by the application of fluorescence-mediated tomography, which delivers tomographic recon-struction of the image by mathematical modeling of

FIG. 4. Ex vivo apoptosis detection using annexin VCy5.5 in pre-diabetic NOD mice. A: Ex vivo NIRF imagingof pancreata derived from 8- and 12-week-old NOD mice.Muscle tissue defined background signal. Strongly fluo-rescing foci were visible in NOD mouse pancreata.Results are representative of three animals. B: Fluores-cence microscopy of adjacent pancreatic tissue sectionsderived from 3-, 6-, 8-, and 12-week-old NOD mice. Pan-creatic islets were identified on H&E-stained sections.One adjacent section was observed under the Cy5.5 filterto evaluate annexin V accumulation. The other adjacentsection was subjected to TUNEL staining (FITC chan-nel). There was a good correlation between TUNEL andannexin V Cy5.5 staining. Arrows point to cells stainingwith annexin V Cy5.5 and with TUNEL.

Z. MEDAROVA AND ASSOCIATES

DIABETES, VOL. 54, JUNE 2005 1785

diffusion and scattering and has achieved resolution of 1–2mm and nanomolar sensitivity (65). Alternatively, theapplication of a recently described multimodality strategy(66), combining the sensitivity of optical imaging methodswith a modality delivering higher spatial resolution andtomographic capabilities, would represent a step towardin vivo use of similar probes.

In vivo–delivered probes for the detection of apoptosisin intact animals could potentially allow real-time evalua-tion of cell death and thus circumvent the problemsassociated with using “frame-by-frame” in situ methods tomeasure apoptosis, a process that occurs over time. Suchin vivo–delivered probes would permit the tracking of thetime course of diabetes-associated �-cell loss and wouldexpand the range of tools available for study and ulti-mately for early detection of the disease. Furthermore, thistechnique could be used for evaluation of islet viabilityafter transplantation, a therapy for patients with type 1diabetes that has become more effective in the past few

years (67,68). The utility of our approach in this scenario isunderscored by the evidence that even in optimal condi-tions and in the absence of graft rejection, �60% oftransplanted islet tissue is lost 3 days after transplantationdue to cell death (69,70). Still, we need to emphasize thatthis study only represents a first step toward a potentiallynovel and valuable imaging strategy, and its true utility asa diagnostic tool remains to be validated.

ACKNOWLEDGMENTS

We acknowledge Jennifer Lock (Joslin Diabetes Center)for help with islet isolation and John Moore (MartinosCenter) for help with animal handling.

REFERENCES

1. O’Brien B, Harmon B, Cameron D, Allan D: Beta-cell apoptosis is respon-sible for the development of IDDM in the multiple low-dose streptozotocinmodel. J Pathol 178:176–181, 1996

2. Trudeau JD, Dutz JP, Arany E, Hill DJ, Fieldus WE, Finegood DT: Neonatal

FIG. 5. Apoptosis detection using annexin V Cy5.5 in dia-betic db/db mice. A: Flow cytometry of cells derived frompancreatic islets of db/db and control mice. There was asignificant difference in the percentage of �-cells (FL1channel, insulin) that stained with annexin V Cy5.5 (FL4channel, annexin V Cy5.5) between diabetic and controlmice. B: Ex vivo NIRF imaging of pancreata derived fromdb/db and C57BLKS/J (control) mice. Strong fluorescencewas associated with db/db mouse pancreata. Results arerepresentative of three animals. C: Fluorescence micros-copy of pancreatic tissue sections derived from db/db andcontrol mice. Pancreatic islets were identified on H&E-stained sections. One adjacent section was observed underthe Cy5.5 filter to evaluate annexin V accumulation. Theother adjacent section was subjected to TUNEL staining(FITC channel). There was a population of stained cells indb/db but not in control mice. Arrows point to cells stainingboth with annexin V Cy5.5 and TUNEL. D: Average bloodglucose levels in db/db and control mice. Blood glucose wassignificantly elevated in diabetic mice (P � 0.018).

IMAGING �-CELL DEATH

1786 DIABETES, VOL. 54, JUNE 2005

beta-cell apoptosis: a trigger for autoimmune diabetes? Diabetes 49:1–7,2000

3. O’Brien BA, Harmon BV, Cameron DP, Allan DJ: Apoptosis is the mode ofbeta-cell death responsible for the development of IDDM in the nonobesediabetic (NOD) mouse. Diabetes 46:750–757, 1997

4. Saini KS, Thompson C, Winterford CM, Walker NI, Cameron DP: Strepto-zotocin at low doses induces apoptosis and at high doses causes necrosisin a murine pancreatic beta cell line, INS-1. Biochem Mol Biol Int

39:1229–1236, 19965. Zhang Y, O’Brien B, Trudeau J, Tan R, Santamaria P, Dutz JP: In situ beta

cell death promotes priming of diabetogenic CD8 T lymphocytes. J Im-

munol 168:1466–1472, 20026. Green EA, Eynon EE, Flavell RA: Local expression of TNFalpha in

neonatal NOD mice promotes diabetes by enhancing presentation of isletantigens. Immunity 9:733–743, 1998

7. Kurrer MO, Pakala SV, Hanson HL, Katz JD: Beta cell apoptosis in Tcell-mediated autoimmune diabetes. Proc Natl Acad Sci U S A 94:213–218,1997

8. Kagi D, Odermatt B, Ohashi PS, Zinkernagel RM, Hengartner H: Develop-ment of insulitis without diabetes in transgenic mice lacking perforin-dependent cytotoxicity. J Exp Med 183:2143–2152, 1996

9. Kagi D, Odermatt B, Seiler P, Zinkernagel RM, Mak TW, Hengartner H:Reduced incidence and delayed onset of diabetes in perforin-deficientnonobese diabetic mice. J Exp Med 186:989–997, 1997

10. Chervonsky AV, Wang Y, Wong FS, Visintin I, Flavell RA, Janeway CA Jr,Matis LA: The role of Fas in autoimmune diabetes. Cell 89:17–24, 1997

11. Su X, Hu Q, Kristan JM, Costa C, Shen Y, Gero D, Matis LA, Wang Y:Significant role for Fas in the pathogenesis of autoimmune diabetes.J Immunol 164:2523–2532, 2000

12. Eizirik DL, Darville MI: �-Cell apoptosis and defense mechanisms: lessonsfrom type 1 diabetes. Diabetes 50 (Suppl. 1):S64–S69, 2001

13. Yang XD, Tisch R, Singer SM, Cao ZA, Liblau RS, Schreiber RD, McDevittHO: Effect of tumor necrosis factor alpha on insulin-dependent diabetesmellitus in NOD mice. I. The early development of autoimmunity and thediabetogenic process. J Exp Med 180:995–1004, 1994

14. von Herrath MG, Oldstone MB: Interferon-gamma is essential for destruc-tion of beta cells and development of insulin-dependent diabetes mellitus.J Exp Med 185:531–539, 1997

15. Wang B, Andre I, Gonzalez A, Katz JD, Aguet M, Benoist C, Mathis D:Interferon-gamma impacts at multiple points during the progression ofautoimmune diabetes. Proc Natl Acad Sci U S A 94:13844–13849, 1997

16. Weir GC, Laybutt DR, Kaneto H, Bonner-Weir S, Sharma A: �-Celladaptation and decompensation during the progression of diabetes. Dia-

betes 50 (Suppl. 1):S154–S159, 200117. Bar-On H, Ben-Sasson R, Ziv E, Arar N, Shafrir E: Irreversibility of

nutritionally induced NIDDM in Psammomys obesus is related to beta-cellapoptosis. Pancreas 18:259–265, 1999

18. Donath MY, Gross DJ, Cerasi E, Kaiser N: Hyperglycemia-induced �-cellapoptosis in pancreatic islets of Psammomys obesus during developmentof diabetes. Diabetes 48:738–744, 1999

19. Federici M, Hribal M, Perego L, Ranalli M, Caradonna Z, Perego C, UselliniL, Nano R, Bonini P, Bertuzzi F, Marlier LN, Davalli AM, Carandente O,Pontiroli AE, Melino G, Marchetti P, Lauro R, Sesti G, Folli F: High glucosecauses apoptosis in cultured human pancreatic islets of Langerhans: apotential role for regulation of specific Bcl family genes toward anapoptotic cell death program. Diabetes 50:1290–1301, 2001

20. Shimabukuro M, Ohneda M, Lee Y, Unger RH: Role of nitric oxide inobesity-induced beta cell disease. J Clin Invest 100:290–295, 1997

21. Shimabukuro M, Zhou YT, Levi M, Unger RH: Fatty acid-induced beta cellapoptosis: a link between obesity and diabetes. Proc Natl Acad Sci U S A

95:2498–2502, 199822. Maedler K, Spinas GA, Dyntar D, Moritz W, Kaiser N, Donath MY: Distinct

effects of saturated and monounsaturated fatty acids on �-cell turnoverand function. Diabetes 50:69–76, 2001

23. Piro S, Anello M, Di Pietro C, Lizzio MN, Patane G, Rabuazzo AM, VigneriR, Purrello M, Purrello F: Chronic exposure to free fatty acids or highglucose induces apoptosis in rat pancreatic islets: possible role of oxida-tive stress. Metabolism 51:1340–1347, 2002

24. Lorenzo A, Razzaboni B, Weir GC, Yankner BA: Pancreatic islet celltoxicity of amylin associated with type-2 diabetes mellitus. Nature 368:756–760, 1994

25. Schubert D, Behl C, Lesley R, Brack A, Dargusch R, Sagara Y, Kimura H:Amyloid peptides are toxic via a common oxidative mechanism. Proc Natl

Acad Sci U S A 92:1989–1993, 199526. Janson J, Ashley RH, Harrison D, McIntyre S, Butler PC: The mechanism of

islet amyloid polypeptide toxicity is membrane disruption by intermediate-sized toxic amyloid particles. Diabetes 48:491–498, 1999

27. Harding HP, Zeng H, Zhang Y, Jungries R, Chung P, Plesken H, Sabatini DD,Ron D: Diabetes mellitus and exocrine pancreatic dysfunction in perk�/�

mice reveals a role for translational control in secretory cell survival. Mol

Cell 7:1153–1163, 200128. Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA, Butler PC: �-Cell

deficit and increased �-cell apoptosis in humans with type 2 diabetes.Diabetes 52:102–110, 2003

29. Reddy S, Bradley J, Ginn S, Pathipati P, Ross JM: Immunohistochemicalstudy of caspase-3-expressing cells within the pancreas of non-obesediabetic mice during cyclophosphamide-accelerated diabetes. Histochem

Cell Biol 119:451–461, 200330. Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM:

Exposure of phosphatidylserine on the surface of apoptotic lymphocytestriggers specific recognition and removal by macrophages. J Immunol

148:2207–2216, 199231. Verhoven B, Schlegel RA, Williamson P: Mechanisms of phosphatidylser-

ine exposure, a phagocyte recognition signal, on apoptotic T lymphocytes.J Exp Med 182:1597–1601, 1995

32. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C: A novel assayfor apoptosis: flow cytometric detection of phosphatidylserine expressionon early apoptotic cells using fluorescein labelled annexin V. J Immunol

Methods 184:39–51, 199533. Blankenberg FG, Katsikis PD, Tait JF, Davis RE, Naumovski L, Ohtsuki K,

Kopiwoda S, Abrams MJ, Darkes M, Robbins RC, Maecker HT, Strauss HW:In vivo detection and imaging of phosphatidylserine expression duringprogrammed cell death. Proc Natl Acad Sci U S A 95:6349–6354, 1998

34. Hofstra L, Liem IH, Dumont EA, Boersma HH, van Heerde WL, DoevendansPA, De Muinck E, Wellens HJ, Kemerink GJ, Reutelingsperger CP, Heiden-dal GA: Visualisation of cell death in vivo in patients with acute myocardialinfarction. Lancet 356:209–212, 2000

35. van de Wiele C, Lahorte C, Vermeersch H, Loose D, Mervillie K, SteinmetzND, Vanderheyden JL, Cuvelier CA, Slegers G, Dierck RA: Quantitativetumor apoptosis imaging using technetium-99m-HYNIC annexin V singlephoton emission computed tomography. J Clin Oncol 21:3483–3487, 2003

36. Boersma HH, Liem IH, Kemerink GJ, Thimister PW, Hofstra L, Stolk LM,van Heerde WL, Pakbiers MT, Janssen D, Beysens AJ, Reutelingsperger CP,Heidendal GA: Comparison between human pharmacokinetics and imag-ing properties of two conjugation methods for 99mTc-annexin A5. Br J

Radiol 76:553–560, 200337. Schellenberger EA, Bogdanov A Jr, Hogemann D, Tait J, Weissleder R,

Josephson L: Annexin V-CLIO: a nanoparticle for detecting apoptosis byMRI. Mol Imaging 1:102–107, 2002

38. Petrovsky A, Schellenberger E, Josephson L, Weissleder R, Bogdanov A Jr:Near-infrared fluorescent imaging of tumor apoptosis. Cancer Res 63:1936–1942, 2003

39. Schellenberger EA, Bogdanov A Jr, Petrovsky A, Ntziachristos V,Weissleder R, Josephson L: Optical imaging of apoptosis as a biomarker oftumor response to chemotherapy. Neoplasia 5:187–192, 2003

40. Montana E, Bonner-Weir S, Weir GC: Beta cell mass and growth aftersyngeneic islet cell transplantation in normal and streptozocin diabeticC57BL/6 mice. J Clin Invest 91:780–787, 1993

41. Cardinal JW, Margison GP, Mynett KJ, Yates AP, Cameron DP, Elder RH:Increased susceptibility to streptozotocin-induced beta-cell apoptosis anddelayed autoimmune diabetes in alkylpurine-DNA-N-glycosylase-deficientmice. Mol Cell Biol 21:5605–5613, 2001

42. George M, Ayuso E, Casellas A, Costa C, Devedjian JC, Bosch F: Beta cellexpression of IGF-I leads to recovery from type 1 diabetes. J Clin Invest

109:1153–1163, 200243. Hoglund P, Mintern J, Waltzinger C, Heath W, Benoist C, Mathis D:

Initiation of autoimmune diabetes by developmentally regulated presenta-tion of islet cell antigens in the pancreatic lymph nodes. J Exp Med

189:331–339, 199944. Katz JD, Benoist C, Mathis D: T helper cell subsets in insulin-dependent

diabetes. Science 268:1185–1188, 199545. Stratmann T, Martin-Orozco N, Mallet-Designe V, Poirot L, McGavern D,

Losyev G, Dobbs CM, Oldstone MB, Yoshida K, Kikutani H, Mathis D,Benoist C, Haskins K, Teyton L: Susceptible MHC alleles, not backgroundgenes, select an autoimmune T cell reactivity. J Clin Invest 112:902–914,2003

46. Turley S, Poirot L, Hattori M, Benoist C, Mathis D: Physiological beta celldeath triggers priming of self-reactive T cells by dendritic cells in a type-1diabetes model. J Exp Med 198:1527–1537, 2003

47. Gagnerault MC, Luan JJ, Lotton C, Lepault F: Pancreatic lymph nodes are

Z. MEDAROVA AND ASSOCIATES

DIABETES, VOL. 54, JUNE 2005 1787

required for priming of beta cell reactive T cells in NOD mice. J Exp Med

196:369–377, 200248. Hugues S, Mougneau E, Ferlin W, Jeske D, Hofman P, Homann D,

Beaudoin L, Schrike C, Von Herrath M, Lehuen A, Glaichenhaus N:Tolerance to islet antigens and prevention from diabetes induced bylimited apoptosis of pancreatic beta cells. Immunity 16:169–181, 2002

49. Leiter EH, Coleman DL, Hummel KP: The influence of genetic backgroundon the expression of mutations at the diabetes locus in the mouse. III.Effect of H-2 haplotype and sex. Diabetes 30:1029–1034, 1981

50. Maedler K, Spinas GA, Lehmann R, Sergeev P, Weber M, Fontana A, KaiserN, Donath MY: Glucose induces �-cell apoptosis via upregulation of theFas receptor in human islets. Diabetes 50:1683–1690, 2001

51. Maedler K, Sergeev P, Ris F, Oberholzer J, Joller-Jemelka HI, Spinas GA,Kaiser N, Halban PA, Donath MY: Glucose-induced beta cell production ofIL-1beta contributes to glucotoxicity in human pancreatic islets. J Clin

Invest 110:851–860, 200252. Maedler K, Fontana A, Ris F, Sergeev P, Toso C, Oberholzer J, Lehmann R,

Bachmann F, Tasinato A, Spinas GA, Halban PA, Donath MY: FLIPswitches Fas-mediated glucose signaling in human pancreatic beta cellsfrom apoptosis to cell replication. Proc Natl Acad Sci U S A 99:8236–8241,2002

53. Donath MY, Halban PA: Decreased beta-cell mass in diabetes: significance,mechanisms and therapeutic implications. Diabetologia 47:581–589, 2004

54. Mathis D, Vence L, Benoist C: Beta-cell death during progression todiabetes. Nature 414:792–798, 2001

55. Bellone M, Iezzi G, Rovere P, Galati G, Ronchetti A, Protti MP, Davoust J,Rugarli C, Manfredi AA: Processing of engulfed apoptotic bodies yields Tcell epitopes. J Immunol 159:5391–5399, 1997

56. Kown MH, Strauss HW, Blankenberg FG, Berry GJ, Stafford-Cecil S, TaitJF, Goris ML, Robbins RC: In vivo imaging of acute cardiac rejection inhuman patients using 99mtechnetium labeled annexin V. Am J Transplant

1:270–277, 200157. Hofstra L, Dumont EA, Thimister PW, Heidendal GA, DeBruine AP,

Elenbaas TW, Boersma HH, van Heerde WL, Reutelingsperger CP: In vivodetection of apoptosis in an intracardiac tumor. J Am Med Assoc 285:1841–1842, 2001

58. Kemerink GJ, Boersma HH, Thimister PW, Hofstra L, Liem IH, PakbiersMT, Janssen D, Reutelingsperger CP, Heidendal GA: Biodistribution anddosimetry of 99mTc-BTAP-annexin-V in humans. Eur J Nucl Med 28:1373–1378, 2001

59. Kemerink GJ, Liem IH, Hofstra L, Boersma HH, Buijs WC, ReutelingspergerCP, Heidendal GA: Patient dosimetry of intravenously administered99mTc-annexin V. J Nucl Med 42:382–387, 2001

60. Kemerink GJ, Liu X, Kieffer D, Ceyssens S, Mortelmans L, Verbruggen AM,Steinmetz ND, Vanderheyden JL, Green AM, Verbeke K: Safety, biodistri-bution, and dosimetry of 99mTc-HYNIC-annexin V, a novel human recom-binant annexin V for human application. J Nucl Med 44:947–952, 2003

61. Post AM, Katsikis PD, Tait JF, Geaghan SM, Strauss HW, Blankenberg FG:Imaging cell death with radiolabeled annexin V in an experimental modelof rheumatoid arthritis. J Nucl Med 43:1359–1365, 2002

62. Tokita N, Hasegawa S, Maruyama K, Izumi T, Blankenberg FG, Tait JF,Strauss HW, Nishimura T: 99mTc-Hynic-annexin V imaging to evaluateinflammation and apoptosis in rats with autoimmune myocarditis. Eur

J Nucl Med Mol Imaging 30:232–238, 200363. Yang DJ, Azhdarinia A, Wu P, Yu DF, Tansey W, Kalimi SK, Kim EE,

Podoloff DA: In vivo and in vitro measurement of apoptosis in breastcancer cells using 99mTc-EC-annexin V. Cancer Biother Radiopharm

16:73–83, 200164. Mochizuki T, Kuge Y, Zhao S, Tsukamoto E, Hosokawa M, Strauss HW,

Blankenberg FG, Tait JF, Tamaki N: Detection of apoptotic tumor re-sponse in vivo after a single dose of chemotherapy with 99mTc-annexin V.J Nucl Med 44:92–97, 2003

65. Ntziachristos V, Tung CH, Bremer C, Weissleder R: Fluorescence molec-ular tomography resolves protease activity in vivo. Nat Med 8:757–760,2002

66. Moore A, Medarova Z, Potthast A, Dai G: In vivo targeting of underglyco-sylated MUC-1 tumor antigen using a multimodal imaging probe. Cancer

Res 64:1821–1827, 200467. Shapiro A, Lakey J, Ryan E, Korbutt G, Toth E, Warnock G, Kneteman N,

Rajotte R: Islet transplantation in seven patients with type 1 diabetesmellitus using a glucocorticoid-free immunosuppressive regimen. N Engl

J Med 343:230–238, 200068. Shapiro A, Ricordi C, Hering B: Edmonton’s islet success has indeed been

replicated elsewhere. Lancet 362:1242, 200369. Biarnes M, Montolio M, Nacher V, Raurell M, Soler J, Montanya E: �-Cell

death and mass in syngeneically transplanted islets exposed to short- andlong-term hyperglycemia. Diabetes 51:66–72, 2002

70. Davalli A, Scaglia L, Zangen D, Hollister J, Bonner-Weir S, Weir G:Vulnerability of islets in the immediate posttransplantation period: dy-namic changes in structure and function. Diabetes 45:1161–1167, 1996

IMAGING �-CELL DEATH

1788 DIABETES, VOL. 54, JUNE 2005