 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*
Division of Rheumatology, Department of Medicine, Duke University School
of Medicine, Durham, North Carolina
Division of Nephrology, Department of Medicine, Duke University School of
Medicine, Durham, North Carolina
Center for In Vivo Microscopy, Department of Radiology, and
Department of Biochemistry, Duke University School of
Medicine, Durham, North Carolina
||
Mountain View Pharmaceuticals, Inc., Menlo Park, California.
Correspondence to Dr. Michael S. Hershfield, Box 3049, Duke University Medical Center, Durham, NC 27710. Phone: 919-684-4184; Fax: 919-684-4168; E-mail: msh{at}biochem.duke.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Blockage of urate synthesis by xanthine oxidase inhibition or promotion of uric acid excretion is usually effective for the treatment of urate-induced disorders (5,10,11). Indeed, no new therapies for controlling hyperuricemia have been introduced in the United States since allopurinol was introduced in the 1960s. In the past four decades, the prevalence of gout has been increasing (12) and an often-aggressive form of gout has been recognized as a complication of cyclosporine therapy among recipients of kidney and other organ transplants (13,14,15). Hyperuricemia has proved difficult to control in some patients (5,16,17). Multiple factors may contribute to this problem, including hypersensitivity to allopurinol, renal insufficiency, coexisting chronic diseases in elderly patients, azathioprine therapy in organ transplant recipients, and marked uric acid overproduction in patients with inherited metabolic disorders. "Replacement" therapy with a long-acting form of uricase is a potential alternative for such patients.
Daily infusion of uricase derived from Aspergillus flavus has been used to control acute hyperuricemia during chemotherapy of malignancies (18,19). However, use of the fungal enzyme for chronic therapy of gout has been limited by rapid clearance, which requires daily infusion, and by potential immunogenicity (19,20,21). Very low solubility at physiologic pH is an additional barrier to therapy with unmodified mammalian urate oxidases (22). Covalent linkage of poly(ethylene glycol) (PEG) to exposed amino groups has been demonstrated to enhance solubility, prolong circulating life, and reduce the immunogenicity of several proteins in animals (23). PEG-modified bovine adenosine deaminase, injected once or twice per week, has been well tolerated and effective as chronic replacement therapy for inherited adenosine deaminase deficiency (24,25).
We have produced PEG-modified, recombinant, mammalian urate oxidases and tested their efficacy in a strain of mice homozygous for a targeted disruption of the Uox gene (Uox-/-). These mice have been demonstrated to develop obstructive nephropathy, akin to that associated with tumor lysis in human patients; most died by 4 wk of age, unless they were treated with allopurinol (26). Renal function in these mice was not studied. Here we report that untreated Uox-/- mice have a profound defect in the ability to concentrate urine, resulting in nephrogenic diabetes insipidus (DI). This condition can be largely prevented by treatment with PEG-uricase. These findings provide a "proof of principle" for the clinical use of PEG-uricase; they also suggest factors that underlie the evolutionary loss or retention of uricase among species.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Biochemical and Physiologic Studies
Uricase activity was determined spectrophotometrically, at 23 to 25°C,
by monitoring the decrease in absorbance at 292 nm in a reaction mixture
containing 0.1 M sodium borate (pH 8.6) and 0.1 mM uric acid (1 unit was
defined as a decrease of 1 µmol of uric acid/min)
(27). Commercial kits were
used to determine the concentrations of uric acid (Sigma kit 685-10) and
creatinine (Sigma kit 557) in serum and urine. Protein was determined by the
Lowry or bicinchoninic acid (Pierce, Rockford, IL) method, with bovine serum
albumin as the standard. Some serum samples were analyzed for urea nitrogen,
creatinine, calcium, phosphorus, albumin, and urate levels by AniLytics, Inc.
(Gaithersburg, MD).
Blood was obtained from the retro-orbital plexus using methoxy-flurane anesthesia. In some studies of mice treated with uricase, ex vivo degradation of uric acid was prevented by rapid mixing of blood with an equal volume of 2 N perchloric acid on ice; after centrifugation (2 min at 4°C at 11,000 x g), the supernatant was neutralized with 3.0 M KOH/1.0 M KHCO3. The serum urate concentration was calculated from the concentration in the neutralized extract, assuming a hematocrit value of 0.44. Spontaneously voided urine was collected from a clean surface. Mice were placed in metabolic cages for quantitation of water intake and urine output. Urine osmolality was determined with a vapor pressure osmometer (Westcor Instruments, Logan, UT).
PEG-Uricase
Standard methods were used to amplify, clone, and sequence cDNA. Reverse
transcription-PCR primers for pig and baboon liver urate oxidases were based
on published sequences
(1,28).
A pig-baboon chimeric (PBC) uricase cDNA, consisting of pig codons 1 to 225
joined in-frame to baboon codons 226 to 304, was constructed. The PBC cDNA
encodes 30 lysines, one more than either "parental" uricase cDNA,
because of the presence of lysine codons replacing Asn-103 in the baboon cDNA
and Arg-291 in the pig cDNA. Other cDNA based on the pig coding sequence but
possessing a Lys-291 codon were subsequently constructed; proteins derived
from the latter are referred to collectively as PigK uricase. Uricase cDNA
were overexpressed in Escherichia coli BL21(DE3)pLysS, using pET
vectors (Novagen, Madison, WI), or in other expression systems not described
here. Purified preparations of the recombinant pig, PBC, and PigK uricases had
specific activities of 3 to 7 U/mg. Similarly prepared recombinant baboon
uricase was four- to fivefold less active; baboon uricase purified from liver
in also less active than pig liver uricase
(29). These recombinant
uricases were soluble and stable in carbonate buffer (pH 10.2) but, like the
mammalian urate oxidases purified from tissues
(22), were sparingly soluble
at pH 7.4.
PBC uricase purified to >95% homogeneity (Kelly SJ, Hershfield MS, unpublished procedure) was provided to Mountain View Pharmaceuticals, Inc. (Menlo Park, CA), where it was modified by covalent attachment of monomethoxy-PEG (i.e., PEGylated), using a modification of a method described by Sherman et al. (30). Approximately 60 kD of PEG was coupled to each uricase subunit, determined as described (31). The resulting PEG-uricase had specific activity ranging from 2.3 to 5.6 U/mg; it was soluble and stable for at least 6 mo at 4°C in phosphate-buffered saline (PBS) (pH 7.4). A PEGylated recombinant PigK uricase with similar properties was prepared by Drs. J. R. Hartman and S. Mendelovitz of Bio-Technology General Ltd. (Rehovot, Israel).
Uricase Replacement Therapy
Serum uricase activity was observed to peak 20 to 24 h after
intraperitoneal injection of Uox-/- mice with PEG-uricase; the
circulating half-life was approximately 78 h. In experiments involving
repeated intraperitoneal administration, the 24-h postinjection serum uricase
activity was used as an index of the rate of enzyme clearance (as in analogous
studies of PEG-modified E. coli purine nucleoside phosphorylase)
(32). The concentration of
uric acid in urine provided a noninvasively measured index of the capacity to
catabolize urate. Use of this index reduced the need for blood sampling and
the associated risk of death resulting from anesthesia and blood volume
depletion.
Enzyme-Linked Immunosorbent Assay for Anti-Uricase Antibody
Microtiter plates (Immulon 2HB; Dynatech, Chantilly, VA) were treated
overnight at 4°C with 1 µg/well of purified recombinant uricase in 0.1
M sodium carbonate (pH 10.2) or with buffer alone. After blocking with 2% goat
serum (Life Technologies BRL, Gaithersburg, MD) in PBS, the plate was washed
with PBS containing 0.1% Tween 20 (PBST). Aliquots (0.1 ml) of mouse serum
(diluted in PBS with 2% goat serum) were added to adjacent rows of
uricase-coated and antigen-free wells. After 1 h at 37°C, plates were
washed with PBST and then incubated for 45 min at 37°C with a goat
anti-mouse Ig-peroxidase conjugate (Calbiochem, San Diego, CA). After washing
with PBST, 0.1 ml/well of o-phenylenediamine peroxidase substrate
solution (Calbiochem) was added. After 25 min at 25°C, 0.1 ml/well of 1 N
HCl was added, and absorbance at 490 nm was measured with a Vmax kinetic
microplate reader (Molecular Devices, Sunnyvale, CA). The
A490 values for uricase-free wells were subtracted to
determine bound antigen-specific mouse Ig. Results are expressed in
enzyme-linked immunosorbent assay units (A490/10 µl of
serum), as defined by Chaffee et al.
(33).
Magnetic Resonance Microscopy
Mice were euthanized by barbiturate overdose. The blood supply to the left
kidney was ligated immediately, to maintain blood in the organ [as a negative
contrast agent for magnetic resonance (MR) imaging]. The body was perfused
with 0.9% saline solution and then with 10% neutral buffered
formalin-glutaraldehyde solution. The perfused (right) kidney was used for
histologic analyses. The nonperfused (left) kidney was placed in a cylindrical
container and immersed in Fomblin (perfluoro-polyether; Ausimont, Morristown,
NJ) to limit magnetic susceptibility variations at the tissue surface. MR
imaging was performed immediately, using a custom-designed 11-mm sole-noid
radiofrequency coil. MR images were acquired at 9.4 T with a Bruker CSI system
(Fremont, CA) equipped with actively shielded gradients
(34). A spin-echo pulse
sequence was used (repetition time, 500 ms; echo time, 10 ms; number of
excitations, 2). Spatial encoding was accomplished using three-dimensional
Fourier transform, which allowed the simultaneous imaging of 128 contiguous
planes (each 39 µm thick) through the specimen. The 10-mm field of view was
reconstructed on a 256 x 256 matrix, leading to a voxel size of (39
µm)3. Images were analyzed on a Silicon Graphics workstation
(Reality Engine 2; SGI, Mountain View, CA), using VoxelView software (Vital
Images, Fairfield, IA).
Histopathologic Analyses
Formalin/glutaraldehyde-fixed kidneys were embedded in paraffin, sectioned
(3 µm) in the axial plane (corresponding to the axial MR images), and
stained with hematoxylin and eosin. Histologic evaluation was performed
without knowledge of genotype or the MR microscopy results.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
Except for moderate azotemia, we did not observe signs of renal insufficiency in adult Uox-/- mice. Thus, when evaluated at 50 to 60 d and 9 to 13 mo of age, the serum urea nitrogen concentrations for Uox-/- mice were 56 ± 20 and 43 ± 10 mg/dl (mean ± SD), respectively, or 1.5- to 2-fold higher than for heterozygotes. BP and serum creatinine, calcium, phosphorus, and albumin levels were normal (data not shown), and ascites was never observed.
Treatment with PEG-Uricase
Although it is stable and active at high pH, unmodified (native)
recombinant uricase was not effective in vivo in Uox-/-
mice. Thus, serum uricase activity remained undetectable (<0.005 U/ml) at 4
and 24 h after intraperitoneal injection of 1 unit of native uricase (diluted
20-fold, to 0.23 mg/ml, in PBS, pH 7.4, just before injection, to maintain
solubility). Serum urate levels for these mice (n = 12) decreased
from 10.2 ± 2.5 mg/dl before treatment to 6.3 ± 1.8 mg/dl at 4 h
after injection but returned to baseline values by 24 h. The urinary uric acid
concentration transiently decreased from 164 ± 19 mg/dl before
injection to 139 and 146 mg/dl at 5 and 9 h after injection, respectively, but
no decrease was observed at 3, 19, or 30 h. With weekly administration,
anti-uricase antibody appeared in all 12 mice after the second injection and
levels increased after the third and fourth injections
(Figure 3, A to C).
|
Compared with the native enzyme, PEG-uricase exhibited greater bioavailability and was far more effective in decreasing urate levels, including results after repeated injections. In one study, 4-mo-old Uox-/- mice (n = 10) received intraperitoneal injections of 0.34 U of PEG-uricase every 5 d. Each injection produced a marked reduction in the urinary level of uric acid; reductions were greatest after 24 h but were evident throughout the period between injections (Figure 4). In urine obtained 5 d after the 11th injection, the uric acid/creatinine ratio was 1.7 ± 2.5, compared with 10.6 ± 1.5 before treatment. Sera obtained 9 d after the 11th injection exhibited detectable uricase activity (0.053 ± 0.025 U/ml) and lacked significant anti-uricase antibody (Figure 3D); the mean serum urate concentration was 1.3 ± 2.1 mg/dl, compared with 7.3 ± 0.8 mg/dl before treatment. After the 10th PEG-uricase dose, we evaluated renal concentrating ability. During a 12-h period of water deprivation, urine osmolality for the treated mice increased by 623 ± 364 mosmol/kg, compared with an increase of 319 ± 236 mosmol/kg before treatment (P < 0.05). The residual concentrating defect may have been attributable to irreversible structural renal damage (see below).
|
PEG-uricase therapy initiated early in life was effective in preventing the development of renal injury. We determined this by studying two groups of six Uox-/- mice, one of which was given a series of 10 or 11 intraperitoneal injections of PEG-uricase and the other of which was given saline solution, administered at 4- to 10-d intervals (mean, 6.7 d) between the first 10 d and the 10th week of life (0.095 U/dose until 6 wk of age and then 0.19 U/dose). After 10 injections, none of the PEG-uricase-treated mice exhibited detectable anti-uricase antibody (Figure 3E). When measured at 6 to 9 wk of age, the mean urinary uric acid/creatinine ratio for the six PEG-uricase-treated Uox-/- mice was 1.2 to 2, which was much lower than that for the untreated Uox-/- mice (Figure 1). At 10 wk, the treated mice demonstrated significantly greater renal concentrating ability than did the untreated Uox-/- control mice, with nearly normal fluid balance. Thus, after 12 h of water deprivation, the untreated control animals exhibited a mean urine osmolality of approximately 750 mosmol/kg; the treated mice exhibited a mean basal urine osmolality of approximately 1850 mosmol/kg, which increased to 2700 mosmol/kg (Figure 5A). With free access to water, the untreated mice ingested 4.4-fold more water and produced 6.3-fold more urine than did normal (Uox+/+) mice, compared with 1.7-fold more water and 1.4-fold more urine for the treated mice (Figure 5B).
As assessed by MR microscopy (Figure 6), the renal architecture was largely preserved in the PEG-uricase-treated mice, in contrast to the scarred, irregularly shaped, hydronephrotic kidneys of untreated Uox-/- mice. Histologic sections of kidneys from untreated mice exhibited cortical foci of fibrosis and tubular atrophy, with apparent crowding of glomeruli (Figure 7). In both the cortex and medulla, collecting ducts were markedly dilated, possibly because of elevated tubular flow. In contrast, kidneys from treated mice exhibited no gross irregularities. Atrophy of the inner medulla was absent, tubules were packed tightly, and glomeruli exhibited normal distribution.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The DI of uricase-deficient mice resembles the persistent concentrating defect in Gunn rats, which lack hepatic glucuronyl transferase; the defect was reversible if unconjugated bilirubinemia was corrected in young animals, before permanent structural renal damage occurred (37). The DI is also related mechanistically to the transient DI that may follow renal tubular injury caused by ischemia, toxins, or obstruction (e.g., postobstructive diuresis). This transient DI might be a manifestation of injury or of the recovery of tubular epithelium. It can result in severe volume depletion, but it may also serve a protective function, i.e., to reduce the concentration of an offending toxin or obstructing agent. The persistent DI of uricase-deficient mice may be lethal in nursing animals but essential for the survival of adults.
Structural kidney damage and DI are evident in Uox-/- mice by 2 wk of age, when mortality (26) and uric acid excretion (Figure 1) are highest. Adult Uox-/- mice with severe DI exhibited little evidence of progressive renal insufficiency, and many exhibited normal lifespans. However, before our recognition of their fivefold greater water requirement, some died as a result of dehydration after exhausting water supplies adequate for normal mice. The high mortality rate before weaning is probably attributable to the inadequacy of maternal milk as the only source of fluid replacement. Death resulting from dehydration was similarly attributed to polyuria in nursing mice in which the gene for the NaK2Cl cotransporter had been disrupted (38). However, after weaning, a sixfold elevated urine volume, compensated by adequate water intake, may limit further deposition of uric acid in renal tubules, preventing progressive renal failure. Such a protective role for nephrogenic DI is consistent with the finding that high tubular fluid flow attributable to inherited pituitary DI prevented acute uric acid nephropathy in a rat model in which the serum urate concentration was increased to 20 mg/dl by the infusion of both urate and oxonic acid (an inhibitor of uricase) (39). Maintenance of high urine volume is used clinically to treat uric acid renal stones and acute uric acid nephropathy in human patients (5,6).
Relevance to Evolution
Discussions of uricase evolution have often speculated on advantages
derived from loss of the enzyme, such as increased longevity resulting from
the radical-scavenging ability of urate
(3). A less anthropocentric
view might consider the benefit of uricase to other species for which renal
water conservation has been an essential adaptation to seasonal drought or
arid climates. The avoidance of uric acid precipitation in maximally
concentrated urine would strongly favor the retention of uricase; its loss
might lead to lethal nephropathy, as in uricase-knockout mice. In contrast,
ancestors of New World monkeys, hominoid apes, and humans, in which
Uox gene mutations became fixed
(2), may have lived in rain
forests. With abundant water, maintenance of relatively dilute urine may have
been better tolerated and favored as an adaptation to environmental toxins, as
well as to the loss of uricase. Other protective mechanisms might have been in
place or selected subsequently, e.g., efficient proximal tubular
reabsorption to limit the access of uric acid to collecting ducts, where
maximal concentration and acidification occur, or shorter renal papillae to
limit distal water reabsorption. The reduced ability of the neonatal kidney to
concentrate and acidify urine may protect human infants from uric acid
nephropathy at a time when urate clearance is considerably higher than that in
adults (40).
PEG-Uricase as a Potential Therapy for Refractory Gout
Fungal and pig liver uricases, PEGylated in various ways, have been studied
in uricase-expressing animals
(41,42,43).
Where reduced immunogenicity has been achieved, it has been at levels of PEG
coupling that cause substantial loss of catalytic activity. PEGylated
Candida and Arthrobacter uricases were demonstrated to
decrease serum urate concentrations in a few patients with malignancies, but
neither preparation was subsequently produced for clinical trials
(44,45).
We have taken several steps aimed at producing a PEGuricase more suitable for long-term therapy. (1) We chose porcine uricase, which has >85% identity with the deduced amino acid sequence of human uricase, rather than a microbial enzyme with <40% identity or baboon uricase, which has slightly greater sequence similarity but is much less active. We considered "reactivating" human uricase by eliminating both known nonsense mutations but concluded that it would be difficult to identify and "correct" any missense mutations acquired during evolution. (2) Addition of potential PEG attachment sites (lysine residues) enhanced the ability of PEGylation to reduce the immunogenicity of a bacterial enzyme in mice (32). This presumably resulted from better epitope masking. We applied this strategy to pig uricase by changing Arg-291 to lysine. The resulting PBC and related enzymes are fully active and possess one to four more lysines than other cloned mammalian uricases (mouse, rat, rabbit, and baboon uricases). (3) We optimized PEGylation to achieve a PEG-uricase that retains the activity of the native enzyme but is substantially more soluble at physiologic pH and is much less immunogenic in mice.
Because of their extraordinary excretion of uric acid and the resulting renal damage, Uox—/— mice provide an excellent model for testing efficacy. Injections of PEG-uricase at 5- to 7-d intervals controlled hyperuricemia and markedly reduced urate excretion in uricase-knockout mice and, when initiated before weaning, prevented the development of nephropathy. This treatment elicited minimal antibody response. In contrast to PEG-uricase, unmodified recombinant mammalian uricase was ineffective in decreasing uric acid levels and was highly immunogenic. These results suggest that this form of PEG-uricase could be effective in treating both acute and chronic hyperuricemia and uricosuria. In addition to use in the acute setting of chemotherapy for malignancies, PEG-uricase could provide a much-needed therapy for controlling chronic hyperuricemia in patients with inherited metabolic disorders and in patients with severe gout who are allergic or unresponsive to conventional therapy.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. A. Fenton and M. A. Knepper Mouse Models and the Urinary Concentrating Mechanism in the New Millennium Physiol Rev, October 1, 2007; 87(4): 1083 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Safra, G. V. Ling, R. H. Schaible, and D. L. Bannasch Exclusion of Urate Oxidase as a Candidate Gene for Hyperuricosuria in the Dalmatian Dog Using an Interbreed Backcross J. Hered., November 1, 2005; 96(7): 750 - 754. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bachmann, K. Mutig, J. Bates, P. Welker, B. Geist, V. Gross, F. C. Luft, N. Alenina, M. Bader, B. J. Thiele, et al. Renal effects of Tamm-Horsfall protein (uromodulin) deficiency in mice Am J Physiol Renal Physiol, March 1, 2005; 288(3): F559 - F567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tinschert, N. Ruf, I. Bernascone, K. Sacherer, G. Lamorte, H.-H. Neumayer, P. Nurnberg, F. C. Luft, and L. Rampoldi Functional consequences of a novel uromodulin mutation in a family with familial juvenile hyperuricaemic nephropathy Nephrol. Dial. Transplant., December 1, 2004; 19(12): 3150 - 3154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Maronpot, R. C. Sills, and G. A. Johnson Applications of Magnetic Resonance Microscopy Toxicol Pathol, February 1, 2004; 32(2_suppl): 42 - 48. [Abstract] [PDF]
|
|
|
|
 |
|
 G. A. Johnson, G. P. Cofer, S. L. Gewalt, and L. W. Hedlund Morphologic Phenotyping with MR Microscopy: The Visible Mouse Radiology, March 1, 2002; 222(3): 789 - 793. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
)
and three to 12 Uox-/- mice (
). Data for six Uox-/-
mice that had been treated with weekly intraperitoneal injections of
pig-baboon chimeric (PBC) poly(ethylene glycol)-uricase (PEG-uricase) (see
legend to 
).
