Proteasome lid bridges mitochondrial stress with Cdc53/Cullin1 NEDDylation status.

Cycles of Cdc53/Cullin1 rubylation (a.k.a NEDDylation) protect ubiquitin-E3 SCF (Skp1-Cullin1-F-box protein) complexes from self-destruction and play an important role in mediating the ubiquitination of key protein substrates involved in cell cycle progression, development, and survival. Cul1 rubylation is balanced by the COP9 signalosome (CSN), a multi-subunit derubylase that shows 1:1 paralogy to the 26S proteasome lid. The turnover of SCF substrates and their relevance to various diseases is well studied, yet, the extent by which environmental perturbations influence Cul1 rubylation/derubylation cycles per se is still unclear. In this study, we show that the level of cellular oxidation serves as a molecular switch, determining Cullin1 rubylation/derubylation ratio. We describe a mutant of the proteasome lid subunit, Rpn11 that exhibits accumulated levels of Cullin1-Rub1 conjugates, a characteristic phenotype of csn mutants. By dissecting between distinct phenotypes of rpn11 mutants, proteasome and mitochondria dysfunction, we were able to recognize the high reactive oxygen species (ROS) production during the transition of cells into mitochondrial respiration, as a checkpoint of Cullin1 rubylation in a reversible manner. Thus, the study adds the rubylation cascade to the list of cellular pathways regulated by redox homeostasis.

Keepers at the final gates: regulatory complexes and gating of the proteasome channel.

The proteolytic active sites of the 26S proteasome are sequestered within the central chamber of its 20S catalytic core particle. Access to this chamber is through a narrow channel defined by the outer alpha subunits. Free proteasome 20S core particles are found in an autoinhibited state in which the N-termini of neighboring alpha subunits are anchored by an intricate lattice of interactions blocking access to the channel. Entry of substrates into proteasomes can be enhanced by attachment of activators or regulatory particles. An important part of this activation is channel gating; regulatory particles rearrange the blocking residues to form an open pore and promote substrate entry into the proteolytic chamber. Interestingly, some substrates can open the entrance themselves and thus facilitate their own destruction. In this review, we will discuss the mechanisms proposed for channel gating and the interactions required to maintain stable closed and open conformations.

Is ultrasound more accurate than axial computed tomography for determination of maximal abdominal aortic aneurysm diameter?

OBJECTIVE(S): Clinical assessment of maximal abdominal aortic aneurysm (AAA) diameter assumes clinical equivalency between ultrasound (US) and axial computed tomography (CT). Three-dimensional (3D) CT reconstruction allows for the assessment of AAA in the orthogonal plane and avoids oblique cuts due to AAA angulation. This study was undertaken to compare maximal AAA diameter by US, axial CT, and orthogonal CT, and to assess the effect that AAA angulation has on each measurement. METHODS: Maximal AAA diameter by US (US(max)), axial CT (axial(max)), and orthogonal CT (orthogonal(max)) along with aortic angulation and minor axis diameters were measured prospectively. Spiral CT data was processed by Medical Media Systems (West Lebanon, NH) to produce computerized axial CT and reformatted orthogonal CT images. The US technologists were blinded to all CT results and vice versa. RESULTS: Thirty-eight patients were analyzed. Mean axial(max) (58.0 mm) was significantly larger (P<0.05) than US(max) (53.9 mm) or orthogonal(max) (54.7 mm). The difference between US(max) and orthogonal(max) (0.8 mm) was insignificant (P>0.05). When aortic angulation was <==25 degrees, axial(max) (55.3 mm), US(max) (54.3 mm), and orthogonal(max) (54.1 mm) were similar (P>0.05); however, when aortic angulation was >25 degrees, axial(max) (60.1 mm) was significantly larger (P<0.001) than US(max) (53.8 mm) and orthogonal(max) (55.0 mm). The limits of agreement (LOA) between axial(max) and both US(max) and orthogonal(max) was poor and exceeded clinical acceptability (+/-5 mm). The variation between US(max) and orthogonal(max) was minimal with an acceptable LOA of -2.7 to 4.5 mm. CONCLUSION: Compared to axial CT, US is a better approximation of true perpendicular AAA diameter as determined by orthogonal CT. When aortic angulation is greater than 25 degrees axial CT becomes unreliable. However, US measurements are not affected by angulation and agree strongly with orthogonal CT measurements.

Challenges of hemodialysis access for high risk patients: Impact of mesenteric vein bioprosthetic graft.

OBJECTIVE: The purpose of this study is to compare in a prospective fashion the performance of a new bioprosthesis, the mesenteric vein bioprosthesis (MVB), in patients who have had multiple failed ePTFE grafts. Performance measures include primary patency rates, assisted-primary patency rates, secondary patency rates, complications, and the number of interventions required to maintain graft patency. STUDY: From October 1999 to February 2002, 276 hemodialysis access grafts were implanted in a multicenter study. Of those grafts, 74 were placed in patients with a prior history of 3 failed prosthetic grafts (mean = 3.5 grafts, range = 3-6 grafts). Fifty-nine grafts were constructed with MVB, and 15 grafts with ePTFE as a concomitant control. Mean follow-up was 11.5 months. In the MVB group, 79.7% were African-Americans, 61% were females, and 23.7% were hypercoagulable. Of the ePTFE group, 86.7% were African-Americans, 46.7% were female, and 13.2% were hypercoagulable. Results : Per Kaplan-Meier curves, the primary patency rate of the MVB group at 12 months was 33% vs the ePTFE group of 18% (p=0.120); the assisted-primary patency rates at 12 months were 45% MVB vs 18% ePTFE (p=0.011). The secondary patency rates at 12 and 24 months for the MVB group were 67% and 59%, respectively, vs 45% and 15% for the ePTFE group (p=0.006). During the follow-up time period, 80% of the ePTFE grafts were abandoned compared to 34% of the MVB group. Infection and thrombosis rates in the MVB group were lower than the ePTFE group. The infection rate for the MVB group requiring intervention was 0.07 events/graft year (gt/y) compared to 0.30 events/gt-y for ePTFE (p=0.04). A thrombosis rate of 0.69 events/gt-y occurred in the MVB group whereas 2.50 events/gt-y presented in the ePTFE group (p<0.01). CONCLUSION: In this study, high-risk patients (defined as those having multiple failed prosthetic grafts for hemodialysis) in whom the MVB conduit for hemoaccess was implanted, showed significant improvement in assisted-primary and secondary patency rates compared to the ePTFE cohort. The MVB group, however, did not have a statistically better primary patency rate compared to the ePTFE group. The MVB patient also had fewer thrombotic and infectious events and an overall reduction in the number of interventions while maintaining a permanent access site. This new bioprosthesis should be the conduit of choice in the complex group of patients as it offers assisted-primary and secondary patency rates similar to those commonly experienced by patients without a history of multiple graft failures.

Regulating the 26S proteasome.

Despite the fact that the composition of proteasomes purified from different species is almost identical, and the basic components of the proteasome are remarkably conserved among all eukaryotes, there are quite a few additional proteins that show up in certain purifications or in certain screens. There is increasing evidence that the proteasome is in fact a dynamic structure forming multiple interactions with transiently associated subunits and cellular factors that are necessary for functions such as cellular localization, presentation of substrates, substrate-specific interactions, or generation of varied products. Harnessing the eukaryotic proteasome to its defined regulatory roles has been achieved by a number of means: (a) increasing the complexity of the proteasome by gene duplication, and differentiation of members within each gene family (namely the CP and RPT subunits); (b) addition of regulatory particles, complexes, and factors that influence both what enters and what exits the proteasome; and (c) signal-dependent alterations in subunit composition (for example, the CP beta to beta i exchange). It is not be surprising that the proteasome plays diverse roles, and that its specific functions can be fine-tuned depending on biological context or need.

Subunit interaction maps for the regulatory particle of the 26S proteasome and the COP9 signalosome.

The 26S proteasome plays a major role in eukaryotic protein breakdown, especially for ubiquitin-tagged proteins. Substrate specificity is conferred by the regulatory particle (RP), which can dissociate into stable lid and base subcomplexes. To help define the molecular organization of the RP, we tested all possible paired interactions among subunits from Saccharomyces cerevisiae by yeast two-hybrid analysis. Within the base, a Rpt4/5/3/6 interaction cluster was evident. Within the lid, a structural cluster formed around Rpn5/11/9/8. Interactions were detected among synonymous subunits (Csn4/5/7/6) from the evolutionarily related COP9 signalosome (CSN) from Arabidopsis, implying a similar quaternary arrangement. No paired interactions were detected between lid, base or core particle subcomplexes, suggesting that stable contacts between them require prior assembly. Mutational analysis defined the ATPase, coiled-coil, PCI and MPN domains as important for RP assembly. A single residue in the vWA domain of Rpn10 is essential for amino acid analog resistance, for degrading a ubiquitin fusion degradation substrate and for stabilizing lid-base association. Comprehensive subunit interaction maps for the 26S proteasome and CSN support the ancestral relationship of these two complexes.

Multicenter evaluation of a polyurethaneurea vascular access graft as compared with the expanded polytetrafluoroethylene vascular access graft in hemodialysis applications.

OBJECTIVES: The purpose of this study was to compare in a randomized, prospective, and controlled study, the performance of a multilayered, self-sealing polyurethane vascular access graft (PVAG) and expanded polytetrafluoroethylene (ePTFE) vascular access grafts in hemodialysis applications. Performance measures included graft survival, complications, time to early cannulation, and hemostasis times after cannulation. STUDY DESIGN: A total of 142 patients were randomized equally to receive one of the two grafts after meeting all eligibility requirements. All patients were followed up prospectively to 12 months or to the end of secondary patency. Specifically, this study documented the performance of the PVAG and ePTFE grafts by determining the patencies and complications for both grafts. RESULTS: Patient characteristics between the two groups were similar with respect to risk factors and demographic characteristics (P >.05). Life-table patencies from the date of first dialysis were primary patency: PVAG 55% versus ePTFE 47% (6 months) and PVAG 44% versus ePTFE 36% (12 months) and secondary patency: PVAG 87% versus ePTFE 90% (6 months) and PVAG 78% versus ePTFE 80% (12 months). None of these differences were significant (P >.05). Both primary and secondary patencies were also not significantly different when the date of implantation was the starting point. Adverse events and complications were similar for the two groups, except the PVAG group had a higher incidence of technical complications manifested by graft kinking when compared with the control cohort (P <.05). Additionally, there was no significant difference in complication rates between these two groups with regard to infection and bleeding. When the time to hemostasis after cannulation was compared at 5minutes or less, there were more PVAG cannulation sites that achieved hemostasis compared with ePTFE sites, and this difference was significant (P <.0001). When time to first dialysis access was compared between the two grafts, 53.9% of all PVAG grafts were cannulated before 9 days versus none with the ePTFE grafts (P <.001). However, long-term graft survival was not significantly different when PVAG patients were stratified into early (< 9 days) and the late access (9 >/= days) groups (P =.29). CONCLUSIONS: The PVAG graft allows for early access without compromising long-term performance. Both PVAG and standard ePTFE grafts have similar long-term outcomes, despite early access with the PVAG vascular access grafts.

The substrate translocation channel of the proteasome.

The core particle (CP) of the yeast proteasome is composed of four heptameric rings of subunits arranged in a hollow, barrel-like structure. We have found that the CP is autoinhibited by the N-terminal tails of the outer (alpha) ring subunits. Crystallographic analysis showed that deletion of the tail of the alpha3 subunit opens a channel into the proteolytically active interior chamber of the CP, thus derepressing peptide hydrolysis. In the latent state of the particle, the tails prevent substrate entry by imposing topological closure on the CP. Inhibition by the alpha subunit tails is relieved upon binding of the regulatory particle to the CP to form the proteasome holoenzyme. Opening of the CP channel by assembly of the holoenzyme is regulated by the ATPase domain of Rpt2, one of 17 subunits in the RP. Thus, open-channel mutations in CP subunits suppress the closed-channel phenotype of an rpt2 mutant. These results identify a specific mechanism for allosteric regulation of the CP by the RP.

A gated channel into the proteasome core particle.

The core particle (CP) of the yeast proteasome is composed of four heptameric rings of subunits arranged in a hollow, barrel-like structure. We report that the CP is autoinhibited by the N-terminal tails of the outer (alpha) ring subunits. Crystallographic analysis showed that deletion of the tail of the alpha 3-subunit opens a channel into the proteolytically active interior chamber of the CP, thus derepressing peptide hydrolysis. In the latent state of the particle, the tails prevent substrate entry by imposing topological closure on the CP. Inhibition by the alpha-subunit tails is relieved upon binding of the regulatory particle to the CP to form the proteasome holoenzyme.

Getting in and out of the proteasome.

By far the best understood role of the proteasome is to remove ubiquitin-conjugated proteins from eukaryotric cells by hydrolysing them into small peptides of varying lengths. These include both misfolded/abnormal proteins, as well as 'normal' proteins that need to be rapidly removed for regulatory purposes. However, the proteasome is also present in numerous prokaryotic organisms, while ubiquitin, and the ubiquitin conjugating system, are not. The eukaryotic proteasome has been adapted to degrading proteins in a ubiquitin-dependent fashion by the addition of regulatory factors that assemble in different layers onto the proteolytic core of the proteasome, and by increasing the diversity of the core subunits as well. In addition to hydrolysing ubiquitinated proteins into amino acids, the proteasome can also proteolyse selected non-ubiquitinated proteins, process proteins, and possibly refold misfolded proteins. This review will focus on the different proteasome functions, and how these are used in the multiple regulatory roles the proteasome plays in eukaryotic cells.

Functional analysis of the proteasome regulatory particle.

We have developed S. cerevisiae as a model system for mechanistic studies of the 26S proteasome. The subunits of the yeast 19S complex, or regulatory particle (RP), have been defined, and are closely related to those of mammalian proteasomes. The multiubiquitin chain binding subunit (S5a/Mcb1/Rpn10) was found, surprisingly, to be nonessential for the degradation of a variety of ubiquitin-protein conjugates in vivo. Biochemical studies of proteasomes from deltarpn10 mutants revealed the existence of two structural subassemblies within the RP, the lid and the base. The lid and the base are both composed of 8 subunits. By electron microscopy, the base and the lid correspond to the proximal and distal masses of the RP, respectively. The base is sufficient to activate the 20S core particle for degradation of peptides, but the lid is required for ubiquitin-dependent degradation. The lid subunits share sequence motifs with components of the COP9/signalosome complex, suggesting that these functionally diverse particles have a common evolutionary ancestry. Analysis of equivalent point mutations in the six ATPases of the base indicate that they have well-differentiated functions. In particular, mutations in one ATPase gene, RPT2, result in an unexpected defect in peptide hydrolysis by the core particle. One interpretation of this result is that Rpt2 participates in gating of the channel through which substrates enter the core particle.

Rheolytic thrombectomy in the treatment of acute limb-threatening ischemia: immediate results and six-month follow-up of the multicenter AngioJet registry. Possis Peripheral AngioJet Study AngioJet Investigators.

We tested the efficacy of rheolytic thrombectomy in treating 21 patients (mean age 68+/-12 years; 66% male) and 22 vessels (limbs) who presented to the hospital within 2 weeks of the development of limb-threatening ischemia. Fifty-two percent had contraindications to use thrombolytics, and 57% had severe comorbidities. All of the vessels were occluded with thrombus on the initial angiogram. Procedural success was achieved in 20 limbs (91%). Three patients expired in the hospital, and one expired at follow-up due to nonvascular causes. Acute limb salvage was achieved in 18 of 19 limbs (95%) in the 18 survivors, and 6-month limb salvage was achieved in 16 of 18 limbs (89%) in the 17 survivors. Rheolytic thrombectomy is effective in restoring immediate blood flow in acute limb-threatening ischemia, especially in high-risk surgical patients or patients with contraindications to thrombolytic therapy.

A subcomplex of the proteasome regulatory particle required for ubiquitin-conjugate degradation and related to the COP9-signalosome and eIF3.

The proteasome consists of a 20S proteolytic core particle (CP) and a 19S regulatory particle (RP), which selects ubiquitinated substrates for translocation into the CP. An eight-subunit subcomplex of the RP, the lid, can be dissociated from proteasomes prepared from a deletion mutant for Rpn10, an RP subunit. A second subcomplex, the base, contains all six proteasomal ATPases and links the RP to the CP. The base is sufficient to activate the CP for degradation of peptides or a nonubiquitinated protein, whereas the lid is required for ubiquitin-dependent degradation. By electron microscopy, the base and the lid correspond to the proximal and distal masses of the RP, respectively. The lid subunits share sequence motifs with components of the COP9/signalosome complex and eIF3, suggesting that these functionally diverse particles have a common evolutionary ancestry.

Active site mutants in the six regulatory particle ATPases reveal multiple roles for ATP in the proteasome.

A family of ATPases resides within the regulatory particle of the proteasome. These proteins (Rpt1-Rpt6) have been proposed to mediate substrate unfolding, which may be required for translocation of substrates through the channel that leads from the regulatory particle into the proteolytic core particle. To analyze the role of ATP hydrolysis in protein breakdown at the level of the individual ATPase, we have introduced equivalent site-directed mutations into the ATPbinding motif of each RPT gene. Non-conservative substitutions of the active-site lysine were lethal in four of six cases, and conferred a strong growth defect in two cases. Thus, the ATPases are not functionally redundant, despite their multiplicity and sequence similarity. Degradation of a specific substrate can be inhibited by ATP-binding-site substitutions in many of the Rpt proteins, indicating that they co-operate in the degradation of individual substrates. The phenotypic defects of the different rpt mutants were strikingly varied. The most divergent phenotype was that of the rpt1 mutant, which was strongly growth defective despite showing no general defect in protein turnover. In addition, rpt1 was unique among the rpt mutants in displaying a G1 cell-cycle defect. Proteasomes purified from an rpt2 mutant showed a dramatic inhibition of peptidase activity, suggesting a defect in gating of the proteasome channel. In summary, ATP promotes protein breakdown by the proteasome through multiple mechanisms, as reflected by the diverse phenotypes of the rpt mutants.

The regulatory particle of the Saccharomyces cerevisiae proteasome.

The proteasome is a multisubunit protease responsible for degrading proteins conjugated to ubiquitin. The 670-kDa core particle of the proteasome contains the proteolytic active sites, which face an interior chamber within the particle and are thus protected from the cytoplasm. The entry of substrates into this chamber is thought to be governed by the regulatory particle of the proteasome, which covers the presumed channels leading into the interior of the core particle. We have resolved native yeast proteasomes into two electrophoretic variants and have shown that these represent core particles capped with one or two regulatory particles. To determine the subunit composition of the regulatory particle, yeast proteasomes were purified and analyzed by gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Resolution of the individual polypeptides revealed 17 distinct proteins, whose identities were determined by amino acid sequence analysis. Six of the subunits have sequence features of ATPases (Rpt1 to Rpt6). Affinity chromatography was used to purify regulatory particles from various strains, each of which expressed one of the ATPases tagged with hexahistidine. In all cases, multiple untagged ATPases copurified, indicating that the ATPases assembled together into a heteromeric complex. Of the remaining 11 subunits that we have identified (Rpn1 to Rpn3 and Rpn5 to Rpn12), 8 are encoded by previously described genes and 3 are encoded by genes not previously characterized for yeasts. One of the previously unidentified subunits exhibits limited sequence similarity with deubiquitinating enzymes. Overall, regulatory particles from yeasts and mammals are remarkably similar, suggesting that the specific mechanistic features of the proteasome have been closely conserved over the course of evolution.

Distal revascularization-interval ligation for limb salvage and maintenance of dialysis access in ischemic steal syndrome.

PURPOSE: Traditional options for treating ischemic steal syndrome related to a functioning dialysis access graft or fistula include banding or ligation. Unfortunately, these techniques usually result in inconsistent limb salvage, loss of a functional access, or both. We report our experience with an alternative method of limb revascularization that eliminates steal while maintaining continuous dialysis access. METHODS: Patients who had critical limb ischemia and functioning arteriovenous fistulae (AVF) underwent color-flow duplex scanning, digital photoplethysmography, and arteriography. Arterial ligation distal to the AVF origin eliminated the steal physiologic mechanism while arterial bypass grafting from above to below the AVF revascularized the extremity (distal revascularization-interval ligation [DRIL] procedure). RESULTS: From March 1994 through December 1996, 21 patients with functioning extremity AVFs presented with critical ischemia and steal syndrome. Eleven patients had chronic ischemia with rest pain, paresthesias, or ulcerations related to nine native fistulae (six brachiocephalic, two basilic vein transpositions, one radiocephalic) and two prosthetic bridge grafts (one upper arm, one lower extremity). Acute ischemia developed in 10 patients related to three native fistulae (two brachiocephalic, one radiocephalic) and seven prosthetic bridge grafts (three forearm, three lower extremity, one upper arm). All 21 patients were treated with the DRIL technique. Three of these patients required treatment for ischemia at the time of AVF construction. Nineteen of 21 bypass procedures were performed with autogenous vein, including nine brachial-brachial, three brachial-radial, two radial-radial, two brachial-ulnar, one popliteal-popliteal, one femoral-popliteal, and one femoral-peroneal. Polytetrafluoroethylene grafts were used for one external iliac-popliteal bypass graft and one axillary-brachial bypass graft. Limb salvage and maintenance of a functional fistula were achieved in 100% and 94%, respectively, at 18 months by life-table analysis. CONCLUSION: The DRIL technique reliably restores antegrade flow to the ischemic limb, eliminates the potential pathway for the steal physiologic mechanism, and maintains continuous dialysis access in these difficult patients.

Lipoxygenase reaction mechanism: demonstration that hydrogen abstraction from substrate precedes dioxygen binding during catalytic turnover.

Molecular oxygen is generally unreactive toward covalent bonds, due to spin conservation rules; a major role for oxygen-utilizing enzymes is, therefore, to activate dioxygen through a change in electronic configuration. In an effort to understand how lipoxygenase catalyzes lipid hydroperoxidation under conditions of catalytic turnover, kinetic deuterium isotope effects have been measured as a function of oxygen concentration. The properties of oxygen binding to lipoxygenase have also been pursued. The results presented herein show that, under steady state conditions, atmospheric oxygen enters the reaction pathway only after abstraction of hydrogen from substrate. Furthermore, it has not been possible to detect any form of lipoxygenase capable of binding molecular oxygen in the absence of activated substrate. We propose that molecular oxygen is not productively bound by lipoxygenase but rather interacts directly with the substrate radical lipoxygenase to form the hydroperoxyl radical of linoleate. A mechanism involving substrate activation, instead of the more familiar oxygen activation pathway, is a unique mechanism for a metallo-oxygenase.

Nature of rate-limiting steps in the soybean lipoxygenase-1 reaction.

A series of kinetic isotope effect experiments were performed with the goal of understanding the nature of rate-limiting steps in the soybean lipoxygenase-1 (SBL-1) reaction. SBL-1 was reacted with linoleic acid (LA) and deuterated linoleic acid (D-LA) under a variety of experimental conditions involving changes in temperature, pH, viscosity, and replacement of H2O with D2O. The extrapolated intrinsic primary H/D isotope effect can be estimated to be possibly as large as 80. This value is probably the largest isotope effect published for an enzymatic reaction, and much larger than that predicted from semiclassical models. Due to this large primary isotope effect, the C-D bond cleavage fully limits the rate of reaction under all conditions tested. In the case of protonated linoleic acid, a number of steps are partially rate-limiting at room temperature; three distinct mechanistic steps which include substrate binding, an H2O/D2O sensitive step, and C-H bond cleavage have been characterized. Use of glucose as a solvent viscosogen demonstrates that substrate binding is approximately 48% rate-limiting for LA at 20 degrees C. SBL-1 is one of the few enzymes that fit the definition of a "perfect enzyme", in the sense that further optimization of any single step at room temperature will not significantly increase the overall rate. At lower temperatures, the step sensitive to solvent deuteration begins to dominate the reaction, whereas at higher temperatures, the hydrogen abstraction step is rate-limiting. The pH dependence of kcat/Km for SBL-1 can be explained as arising from two pKa's, one controlling substrate binding and the other substrate release. Below pH 7.8, the rate of substrate release increases, thus decreasing the commitment to catalysis and unmasking the large intrinsic isotope effect on the subsequent hydrogen abstraction. An abnormally high pKa, in the range of 7-8, has been determined for LA in the concentration range employed in these studies. We propose that the negatively charged form of LA, predominating above pH 7.8, is the preferred substrate with larger commitments to catalysis.

Failure of foot salvage in patients with end-stage renal disease after surgical revascularization.

PURPOSE: This report ascertained factors responsible for for failure of foot salvage in patients with end-stage renal disease (ESRD) after undergoing infrainguinal bypass for critical ischemia. METHODS: A retrospective review of 69 distal arterial reconstructions performed in 53 patients with ESRD (hemodialysis [n = 37], kidney transplantation [n = 10], peritoneal dialysis [n = 6]) for foot gangrene (n = 28), nonhealing ulcer (n = 25), or ischemic rest pain (n = 16) was conducted. Endpoints of surgical morbidity, limb loss, and graft patency were correlated with extent of preoperative tissue loss and presence of diabetes mellitus. RESULTS: The 30-day operative mortality rate was 10%, and the patient survival rate at 2 years was 38%. The primary graft patency rate was 96% at 30 days, 72% at 1 year, and 68% at 2 years. Eleven of 22 foot amputations performed during the mean follow-up period of 14 months (range 3 to 96 months) occurred within 2 months of revascularization. Mechanisms responsible for limb loss included graft failure (n = 9), foot ischemia despite a patent bypass (n = 8), and uncontrolled infection (n = 5). Overall, 59% of amputations were performed in limbs with a patent bypass to popliteal or tibial arteries. Healing of forefoot amputations was prolonged, but all limb loss beyond 9 months of revascularization was due to graft failure. The limb salvage rate at 1 year decreased (p = 0.13) from 74% to 51% in patients admitted with gangrene. Only two of seven patients admitted with forefoot gangrene experienced foot salvage. CONCLUSION: Failure of foot salvage in patients with ESRD and critical ischemia was due to wound healing problems rather than graft thrombosis. Earlier referral for revascularization, before development of extensive tissue ischemia and infection, is recommended. Primary amputation should be considered in patients admitted with forefoot gangrene, particularly if it is complicated by infection.