Journal of Surgical Research 143, 195–199 (2007)

Development of PTH Eluting Microspheres for the Treatmentof Hypoparathyroidism

Peter Fong, Ph.D.,*,†,1 Amit Goyal, M.D.,* Matthew Brennan, M.D.,* Jason Park,† Lawrence Moss, M.D.,*W. Mark Saltzman, Ph.D.,† and Christopher K. Breuer, M.D.*

*Department of Surgery, †Department of Biomedical Engineering, Yale University School of Medicine, New Haven, Connecticut

Submitted for publication August 4, 2005

doi:10.1016/j.jss.2006.04.009

Background. Parathyroid hormone (PTH) replace-ment has been demonstrated to be superior to conven-tional treatment with calcium supplementation andvitamin D analogs for the treatment of hypoparathy-roidism. In this investigation we evaluated the feasi-bility of using PTH microsphere encapsulation as apotential delivery system for PTH.

Materials and methods. Using the spontaneous emul-sion technique, PTH microspheres were created by en-capsulating PTH (1–34) in a copolymer of polyglycolicand polylactic acid (PLGA). Additional microsphereswere constructed by coencapsulating calmodulin withPTH (1–34) in the PLGA microspheres. Microsphere pro-duction was confirmed using electron microscopy. PTHrelease was measured in vitro using an enzyme-linkedimmunosorbent assay. The bioactivity of PTH releasedfrom the microspheres was confirmed in vivo using ahypoparathyroid rat model by measuring serum cal-cium concentrations before and 3 h after subcutaneousinjection of PTH microspheres.

Results. PTH microsphere and PTH/calmodulin mi-crospheres could be created using the spontaneousemulsion technique. Physiologically significant PTHrelease was measured in vitro for 20 days. PTH releasewas calcium sensitive and exhibited negative feed-back. This effect was augmented by coencapsulationwith calmodulin. PTH released from the microspherescaused a significant rise in serum calcium levels froman average of 6.35 (6.19–6.48 mg/dL) to 8.55 mg/dL(8.22–8.73). PTH released from the PTH/calmodulinmicrospheres resulted in an increase in serum calciumfrom a mean of 6.8 (6.7–6.9 mg/dL) to 8.1 mg/dL(7.8–8.2).

Conclusions. The PLGA microspheres can be used to

1 To whom correspondence and reprint requests should be ad-dressed at Department of Surgery, 333 Cedar Street, P.O. Box

208062, New Haven, CT 06520. E-mail: christopher.breuer@yale.edu.

195

provide calcium sensitive controlled release of biolog-ically active PTH and offer a potential mean of provid-ing biomimetic hormone replacement therapy. © 2007

Elsevier Inc. All rights reserved.

Key Words: parathyroid hormone (PTH); microspheres;drug delivery; negative feedback.

INTRODUCTION

Hypoparathyroidism is a serious metabolic disorder.The most common cause of hypoparathyroidism is iat-rogenic injury to the gland during thyroid or parathy-roid surgery. Estimates of transient and permanenthypoparathyroidism after surgery range between 6.9to 46% and 0.4 to 33%, respectively [1–5]. Unlike otherhormone insufficiency states, the conventional treat-ment of hypoparathyroidism does not involve hormonereplacement. Rather, conventional treatment for hypo-parathyroidism involves the use of calcium supplemen-tation and vitamin D analogs. parathyroid hormone(PTH) replacement therapy has been demonstrated tobe superior to conventional therapy for the treatmentof hypoparathyroidism [6, 7]. Unfortunately, daily ortwice daily subcutaneous injections of PTH are re-quired for effective treatment [7, 8].

The purpose of this study was to explore the use ofcontrolled release technology to deliver PTH for thepurpose of maintaining calcium homeostasis withoutthe need for daily subcutaneous injections. Controlledrelease technology enables the delivery of a biologicallyactive agent at specific times with defined concentra-tions by encapsulating the biologically active agentwithin a biodegradable polymer in the form of a micro-sphere. In euparathyroid individuals PTH release oc-curs in response to serum calcium levels via a nega-tive feedback mechanism; as serum calcium decreases,

PTH release increases and vice versa. Using controlled

0022-4804/07 $32.00© 2007 Elsevier Inc. All rights reserved.

196 JOURNAL OF SURGICAL RESEARCH: VOL. 143, NO. 2, DECEMBER 2007

release microsphere technology, we developed a rudi-mentary, calcium-sensitive, negative feedback PTH de-livery system by co-encapsulating PTH with calmodulinin poly(lactic-co-glycolic) acid (PLGA) microspheres andcharacterized its function both in vitro and in vivo.

MATERIALS AND METHODS

To explore the use of controlled release technology to deliver PTH,PTH (1-34) was incorporated into PLGA microspheres that werestudied in vitro and in vivo. In vitro studies were designed to dem-onstrate release of PTH from these particles, while in vivo studieswere designed to show a physiological effect thereby confirming thebioactivity of the PTH released from the microspheres.

PTH Microsphere Production and Characterization

PTH Fluorescent Labeling

The PTH was fluorescently labeled to enable detection and quan-tification of PTH released from the microspheres. PTH (1-34) wasconjugated with fluorescein isothiocyanate (FITC) by reacting 1 mgof PTH (2.47 �10�7 mol) with 50 times molar excess (1.23 �10�5 mol)of FITC for 1 h at 25°C in 300 �L of sodium borate and 200 �L ofphosphate-buffered saline (PBS). Conjugation of the FITC-NHS toPTH was confirmed by size exclusion high performance liquid chro-matography.

Microsphere Fabrication

The spontaneous emulsion technique was used to make twobatches of PLGA microspheres. One batch contained PTH conjugatedto FITC and calmodulin (PTH/Cal) and another batch contained PTHconjugated with only FITC (FITC-PTH). One milliliter of dichlo-romethane and 4 mL of trifluoroethanol were used to dissolve 200 mgof solid PLGA (50/50). To make (PTH/cal) microspheres 250 �gCalmodulin (in 100 �L PBS) and 25 ug FITC-PTH (in 100 �L solu-tion) was added to the polymer solution, giving a 10:1 by weightCaM:PTH ratio. To make FITC-PTH microspheres 25 �g of FITC-conjugated PTH was added in 100-�L solution. The protein solutionswere added together with the surfactant, then immediately pipettedinto the polymer solution. The mixture was then added slowly to 100mL of chilled 5% (in water) poly-(vinyl alcohol) solution in a 400 mLbeaker and stirred for 4 h at medium speed [5, 6] on a stir plate at25°C. After stirring, the solution was spun down in a centrifuge at4500 rpm for 5 min and the supernatant was removed and discarded.The microspheres were then resuspended in water and centrifugedagain; this wash process was repeated a total of three times. Afterthe final wash, the microspheres were resuspended in minimumamount of water, frozen at �70°C for 30 min, and lyophilized for 2days.

Microsphere Characterization

To confirm microsphere production, the lyophilized PTH micro-spheres were imaged using scanning electron microscopy. The mi-crospheres were fixed on an aluminum stub using two-sided carbontape and then coated with a gold/palladium mixture (60:40) using asputter coater. The samples were imaged via electron microscopywith the acquired imaged analyzed using NIH image (available fromzippy.nimh.nih.gov) to determine microsphere diameter. Average

particle diameter and size distributions were calculated.

In Vitro Studies of PTH Release From Microspheres

In Vitro Release of PTH From PTH Microspheres inPBS Without Calcium

Ten milligrams of the PTH-loaded microspheres were suspendedin 1 mL of PBS. The eppendorf tubes were incubated at 37°C on arotary shaker. At various times points the tubes were removed,centrifuged, and the supernatant was removed and stored at �80°Cfor later determination of protein content. After removal of thesupernatant a fresh 1 mL aliquot of additional PBS was added tothe microspheres for further incubation at 37°C. After collecting thesamples, the concentration of PTH (1-34) in the release medium wasmeasured in triplicate using a fluorimeter to detect the presence ofFITC. Release studies were carried out in triplicate.

In Vitro Release Profile of PTH/Cal and FITC-PTHMicrospheres in PBS With VariableCalcium Concentrations

To determine the release profile of PTH, 5 mg of PTH/Cal andFITC-PTH microspheres were suspended in two buffer solutionscreated using 20 mM Tris-HCl and either 1 mM CaCl2 or 2 mM CaCl2.One set was suspended in a solution of 1 mM Ca2�, and the other ina 2 mM Ca2� solution. One milliliter of buffer was added to each tube.The tubes were placed in a Lab-Line Environ Shaker at 37°C. Mea-surements of PTH release were taken by centrifuging the eppendorftubes and removing the supernatant to measure FITC presence withthe fluorimeter. The microspheres were resuspended in buffer andplaced back in the shaker. Measurements were taken at 2, 4, 6, 21,24, 48, and 72 h. At 21 h, the buffer for each set was switched to theother calcium concentration to determine if the release profile fol-lowed the change in calcium concentration. All release studies werecarried out in triplicate.

In Vivo Characterization of PTH Microspheres

Hypoparathyroid Animal Model

All animal studies were approved by the Institutional AnimalCare and Use Committee at Yale University. The male Sprague-Dawley rats (Charles Rivers Labs, Portage, MI) (175-225 g bodyweight) underwent subtotal thyroidectomy and total parathyroidec-tomy under general anesthesia with isoflourane. Rats were fed ratchow (Harlan Teklad Global 2018, 1.01% calcium and 0.65% phos-phorous) and given water ad libitum. They were housed at the YaleAnimal Resource Center in accordance with the institutional poli-cies. Serum calcium and PTH (1-84) levels were measured before thesurgery and 7 days post-operatively to confirm their hypoparathy-roid status. Serum PTH (1-84) levels were measured using rat intactPTH enzyme-linked immunosorbent assay kit (Immutopics, Inc.,San Clemente, CA). Total serum calcium concentration was mea-sured using a standard O-Cresolphthalein. Complexone colorimetricassay (Biotron Diagnostics, Hemet, CA). All measurements wereperformed three times and averaged.

Subcutaneous Implantation of Microspheres

To demonstrate the physiological effect of the PTH microspheresand thereby confirm the bioactivity of the PTH released from themicrospheres, PTH microspheres were subcutaneously injected intohypoparathyroid rats and serum calcium levels were measured. Fivehypoparathyroid rats were injected subcutaneously with 100 mg/kgof FITC-PTH containing microspheres in 1 mL of PBS with 0.5%tween 80. Rat blood was collected 1 day prior, immediately before theinjection of microspheres, and 3 h post-injection. Three hypoparathy-roid rats were injected subcutaneously with 100 mg/kg of PTH/Calmicrospheres and serum calcium was sampled every 30 min for 3 h

post-injection. All rats were sacrificed at the end of the study.

197FONG ET AL.: PTH MICROSPHERES

RESULTS

SEM Micrograph of PTH Microspheres

A scanning electron micrograph of the PTH micro-spheres is shown in Fig. 1 at 700� magnification con-firms microsphere production. The majority of the mi-crospheres were spherical with some irregular surfaces.The average size of FITC-PTH microspheres was 7 �mand that of PTH/Cal microspheres was 20 �m.

In Vitro Release of PTH Without Calcium

The controlled release profile for PTH microspheresis shown in Fig. 2. PTH was successfully encapsulatedand released continually for 20 days. The PLGA micro-spheres gave a characteristic immediate burst releaseover the first 24 h followed by a more steady, continu-ous release over approximately 3 weeks.

FIG. 1. shows an SEM image of FITC-PTH (left) and FITC/Cal (rtechnique.

Controlled Rele

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

0 5

PT

H (

pg)

FIG. 2. In vitro release profile of F

In Vitro Release Profile of PTH/Cal and FITC-PTHMicrospheres With Varying Calcium Concentrations

The effect of calcium concentration on PTH releasefrom the microspheres is shown in Fig. 3. The differencein the rate of release of PTH from microspheres incu-bated in solution A (1 mM CaCl) and solution B (2 mM

CaCl) was determined by subtracting the rate of releaseof microspheres incubated in solution A from micro-spheres incubated in solution B at 2, 4, 6, and 21 h. At21 h the buffers were reversed (solution A changed to 2mM CaCl and solution B changed to 1 Mm CaCl) and thedifference in the rate of release were again determined bysubtracting the rate of release of PTH from microspheresincubated in solution A from microspheres incubated insolution B at 24, 48, and 72 h. A positive value representsan increase in the rate of release of PTH while a negativevalue represents a decrease in the rate of release. PTH

t) microspheres made by spontaneous emulsion solvent evaporation

of PTH 1-34 in PBS

10 15 20

e (Days)

*10 mg's of microspheres

igh

ase

Tim

ITC-PTH microspheres in PBS.

198 JOURNAL OF SURGICAL RESEARCH: VOL. 143, NO. 2, DECEMBER 2007

release was found to be inversely proportional to thecalcium concentration. This effect was more pro-nounced in the PTH-F/Cal microspheres comparedwith the PTH-F microspheres.

In Vivo Evaluation of the Effect of Subcutaneous MicrosphereInjection on Serum Calcium in Hypoparathyroid Rats

After parathyroidectomy the rat serum calcium levelfell to below 7 mg/dL in all rats. The average startingserum calcium before implantation of PTH microsphereswas 6.35 mg/dL (6.19–6.48 mg/dL). Three hours post-implantation the serum calcium rose to an average of8.55 mg/dL (8.22–8.73). When PTH/Cal microsphereswere injected in the rat subcutaneously, the serum cal-cium rose from a baseline average of 6.8 (6.7–6.9 mg/dL)to an average of 8.1 mg/dL (7.8–8.2) by 3 h.

DISCUSSION

The ideal treatment for hypoparathyroidism is PTHreplacement therapy. The clinical efficacy of PTH as atreatment for hypoparathyroidism has been established[6, 7]. In a short term randomized cross over trial, pa-tients with chronic hypoparathyroidism treated with

Differential Rate of Release

-30.00

-20.00

-10.00

0.00

10.00

20.00

30.00

40.00

2h 4h 6h 21h 24h 48h 72h

Time

Rat

e o

f re

leas

e o

f P

TH

(n

g/h

r)

PTH/Cal

FITC-PTH

Buffer Sw itched

FIG. 3. The Effect of the calcium concentration on the rate ofPTH release from PTH-F and PTH/Cal microspheres. The differencein the rate of release of PTH from microspheres incubated in solutionA (1 mM CaCl) and solution B (2 mM CaCl) was determined bysubtracting the rate of release of microspheres incubated in solutionA from the rate of release of microspheres incubated in solution B at2, 4, 6, and 21 h. At 21 h the buffers were reversed (solution Achanged to 2 mM CaCl and solution B changed to 1 Mm CaCl) and thedifference in the rate of release were again determined by subtract-ing the rate of release of PTH from microspheres incubated in solu-tion A from microspheres incubated in solution B at 24, 48, and 72 h.A positive value represents an increase in the rate of release of PTHwhile a negative value represents a decrease in the rate of release.PTH release was found to be inversely proportional to the calciumconcentration. This effect was more pronounced in the PTHF/Calmicrospheres compared to the PTH-F microspheres.

daily injections of human PTH (1–34) maintained serum

calcium in the normal range with decreased urine cal-cium excretion compared to patients treated with con-ventional treatment (oral calcitrol and calcium supple-mentation) [7]. A follow-up study demonstrated thattwice-daily PTH regimen provided effective treatmentof hypoparathyroidism and reduced the variation inserum calcium levels at a lower total daily dose [8]. Thelower total daily dose limits the toxicity associatedwith PTH administration while maintaining tightercontrol over calcium homeostasis.

An optimal method of PTH delivery has not beenestablished. PTH has a short plasma half-life, is un-stable in the gastrointestinal tract, and has low bio-availability because of its large molecular weight andhigh aqueous solubility [9–11]. These properties makedelivery of PTH problematic. A PLGA microsphere drugdelivery system has been investigated for controlled re-lease of a variety of other similar proteins and was foundto be efficacious [12–14]. We hypothesized that contin-uous release of PTH from PLGA microspheres could beused to maintain physiological concentrations of PTHbut with a lower total daily dose when compared withsubcutaneous injections. This could potentially im-prove efficacy while minimizing any side effects relatedto PTH administration. In this study we demonstratedthat PTH can reliably be encapsulated into PLGA mi-crospheres and sustained release of physiological con-centrations of PTH can be achieved over several weeksin vitro. Furthermore, we demonstrated that the re-leased PTH maintains its biological activity and canexert its effect on calcium homeostasis.

Physiological PTH release is calcium sensitive via anegative feedback mechanism. In our initial studieswe noted that PTH microspheres demonstrated dif-ferential release of PTH based on the concentrationof calcium. At higher concentrations of calcium (2 mM

calcium chloride) the release rate of PTH decreased,while at lower concentrations (1 mM calcium chloride)the rate increased. Moreover, this was a reversiblephenomenon: i.e., when the buffer of the microsphereswas switched, the microspheres released less PTH inhigher concentrations of calcium chloride. This effectwas enhanced when calmodulin was co-encapsulatedinto the PTH microspheres. The exact mechanism ofthis differential release in the presence/absence ofcalcium is incompletely understood. Perhaps the con-firmation change that occurs in calmodulin in thepresence/absence of calcium leads to changes in inter-action between PTH and calmodulin within the poly-mer that results in differential release rates of PTH.While studies are ongoing to elucidate the mechanismof release of PTH from these microspheres, it is clearthat the PLGA microspheres demonstrate a rudimen-tary form of calcium sensitive negative feedback thatmimics the calcium sensitive negative feedback exhib-

ited by the parathyroid glands.

199FONG ET AL.: PTH MICROSPHERES

Maintenance of calcium homeostasis in the hypopar-athyroid patient is a complex and challenging problem.The mechanisms by which PTH controls serum cal-cium through an interplay of bone resorption, urinaryexcretion and GI absorption results in a complex tem-poral response of serum calcium to changes in PTHsecretion. As our understanding of this complex phe-nomena unfolds our ability to construct a biomimeticdrug delivery systems with improved efficacy and lim-ited side effects will also improve.

ACKNOWLEDGMENTS

This work was supported in part by an Ohse Grant (Breuer).

REFERENCES1. Thomusch O, Machens A, Sekulla C, Ukkat J, Brauckhoff M,

Dralle H. The impact of surgical technique on postoperativehypoparathyroidism in bilateral thyroid surgery: A multivari-ate analysis of 5846 consecutive patients. Surgery 2003;133:180.

2. Falk SA, Birken EA, Baran DT. Temporary postthyroidectomyhypocalcemia. Arch Otolaryngol Head Neck Surg 1998;114:168.

3. Percival RC, Hargreaves AW, Kanis JA. The mechanism ofhypocalcemia following thyroidectomy. Acta Endocrinol 1985;109:220.

4. See ACH, Soo KC. Hypocalcemia following thyroidectomy forthyrotoxicosis. Br J Surg 1997;84:95.

5. Burney RE, Jones KR, Coon JW, et al. Assessment of patientoutcomes after operation for primary hyperparathyroidism.

Surgery 1996;120:1013.

6. Strogmann W, Bohrn E, Woloszczuk W. First experiences in thesubstitution treatment of hypoparathyroidism with synthetichuman parathyroid hormone. Monatsschr Kinderheilkd 1990;138:141.

7. Winer KK, Yanovski JA, Cutler GB. Synthetic human parathy-roid hormone 1-34 vs calcitrol and calcium in the treatment ofhypoparathyroidism: Results of a short-term randomized cross-over trial. JAMA 1996;276:631.

8. Winer KK, Yanovski JA, Sarani B, Cutler GB. A randomized,cross-over trial of once-daily versus twice-daily parathyroid hor-mone 1-34 in treatment of hypoparathyroidism. J Clin Endocri-nol Metab 1998;83:3480.

9. Leone-Bay A, Sato M, Paton D, et al. Oral delivery of biologi-cally active parathyroid hormone. Pharm Res 2001;18:961.

10. Patton JS, Trinchero P, Platz RM. Bioavailability of pulmonarydelivered peptides and proteins: A-interferon, calcitonins, andparathyroid hormones. J Control Rel 1994;28:79.

11. Condrons B, Vanderbist F, Ucakar B, Preat V, Vanbever R.Impact of formulation and methods of pulmonary delivery onabsorption of parathyroid hormone (1-34) from rat lungs.J Pharm Sci 2004;93:1241.

12. Perets A, Baruch Y, Weisbuch F, Shoshany G, Neufeld G,Cohen S. Enhancing the vascularization of three-dimensionalporous alginate scaffolds by incorporating controlled releasebasic fibroblast growth factor microspheres. J Biomed MaterRes 2003;65A:489.

13. Eiselt P, Kim BS, Chacko B, et al. Development of technologiesaiding large-tissue engineering. Biotechnol Prog 1998;14:134.

14. Saltzman WM, Mak M, Mahoney MH, Duenas ET, Cleland JL.Intracranial delivery of recombinant nerve growth factor: Re-lease kinetics and protein distribution for three delivery sys-

tems. Pharm Res 1999;16:232.