Horm Metab Res 2015; 47(01): 24-30
DOI: 10.1055/s-0034-1394375
Endocrine Care
© Georg Thieme Verlag KG Stuttgart · New York

Oxygen Supply by Photosynthesis to an Implantable Islet Cell Device

Y. Evron1, B. Zimermann1, B. Ludwig1, 2, U. Barkai1, C. K. Colton3, G. C. Weir4, B. Arieli1, S. Maimon1, N. Shalev1, K. Yavriyants1, T. Goldman1, Z. Gendler1, L. Eizen1, P. Vardi5, K. Bloch5, A. Barthel2, 6, S. R. Bornstein2, A. Rotem1
  • 1Beta-O2 Technologies, Rosh Ha’ain, Afek Park, Israel
  • 2University Hospital Carl Gustav Carus, Department of Medicine III, Dresden, Germany
  • 3Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, USA
  • 4Section of islet Transplantation and Cell Biology, Joslin Diabetes Center, Research Division, One Joslin Place, Boston, USA
  • 5Diabetes and Obesity Research Laboratory, Felsenstein Medical Research Center, Sackler Faculty of Medicine, Tel-Aviv University, Beilinson Campus, Petah Tikva, Israel
  • 6Endokrinologikum Ruhr, Bochum, Germany
Further Information

Correspondence

A. Rotem, PhD
Beta-O2 Technologies Ltd.
11 Ha’amal St
POB 11793
Rosh Ha’ain
Afek Park
480990
Israel   
Phone: +972/3/918 0700   
Fax: +972/3/918 0701   

Publication History

received 04 August 2014

accepted 23 September 2014

Publication Date:
03 November 2014 (eFirst)

 

Abstract

Transplantation of islet cells is an effective treatment for type 1 diabetes with critically labile metabolic control. However, during islet isolation, blood supply is disrupted, and the transport of nutrients/metabolites to and from the islet cells occurs entirely by diffusion. Adequate oxygen supply is essential for function/survival of islet cells and is the limiting factor for graft integrity. Recently, we developed an immunoisolated chamber system for transplantation of human islets without immunosuppression. This system depended on daily oxygen supply. To provide independence from this external source, we incorporated a novel approach based on photosynthetically-generated oxygen. The chamber system was packed sandwich-like with a slab of immobilized photosynthetically active microorganisms (Synechococcus lividus) on top of a flat light source (LEDs, red light at 660 nm, intensity of 8 μE/m2/s). Islet cells immobilized in an alginate slab (500–1 000 islet equivalents/cm2) were mounted on the photosynthetic slab separated by a gas permeable silicone rubber-Teflon membrane, and the complete module was sealed with a microporous polytetrafluorethylene (Teflon) membrane (pore size: 0.4 μm) to protect the contents from the host immune cells. Upon illumination, oxygen produced by photosynthesis diffused via the silicone Teflon membrane into the islet compartment. Oxygen production from implanted encapsulated microorganisms was stable for 1 month. After implantation of the device into diabetic rats, normoglycemia was achieved for 1 week. Upon retrieval of the device, blood glucose levels returned to the diabetic state. Our results demonstrate that an implanted photosynthetic bioreactor can supply oxygen to transplanted islets and thus maintain islet viability/functionality.


#

Introduction

Blood glucose control in patients with type 1 diabetes (T1D) usually requires multiple insulin injections and frequent blood glucose level tests per day. Despite this, extended hyperglycemia periods and life-threatening hypoglycemia episodes are common. For example, approximately 6% of deaths in T1D patients are associated with hypoglycemia [1] [2].

A promising therapeutic approach in T1D patients with labile metabolic control and frequent hypoglycemia is allogeneic islet transplantation. However, several challenges limit the widespread application of this approach. First, donor organs are rare and often more than one donor organ per recipient is needed [3]. Second, the life-long immunosuppressive treatment required to sustain the graft in these patients is associated with potentially severe side effects [4] [5]. Third, donor islets isolated from pancreatic tissue by enzymatic digestion are cut from functional vascular system and, hence, no direct blood supply is available to nourish the islets until revascularization occurs. Normally, revascularization is a slow process (reviewed by Pepper et al. 2013 [6]), which is prolonged further with the use of immunosuppressive drugs [7]. Furthermore, even when fully developed, the new vascular network does not resolve oxygen deficiencies [8]. Consequently, transplanted islets initially rely solely on diffusion for transport of nutrients, hormones, and metabolites. In particular, the oxygen supply becomes a limiting factor for graft function and survival. Protection of islets from the host immune system by encapsulation in hydrogels further aggravates this problem due to increased diffusion distances [8] [9] [10] [11] [12] [13] [14] [15] [16]. Moreover, cell necrosis resulting from hypoxia may increase the immune response to the graft [17]. The field of oxygen supply to encapsulated therapeutic cells has recently been reviewed by Colton [18].

Recently, we developed an immunoisolated chamber system that allows transplantation of islets without immunosuppression [19] [20]. Islets in this device were oxygenated using daily oxygen replenishing. Various technical approaches to improve oxygen supply to islets within implantable devices have been investigated theoretically and experimentally. These include increasing oxygen permeability in the encapsulating matrix by adding perflurocarbons [16] [21], adding cross-linked hemoglobin [22], transfection with the cytoglobin gene [23], and reducing islet size by dispersing and aggregating the cells into small pseudo-islets [16]. Continuous exogenous oxygen generation by in situ water electrolysis [13] [24] and transient oxygen generation by chemical decomposition [25] [26] were also tested. Bloch et al. [27] demonstrated that islet functionality can be maintained by photosynthetically-produced oxygen from algae (Chlorella sorokiniana) co-cultured with the islets. However, separating the oxygen-producing microorganisms from the oxygen-consuming islets is required to protect the algae from the immune system.

In this work, we report a novel macro-device, in which oxygen is supplied to the islets by a slab of immobilized photosynthetically-active thermophilic cyanobacteria (Synechococcus Lividus) illuminated by a flat double-sided planar light source. This specific thermophilic strain was chosen as it can survive and function at elevated temperatures [28], and also can perform efficient photosynthesis even at very low light intensities. The microorganisms and the islet modules are encapsulated in a hydrogel and are separated by a silicone-Teflon membrane that has high permeability for oxygen. The 3-layer configuration combines a robust photosynthetic oxygen generator that provides a constant oxygen flux to ensure sustained islet graft survival and function.


#

Materials and Methods

Photosynthetic microorganisms

Synechococcus lividus was obtained from the Pasteur culture collection (PCC 6717). It is a thermophilic cyanobacterium isolated from Yellowstone hot springs with an optimal culture temperature of 52°C (i. e., well above human body temperature). Therefore, to allow adaptation of the organisms to high osmotic pressure and low temperature, the cyanobacteria were cultured for 10 generations in a medium with high osmotic pressure (270 mOsm/l) at 40°C. The cyanobacteria were cultured in 70 ml BG-11 media [29] concentrated 5-fold (osmotic pressure of 197 mOsm) at 40±1°C in transparent 250 ml flasks on a horizontal shaker at 110 rpm with continuous exposure to daytime light (30 W fluorescence tube, Osram, Israel). The light intensity was 40±2 μE/m2/s.


#

Islet isolation

Pancreata were obtained from 9- to 10-week old male Lewis rats weighing 260–280 g. The organs were digested with collagenase following a standard procedure with slight modifications, as previously described [19]. Briefly, each pancreas was infused with 10 ml enzymatic digestive blend containing 15 PZ units of collagenase NB8 (Serva, Heidelberg, Germany) and 1 mg/ml bovine DNAse (Sigma, Rechovot, Israel) dissolved in HBSS solution (Bet-HaEmek, Israel) for 14 min. Islets were purified by centrifugation (20 min at 1 750×g and 6°C) over a discontinuous Histopaque gradient (1.119/1.100/1.077 g/ml in RPMI), washed twice and cultured in complete CR medium (1:1, CMRL:RPMI medium, Bet HaEmek, Israel), and supplemented with 10% fetal bovine serum (Bet-HaEmek, Israel) for 1 week prior to being integrated into the implantable devices. Determination of islet size and number and calculation of volume and number of islet equivalents (IEQ) were performed as previously described [19].


#

Implantable device

The implantable bioartificial pancreas device included 2 modules connected by an electric wire. A power supply (PS) module provided energy to the light source of a bioreactor module. The bioreactor was a square-shaped clear polyethylene terephthalate container measuring 31×31×7 mm ([Fig. 1]).

Zoom Image
Fig. 1 Design and construction of the implantable device. a Schematic view of the device. Inset: cross section of the device and detailed explanation. b Actual device withdrawn from an animal after implantation for 7 days (device dimensions: 31×31×7 mm).

The PS module contained 2 batteries (3.6 V each) suitable for operation at various temperatures (TL-5101, Size 1/2 AA, Tadiran, Holon, Israel) connected in series to obtain a total of 7.2 V, and a magnetic switch (2 mm in diameter and 5 mm length; KSK-1C90, MEDER Electronics, West Wareham, MA) to control the light source. For water resistance, the PS module was coated with 50±10 μm thick parylene (Simtal, Rehovot, Israel), and was then molded in a layer of silicone rubber (5 mm; 4 970, Nusil, Carpenteria, CA, USA). The bioreactor module included 3 major subunits: The islet slab, the photosynthetic slab, and a planar light source (PLS) (supplementary Fig. 1A online).

For the islet slab, 2 000–4 000 IEQ were collected by brief centrifugation (1 min at 140 g). The pellet was gently hand-mixed with 200 μl of 1.8% (w/v) of high guluronic acid alginate (G=0.68, Pronova UPMVG, Novamatrix; Sandvika, Norway), placed in a sterile Teflon mold with a thickness of 500 μm and cross-linked by applying flat sintered glass (Pyrex, UK) saturated with 70 mM strontium chloride dissolved in RPMI medium (total of 270 mOsm) on top of the alginate-islet mixture for 10 min and the resulting slab was then placed into the islet compartment (final islet density: 500–1 000 IEQ/cm2). The islet slab was shielded from mechanical stress by a stainless steel grid placed on top of the compartment.

The photosynthetic slab was formed by mixing the cyanobacteria (harvested from the log-phase by centrifugation for 3 min at 8 000 g) with liquefied agarose (Sigma Type I-B A-0576, Sigma, Rechovot, Israel) at 60°C for a final concentration of 1.5% (w/v) and 300 μg/ml chlorophyll. Exposure of the cyanobacteria to this temperature did not damage their viability or their oxygen production rate (OPR) (data not shown). A 250 μl volume of the hot mixture was applied to the relevant module and solidified. The process provided a 300 μm thick immobilized cyanobacteria layer, slightly thicker than the islet slab. The photosynthetic slab was separated from the islet slab by a 25 μm gas-permeable silicone rubber Teflon membrane (Silon, BMS, Allentown, PA, USA).

The PLS included a double-sided planar light guide (LG) designed to provide a uniform spatial light distribution to both surfaces. The LG was made of polycarbonate with a high refractive index of 1.67 (Eyaloptica, Israel) and dimensions of 20×20×1.4 mm. Surfaces on both sides were patterned with drilled hemispherical holes (300 μm diameter) and polished (Optic, Israel). To prevent loss of photons, the first 500 μm in the proximity of the LEDs (supplementary Fig. 1A online) and the sides of the LG were coated with a 1 μm gold reflective layer (Adionim, Kohav Yair, Israel). The design and optimization of the PLS was performed by software (LightTools 6.2, Germany, Munich, GmbH). LEDs were glued within the LG by UV-cured adhesive (140 Dymax Corp., Burlington, MA, USA). Light input was obtained from 3 light emitting diodes (ELC-660-23-1, Epigap, Berline, GmbH) emitting at a wavelength of 660±3 nm (red light). Light intensity was calibrated with a 2.2 KΩ resistor in series to obtain a photosynthetically active radiation intensity of 8.8 μE/m2/s (3.3 mA) at the surface of the LG. Light uniformity along the light guide surface was analyzed with a CCD camera (Fx-33, Spirieon, North Logan, UT, USA) and the light intensity was analyzed by a light meter (LI-KOR, Cambridge, UK). To provide water resistance, the LEDs and LG were coated with parylene as described above.

A 25 μm, 0.4 μm pore diameter polytetrafluoroethylene (PTFE) hydrophylized membrane (Biopore, Millipore, Billerica, MA, USA) was used to separate the islet slab from the interstitial body fluid in order to protect the islets from the host cellular immune system. The PTFE-Biopore membrane was then fixed onto the device using a Viton O-Ring (McMaster Carr; Aurora, OH, USA) and sealed to the plastic housing with medical silicone glue (MED 2000, Polytek Easton, PA, USA). The fully assembled device was washed in complete CR (CMRL:RPMI, 1:1+10% bovine serum) medium at 37°C with agitation for 2 h before implantation.

In order to ensure that the electric wire connecting the PS and the bioreactor was resistant to extensive banding stress due to animal movements, a special electric wire (35N LT DRF 25Ag, Fort Wayne Metals, IN, USA) was used, allowing maximum mechanical flexibility. The device was sterilized by ethylene oxide (Mediplast, Yavne, Israel) prior to implantation.


#

Animals, induction of diabetes, and implantation of the device

All experiments were approved by the Israeli National Ethical Committee (IL-05-05-012). Lewis rats (260–280 g) were purchased from Harlan (Rehovot, Israel), and diabetes was induced by a single intravenous (IV) injection of 85 mg/kg body weight of streptozotocin (STZ; Sigma, Rehovot, Israel) as previously described [19]. Animals had unrestricted access to food at all times and were considered diabetic when spontaneous blood glucose concentrations exceeded 350 mg/dl for at least 4 consecutive days. For collection of blood samples, the animals were sedated, the tail tip was slit slightly, and glucose levels were measured with a commercial glucometer (Accu-Chek sensor, Roche Diagnostics, Berlin, GmbH). To prepare the diabetic animals for device implantation under nonstressing, close-to-normal blood glucose conditions, 1.5 capsules (Linplant, LinShin, Toronto, Canada) of a sustained release insulin were inserted subcutaneously (SC) into the diabetic animals. Animals were considered ready for implantation of the device when spontaneous blood glucose concentrations were below 250 mg/dl for 3 or more consecutive days.

For device implantation, the animals were anesthetized by intraperitoneal injection of 90 mg/kg ketamine and 10 mg/kg xylazine followed by isofluran inhalation. A 4 cm incision was made on the back of the animal and the PS was implanted subcutaneously (SC). The bioreactor was implanted into a subcutaneous pocket at the upper lateral abdominal quadrant and the subunits connected by wire via tunneling. The sustained release insulin capsules were removed upon the device implantation, leaving the device as the only source for glucose surveillance and insulin secretion. The efficacy of the device in achieving glycemic control was followed for 7 days after implantation by measuring nonfasting blood glucose concentrations twice daily.


#

Measurements

Chlorophyll concentrations

A 1 ml volume of cyanobacteria suspension was incubated with 3 ml of methanol for 5 min at 70°C. The mixture was centrifuged at 5 000 g for 5 min and the optical density of the supernatant was tested at 665 nm. The chlorophyll concentration was calculated according to Schumann et al. [30].


#
#

Oxygen production rate (OPR)

A custom-designed conical oxygen measurement chamber was placed on top of a bioreactor containing only the light source and the immobilized cyanobacteria without the islet chamber. The conical chamber was filled with BG-11 medium concentrated 5-fold to a final volume of 620 μl ([Fig. 2a]). The conical chamber was equipped with a 5-mm diameter magnetic stirrer and a Clark-type oxygen electrode of 500 μm diameter, connected to a picoampermeter (Cat No. OX 500-9893, Unisense, Arhaus, Denmark). The oxygen measurement chamber was placed within a perspex box at an air temperature of 37±1°C using a temperature control unit (Eurotherm 808; Eurotherm Worthing, UK). The stirring speed was increased until the OPR was stable (approximately 70 rpm), assuring minimal boundary layers around the cyanobacteria slab and the oxygen electrode. The electrode was calibrated against BG-11 medium equilibrated with gas containing no oxygen or ambient oxygen concentrations. Oxygen concentrations were reported as oxygen partial pressure p, in units of mm Hg, related to the molar oxygen concentration c by the relation: c=αp, where α is the Bunsen solubility coefficient, 1.34×10−9 mol/(cm3×mm Hg) for a temperature of 37°C. As a result of oxygen production, the oxygen concentration in the medium increased linearly with time. OPR was calculated from the slope obtained by linear regression.

Zoom Image
Fig. 2 Photosynthetic OPR of immobilized cyanobacteria. a Measurement system. b Bioreactor containing light source and immobilized cyanobacteria (S. Lividus) was implanted in rats for various times and the OPR was tested using light intensity of 7 μE/cm2/s. n=2.

#

Oxygen steady state

The conical chamber and oxygen electrode described above were placed on a bioreactor containing light source, immobilized cyanobacteria and immobilized islets ([Fig. 3a]). The system was adjusted to 37°C as described above, illuminated at various light intensities between 3.9 and 20 μE/s/m2, and the oxygen concentration was measured.

Zoom Image
Fig. 3 Balancing microbial oxygen production and islet oxygen consumption. a Schematic view of the measurement system. b Raw data: 1 200 IEQ/cm2 were coupled to immobilized algae separated by 25 μm gas-permeable membrane and illuminated at various intensities. Inset: 660 IEQ. Blue dots: light turned on. Red dots: light turned off.

#

Oxygen concentration across the islet slab

Bioreactor containing islets and cyanobacteria slabs was placed in a petri dish with RPMI medium. The space above the slab was purged with a gas stream (40 mm Hg O2, 40 mm Hg CO2, and 680 mm Hg N2; supplementary Fig. 2A online), simulating SC gas concentrations. An oxygen electrode attached to a micromanipulator, was inserted into the islet-containing slab and advanced at 100 μm increments from the distal side of the islet slab downwards toward the gas permeable membrane. At each step, the oxygen electrode readings reached a near steady-state level before moving to the next step.


#
#

Results

To achieve a uniform light distribution and avoid variability in light intensity that would cause non-uniform oxygen production and insufficient oxygen supply to islets in some regions, LEDs emitting at 660 nm were coupled to a flat wave guide (WG, supplementary Fig. 1A online). Light at 660 nm is optimal for photosynthesis in cyanobacteria and provides energy-efficient system for oxygen production. The light produced by the LED distributed within the WG and was emitted perpendicularly via the dimples on the WG surface. The resulting uniform light across the WG was demonstrated using a CCD camera (supplementary Fig. 1B online).

The cyanobacteria were immobilized within an isolated module that separated it from body fluids while allowing gas exchange (e. g., oxygen, CO2, and water vapor). To evaluate the photosynthetic function of the device over time, illuminated bioreactors without islets were implanted SC for various times in rats and the OPR was measured using the system described in [Fig. 2a]. After one month, the OPR remained constant at a rate of >1.0 nmol/min/cm2 ([Fig. 2b]), suggesting a stable oxygen production for a period of 1 month.

Balancing between islet oxygen consumption rate (OCR) and photosynthetic oxygen production rate (OPR)

Producing oxygen at levels exceeding oxygen consumption is an inefficient use of electrical power and may also damage the cyanobacteria and the islet cells. Therefore, the rate of oxygen production by photosynthesis should be equivalent to oxygen consumption by the graft. To find this equilibrium in a culture system, a chamber containing immobilized cyanobacteria slab at a density of 300 μg chlorophyll/ml was coupled to 1 200 IEQ. Variable light intensities were applied on the oxygen producing module and the oxygen concentration above the islet chamber was measured ([Fig. 3a]). At relatively high intensities of 14 and 20 μE/s/m2 the oxygen production exceeded consumption ([Fig. 3b]). At an illumination intensity of approximately 7 μE/s/m2, the slope of net oxygen concentration was nearly zero, indicating that the rates of oxygen production and consumption were equal. Therefore, we assumed that this intensity was sufficient to maintain 1 200 IEQs that consume oxygen at 3 pmol/IEQ/min. Light intensity of 3.9 μE/s/m2 was sufficient to supply oxygen to 660 islets at a steady state ([Fig. 3b], insert).


#

Validation of sufficient islet oxygen supply

The islets within the device are randomly distributed within the alginate slab (i. e., islets differ in their proximity to the oxygen source) ([Fig. 1a]). To maintain 150 μm diameter islets in a fully functional state, the minimal oxygen concentration on the islet surface should be approximately 50 mm Hg [10]. To ensure that all islets in the slab are exposed to oxygen concentrations of at least 50 mm Hg, we measured the distribution of oxygen across the islet slab. An alginate slab containing 1 200 immobilized islets was placed on top of the cyanobacteria slab and was separated by a 25 μm of gas-permeable membrane. An oxygen electrode was inserted in various depths within the slab at increments of 100 μm (supplementary Fig. 2A online). To simulate SC gas composition, the medium surrounding the islets was continuously purged with oxygen and CO2, both at a concentration of 40 mm Hg. At an illumination intensity of 6.5 μE/s/m2, oxygen concentration furthest from the oxygen source was approximately 30 mm Hg. This concentration is not sufficient to maintain islet functionality across the entire slab thickness (Red line, supplementary Fig. 2B online), indicating that an increased OPR is required. Based on these data, in subsequent experiments, light intensity was set to 8 μE/s/m2 in order to ensure sufficient oxygen supply to all islets.


#

In vivo implantation

To evaluate the ability of the system to serve as a bioartificial islet organ, full devices containing isogeneic rat islets were implanted into STZ-induced diabetic rats. After implantation, blood glucose concentrations rapidly dropped to normoglycemic levels, and upon explantation (7 days later), blood glucose concentrations returned to the diabetic range ([Fig. 4]).

Zoom Image
Fig. 4 Glycemic control after implantation of the device in diabetic rats. One week prior to device implantation, diabetic rats were brought to near normoglycemia by SC implantation of slow release insulin capsule (LP in). Devices containing between 3 000 and 4 000 IEQ were implanted and the slow release insulin capsule was removed (LP out). Devices were illuminated with light intensity of 8 μE/m2/s and the nonfasting blood glucose was measured. Mean±SD (n=4 animals).

#
#

Discussion

Aspects of methods for supplying exogenous oxygen to encapsulated islets in implantable devices using water electrolysis or chemical decomposition reactions have been previously studied [13] [24] [25] [26]. Previously, we demonstrated that direct injection of oxygen to a macro-chamber device is feasible and efficient [19] [20]. Here, we tested a new concept regarding oxygen supply using photosynthetic cyanobacteria as oxygen generator.

Cyanobacteria phylum was chosen as the photosynthetic oxygen source because it can adapt and survive in extreme environmental conditions, and because it can use low-intensity light efficiently [31]. In order to achieve efficient supply of oxygen to the islets, a flat geometry was chosen for the 3 components: light source, cyanobacteria, and islet modules. This design resulted in a homogenous oxygen production in close vicinity with the islet slab. To separate the photosynthetic microbial cells from the host immune system, the cyanobacteria were completely sealed from body fluids by a silicon membrane, allowing only the transfer of gases and water vapor. To avoid transfer of water vapor due to differences in osmotic pressure between the cyanobacteria compartment and body fluids, similar osmotic conditions were established in all compartments.

Every photosynthetic organism has its own set of pigments suitable for optimal harvest of light energy in its habitat. The cyanobacteria S. Lividus contains phycocyanine pigment, suitable for light harvesting at a wavelength band of 550–640 nm, which is the typical light in its habitat. However, light emitted from LEDs has a narrow wavelength band (660 nm) that does not include this bandwidth, and therefore we illuminated the light harvesting complexes directly for maximal efficiency. The ability to spread the light evenly into a uniform sheet ensured maximum illumination efficiency.

In vitro experiments using illuminated, photosynthetically-active microorganisms co-cultured with islets in liquid phase demonstrated the production of adequate amounts of oxygen to support functional islets [27]. However, in the implantable devices, the cyanobacteria were encapsulated in alginate and were totally sealed from body fluids to protect them from the host immune system. Therefore, the organisms were casted into a thin film, thereby creating solid photosynthetic mat, separated from the islet compartment by a thin 25 μm gas-permeable silicone membrane, allowing for gas exchange with the islet compartment.

We explored achieving a balance between light intensity, photosynthetic OPR, and islet OCR. The photosynthetic slab had to produce oxygen at a rate of approximately 4.5 nmol/cm2/min to supply the needs of 1 200 IEQ/cm2. The observed equilibrium between oxygen production (cyanobacteria) and oxygen consumers (islets) demonstrated the ability of the system to produce enough oxygen. The minimum light intensity required for oxygen production above compensation (the light intensity required for switching from oxygen consumption to production) was 0.5 μE/m2/s (data not shown). On the other hand, high light intensity may cause the production of excess oxygen leading to the formation of reactive oxygen species, which could damage the islets [32] as well as the cyanobacteria [33]. Therefore, optimization of the operating conditions was required. We found that a 300 μm thick slab with cyanobacteria at a concentration of 300 μg chlorophyll/ml, illuminated with a light intensity of 7–8 μE/m2/s was sufficient to support 1 200 IEQ. Islet viability and functionality were determined before and after device implantation. Implantation of the devices in diabetic rats cured diabetes for the duration of implantation period (1 week). The OCR of the devices after their explantation showed no loss of activity, indicating that the islets had sufficient oxygen supply.

We have shown that photosynthetic cyanobacteria casted into a solid mat can be implanted into mammalians for oxygen production. The oxygen production was nearly constant over a period of 1 month, capable of supporting functional islets at a density of 1 000 IEQ/cm2. Experiments evaluating oxygen production for longer periods of time are warranted. Notably, to support sufficient insulin production for a 70 kg diabetic patient approximately 5×105 islets are needed. This is translated to a device with a surface area of 450 cm2, which is impractical for implantation in humans. To achieve a smaller device for implantation, the islet density needs to be increased to >4 000 IEQ/cm2. As more islets are packed in the same volume, the oxygen consumption increases sharply and the gradient across a slab should increase as well [34]. To compensate for the increased islet density, more oxygen has to be produced. However, using our solid state, flat cyanobacteria-based bioreactor, we reached the maximal oxygen production from these microorganisms at the current illumination conditions and geometry. Therefore, in order to supply oxygen to a bioartificial pancreas with dimensions that are practical for implantation in humans, the photosynthesis approach should be greatly improved.


#
#

Conflict of Interest

YE, BZ, BL, UB, BA, SM, NS, KY, TG, ZG, LE, and AR are beta-O2 employees. CKC, GCW, PV, and KB are consultants to beta-O2 and have financial interest in the company.

Supporting Information


Correspondence

A. Rotem, PhD
Beta-O2 Technologies Ltd.
11 Ha’amal St
POB 11793
Rosh Ha’ain
Afek Park
480990
Israel   
Phone: +972/3/918 0700   
Fax: +972/3/918 0701   


Zoom Image
Fig. 1 Design and construction of the implantable device. a Schematic view of the device. Inset: cross section of the device and detailed explanation. b Actual device withdrawn from an animal after implantation for 7 days (device dimensions: 31×31×7 mm).
Zoom Image
Fig. 2 Photosynthetic OPR of immobilized cyanobacteria. a Measurement system. b Bioreactor containing light source and immobilized cyanobacteria (S. Lividus) was implanted in rats for various times and the OPR was tested using light intensity of 7 μE/cm2/s. n=2.
Zoom Image
Fig. 3 Balancing microbial oxygen production and islet oxygen consumption. a Schematic view of the measurement system. b Raw data: 1 200 IEQ/cm2 were coupled to immobilized algae separated by 25 μm gas-permeable membrane and illuminated at various intensities. Inset: 660 IEQ. Blue dots: light turned on. Red dots: light turned off.
Zoom Image
Fig. 4 Glycemic control after implantation of the device in diabetic rats. One week prior to device implantation, diabetic rats were brought to near normoglycemia by SC implantation of slow release insulin capsule (LP in). Devices containing between 3 000 and 4 000 IEQ were implanted and the slow release insulin capsule was removed (LP out). Devices were illuminated with light intensity of 8 μE/m2/s and the nonfasting blood glucose was measured. Mean±SD (n=4 animals).