Journal Logo

Clinical and Translational Research

The Impact of Proteasome Inhibition on Alloantibody-Producing Plasma Cells In Vivo

Diwan, Tayyab S.1; Raghavaiah, Suresh1; Burns, Justin M.1; Kremers, Walter K.2; Gloor, James M.3; Stegall, Mark D.1,4

Author Information
doi: 10.1097/TP.0b013e3182081333

In sensitized renal transplant candidates, antibody against donor human leukocyte antigens (HLA) may present a major barrier to successful transplantation due the possibility of both early and late antibody mediated rejection (1–3). Terminally differentiated, long-lived PCs (plasma cells) in the bone marrow and other secondary lymphoid tissue are the most likely source of anti-HLA antibody (4–6). Because PCs are extremely rare, especially in peripheral blood, there have been relatively few studies of PCs in humans.

We previously showed that splenic PCs seem unaffected by protocols that include PE (plasma exchange), Thymoglobulin, Rituximab, and intravenous immunoglobulin (7). Subsequently, we isolated donor-specific antibody (DSA) PCs from the bone marrow of sensitized renal allograft candidates (8) and showed that the number of DSA-PCs in the marrow did not seem to decrease over time after bone marrow aspiration and was unaffected by pretransplant therapy with a combination of PE, Thymoglobulin, and intravenous immunoglobulin. Recently, we demonstrated in vitro that proteasome inhibition using bortezomib caused apoptosis of DSA-PCs thereby blocking alloantibody production (9).

This study extends the previous studies by showing that, in sensitized renal allograft candidates, pretransplant treatment in vivo with bortezomib monotherapy depletes marrow-derived DSA-PCs. Although serum DSA levels did not decrease in the time frame of the current study, bortezomib treatment seemed to enhance reduction of DSA using PE.


Impact of Proteasome Inhibition on Antigen-Specific Plasma Cells

Using an ELISPOT assay, we determined the number of bone marrow derived antigen-specific PCs that reacted against either: (1) tetanus toxoid (control antigen) and (2) the predominant donor HLA antigen (i.e., the HLA specificity of the prospective living donor for which the candidate had the highest serum antibody level). We analyzed sequential marrow samples in eight sensitized renal allograft candidates both 1 day before and 3 days after treatment with the proteasome inhibitor, bortezomib. One patient's baseline marrow clotted and was excluded from further analysis.

Two different dosing regimens were used. Four patients received one 4-dose cycle of bortezomib (single cycle) over 2 weeks and four patients received four 4-dose cycles (multiple cycles) over approximately 3 months. A single cycle decreased the number of DSA-PCs in three of four patients and multiple cycles reduced the number of DSA-PCs in all patients treated (Fig. 1). When the eight patients were considered as one group, proteasome inhibition decreased both the number of tetanus-specific PCs (mean=25.2±15.7 antitetanus PCs/mL pretreatment vs. 13.2±8.1/mL posttreatment, P=0.032, paired t test, Table 1) and the number of DSA-PCs (mean=16.7±14.5 DSA-PCs/mL marrow pretreatment vs. 6.2±3.6 cells/mL marrow after treatment, P=0.048). The number of DSA-PCs varied among patients (range, 6–48.9 cells/mL marrow) and did not correlate well with the serum levels of alloantibody.

Proteasome inhibition depletes antigen-specific marrow-derived plasma cells. (A) Representative flow cytometry of the bone marrow population enriched for plasma cells (CD138+, CD38+). (B) Representative ELISPOT showing numerous individual plasma cells (spots) specific for tetanus and human leukocyte antigen (HLA) pretreatment and a reduction in number of antigen-specific plasma cells after treatment. (C) Relative reduction in antigen-specific plasma cells with proteasome inhibition (P=0.032 for tetanus-specific plasma cells and P=0.048 for HLA-specific plasma cells). Four patients received one 4-dose cycle (open squares) and four received four 4-dose cycles (closed squares).
Impact of proteasome inhibition on bone marrow derived cell types showing a reduction in the number of antigen-specific plasma cells, but no decrease in the total number of cells or in the total plasma cell number

Proteasome inhibition did not significantly reduce the overall number of mononuclear cells isolated from the marrow (14.3±5.1×106 cells/mL marrow pretreatment to 11.6±3.9×106 cells/mL marrow posttreatment, P=0.27). However, the total number of PCs were numerically lower, but not significantly lower after treatment (21.5±8.6×103 pre vs. 15.5±12.1×103 post, P=0.21, Table 1).

The percentage of marrow-derived PCs that were specific for donor HLA and for tetanus was numerically lower after treatment (anti-HLA decreased from 0.07% to 0.04% and antitetanus decreased from 0.12% to 0.09%). This suggests that preexisting PC population was replaced by PCs with different and, likely new specificities.

Impact of Proteasome Inhibition on Peripheral Blood Cell Types

In the peripheral blood, some patients showed a decrease in the different cell types immediately after treatment (see Figure, Supplemental Digital Content 1,, but the decrease was not statistically significant including those for total peripheral blood leukocytes, total lymphocytes, total B lymphocytes, switched memory B cells, and plasmablasts. Platelets showed a significant decrease in their numbers posttreatment compared with baseline. Two patients treated with multiple cycles of bortezomib developed transient thrombocytopenia during treatment. But the platelet count returned to normal within 24 hr and did not require interruption of bortezomib treatment. One patient demonstrated low platelet count at 1 month follow-up but the counts returned to normal within 48 hr. Serum immunoglobulin levels remained normal throughout treatment and after the last dose of bortezomib.

Other Outcomes

Two patients, one treated with single cycle and one treated with multiple cycles of bortezomib developed central line infections during PE which necessitated change of vascular access. No patient developed peripheral neuropathy.

Impact of Proteasome Inhibition on Response to Plasma Exchange

The overall clinical goal of proteasome inhibitor treatment was to reduce DSA production and thus decrease serum DSA levels which might increase the likelihood of living donor kidney transplantation. However, given the long serum half-life of IgG (∼3 weeks) and persistent DSA production by residual PCs, we did not expect bortezomib treatment to lead to a significant decrease in serum DSA levels in the first few weeks after treatment. Indeed, the serum DSA level as determined by BFXM (B cell flow cytometric crossmatch) channel shift and single antigen bead assay was not different between pre and immediately posttreatment (see Table, Supplemental Digital Content 2,

Because patients in this study were motivated to proceed to transplantation, we examined the impact of DSA-PC reduction on DSA production by assessing its effect on the ability of multiple PEs to decrease serum DSA. We contend that this is a valid, clinically relevant endpoint for assessing decreased antibody production after proteasome inhibition.

Five bortezomib patients underwent multiple PEs. However, three bortezomib-treated patients did not undergo PE including the following: (1) one treated with a single cycle who did not demonstrate a reduction in DSA-PCs; (2) one whose donor was found to be unacceptable medically; and (3) one who identified a new donor against whom BFXM less than 300 and was transplanted.

For a comparison group, we identified all patients evaluated at our program between January 1, 2005 and January 1, 2010 who had similar levels of DSA at baseline (BFXM>450, n=8) and who underwent DSA-reduction therapy with PEs, but were not treated with bortezomib. The demographics of this group were similar to the five bortezomib-treated patients who also underwent PE (Table 2) with respect to age, sex, race, cause of ESRD, and history of previous transplant. Detailed antidonor HLA antibody levels (as determined by B FXM) also were similar (see Table, Supplemental Digital Content 2, The control and bortezomib-treated patients underwent a similar number of PEs (11.6±3.9 control vs. 11.4±2.7 bortezomib group, P=0.9, Table 3). Bortezomib treatment resulted in a greater decrease in DSA secondary to PE (Difference in BFXM CS before and after PE −272.6±92.1 channels) compared with the control group that underwent PE alone (−95.4±72.2 channels, P=0.008).

Patient demographics of sensitized renal allograft candidates who underwent antibody-reduction with multiple plasma exchanges including a control group (patients treated with plasma exchange alone, n=8) and the bortezomib group (patients treated with bortezomib, then subsequently treated with plasma exchange, n=5)
Proteasome inhibition potentiates antidonor HLA antibody reduction with plasma exchange

In the PE-alone control group, no patient achieved a low enough serum DSA level (BFXM<300) to proceed to transplant. In contrast, 60% (3 of 5) of patients treated with proteasome inhibition and PE were able to achieve a BFXM less than 300 (Fig. 2). Of these, two received one cycle and one received multiple cycles of bortezomib.

Relative reduction of serum levels of donor-specific alloantibody with plasma exchange therapy as measured by the binding to donor B cells (B flow cytometric crossmatch) showing that in patients starting with similar levels of serum alloantibody, treatment with bortezomib led to a greater response to PE (B) compared with those not treated with bortezomib (A). By clinical protocol, PE therapy must achieve a BFXM channel shift <300 in order for the patient to proceed to transplantation. This was achieved in a greater number of bortezomib-treated patients undergoing plasma exchange (3 of 5) compared with the untreated control group (0 of 8) (P=0.035; (single cycle, open squares; multiple cycles, closed squares).

Of the patients who achieved a BFXM less than 300, one patient's donor was denied due to medical reasons and two others were transplanted.

Posttransplant Outcomes

The two patients who underwent transplantation are 12 and 18 months into their follow-up, respectively. Both have functioning allografts and their serum creatinine values are 1.0 and 1.7 mg/dL, respectively, at latest follow-up. The patient 12 months into follow-up continues to have persistent DSA and shows diffuse C4D positivity on her protocol biopsy. The second patient has no DSA and no evidence of C4D positivity. None of the patients show evidence of rejection (see Table, Supplemental Digital Content 3,


The current study demonstrates that monotherapy using the proteasome inhibitor bortezomib significantly depletes bone marrow derived DSA-PCs in sensitized renal allograft candidates who have high levels of DSA at baseline. In addition, this PC depletion seems to lead to a clinically significant reduction in DSA production that potentiates antibody reduction using PE.

Mechanistically, these studies extend the findings of previous studies. Neubert et al. (10) showed that proteasome inhibition depleted normal plasma cells and ameliorated disease in a mouse model of lupus. Our group previously showed that bortezomib blocks the proteasome pathway of normal human plasma cells in vitro leading to apoptosis and decreased alloantibody production (9).

These data also suggest that while PC depletion is significant, it is not complete. The factors determining sensitivity of normal PCs to proteasome inhibition are not known. While studies from myeloma cell lines suggest that sensitivity to proteasome inhibition correlates with secretory rates (11), other factors including specific cell characteristics of the PC (12) and its location within the marrow also may affect their proclivity for removal (13, 14). The fact that serum DSA levels remained high in this study might be a function of the time frame of the study or may be because the treated patients were those who had high levels of DSA. Further studies are needed to determine the impact of bortezomib in patients with lower levels of DSA.

We contend that the antigen-specific PCs in this study are likely prototypical long-lived PCs (15). Although little is known about their rate of turnover, these studies suggest that the marrow population of long-lived PCs is rapidly replaced after depletion. These data also suggest that replacement PCs have new target antigens and do not arise from a pool of cells with specificities similar to that of preexisting PCs. Other B-cell types in the peripheral blood also were reduced by proteasome inhibition but were rapidly replaced. The proteasome pathway is present in all eukaryotic cells including many involved in antibody production such as T cells, memory B cells, and dendritic cells (16, 17). Thus, it is possible that the reduction in DSA production detected in the current study might be due, in part, to changes in the number and function of non-PC cell types.

The fact that PC depletion is only partial may be a positive aspect of the therapy because complete abolition of existing antibody production could lead to profound immunodeficiencies and increased infection rates in transplant candidates. The number of patients in the current study is too small to clearly determine the optimum dosing of bortezomib and more data will be needed to determine if more prolonged treatment leads to greater PC depletion and DSA reduction.

Most clinical studies of bortezomib in transplant patients have involved posttransplant usage to prevent antibody mediated rejection (usually combined with cellular rejection) or to deplete antibody that developed de novo after transplantation (9, 18–21). The reduction in serum DSA levels observed in several studies could be due to factors other than bortezomib treatment including absorption of antibody by the allograft (22), the development of blocking “antiidiotypic” antibodies (23), or the self-limited nature of an acute alloimmune response during acute cellular rejection (24). Recently, a report of two sensitized renal allograft candidates treated pretransplant similar to our current study, showed only a mild reduction in serum DSA levels after one 4-dose cycle (25). In light of the findings of the current study, this apparent lack of success may be due to a shorter treatment schedule and the lack of concomitant antibody removal with PE.

The current study has several limitations, and these data should be viewed with some caution. The difficult logistics of performing paired marrows and a relatively protracted therapeutic trial have resulted in relatively few patients being included in this detailed protocol. The lack of a randomized, control group with paired bone marrow analyses is another limitation. Although we previously have shown that sequential bone marrow aspirations pretransplant similar to the current study does not lead to depletion of DSA-PCs, we acknowledge that a true randomized, control trial would be preferred to truly assess the role of bortezomib on PC depletion. However, performing two bone marrows on untreated patients raises ethical issues at this time point. Finally, the use of a historical comparison group for PE also could introduce selection biases. However, we contend that the two groups are very similar and represent typical patients that present to our institution for desensitization therapy.

In conclusion, the current study demonstrates that proteasome inhibition causes depletion of DSA-PCs. The depletion of DSA-PCs seems to enhance DSA reduction by PE. However, the overall impact of bortezomib on serum DSA levels seems to be somewhat mild in these patients who have high DSA levels at baseline and much DSA remains. Thus, we contend that the current study is more a proof of principal. Further studies are needed to clarify the optimum dosing regimen and whether combination therapy with bortezomib or newer proteasome inhibitors might someday lead to improved desensitization protocols.


Study Design

The study was a prospective, nonrandomized, trial to determine the impact of treatment with the proteasome inhibitor, bortezomib, on the number of DSA-PCs in bone marrow. A second, clinical endpoint aimed to determine whether the proteasome inhibition reduced anti-HLA antibody production involved measuring the reduction in serum DSA levels in response to multiple PEs and compared bortezomib-treated patients with a historical control group that underwent PE alone.

Donor-Specific Alloantibody Determination

We use the BFXM to estimate the total amount of DSA because B cells express both class I and class II HLA. This allows us to use one number to estimate DSA levels in all patients, including those with DSA against class I only, class II only or both classes I and II. The results of this assay are expressed as the mean channel shift above a known negative control serum. As previously described (24), our channel shift data have been “standardized” using commercially available beads with known fluorescence (molecules of equivalent soluble fluorochrome units [MESF], Quantum TM, FITC MESF Premix, Bangs Laboratories, Fishers, IN). This allows other laboratories to compare their results with our laboratory's results. The conversion formula was determined to be: number of channels (raw or uncorrected=−528+104.22×log[MESF]). From this equation, the average raw fluorescence of our negative control/unstained B-lymphocytes of 200 channels represents a MESF value of approximately 1100. A channel shift of 300 (DSA level required for a patient to proceed to transplantation) is a raw fluorescence of 500 or 300 channels above background and corresponds to a MESF of approximately 19,300.

In addition, the specificity and amount of anti-HLA antibodies in the serum is validated using commercially available solid phase assay containing microspheres coated with a single HLA type (LABscreen, One Lambda, Canoga Park, CA).


This study was conducted with informed consent using a protocol approved by the institutional review board of the Mayo Foundation and Clinic. For this study, all patients had a BFXM channel shift more than 450 (MESF>80,000) against their prospective living donor and all had identifiable antibodies with specificities for at least one donor HLA type by single antigen beads. These patients were chosen for bortezomib treatment because our historical experience demonstrated that despite multiple PEs, candidates with this level of DSA at baseline were unable to achieve a BFXM less than 300. Per our program's protocol (24), a candidate must achieve a BFXM less than 300 to proceed to transplantation. This level was decided on to avoid hyperacute rejection episodes and to avoid an extremely high rate of early antibody mediated rejection after transplantation. In addition, all patients had end-stage renal disease and qualified for a renal transplant using our program's criteria. The patients in the treatment group were sequential sensitized candidates who presented to our program and who met criteria from August 2008 to August 2009. The antidonor HLA antibody mean fluorescence intensity as determined by the SAB assay is presented in Table 3, Supplemental Digital Content 4, The fact that the mean fluorescence intensity obtained in the SAB assay remained high in these patients even after diluting the serum to more than or equal to 1:16 illustrates the fact that these patients had extremely high DSA levels at baseline.

Plasma Exchange

PE was performed using a daily regimen (one plasma volume exchange using albumin replacement). The BFXM was performed using serum collected the day before the first PE and the serum collected immediately after the final PE.

Peripheral Blood Analyses: Leukocytes subtypes, B-Cell phenotypes, and Serum IgG Levels

In addition to the antigen-specific studies, the impact of proteasome inhibition on several components of peripheral blood also was analyzed at baseline and at 3 and 30 days after the final dose of drug. Assessment included: total numbers of leukocytes, lymphocytes, platelets; and B-cell subtypes using multicolor flow cytometry: total B cells (CD19), switched memory B cells (CD27+, IgD, IgM), and plasmablasts (CD38+, IgM). Total serum immunoglobulin levels were also measured.

Antibody-Secreting Plasma Cell Studies

The details of the PC ELISPOT have been described previously (8). Purified HLA (One Lambda, using the same protein as that attached to the microspheres) or tetanus toxoid were plated on synthetic nanofiber polyamine surface membranes (Surmodics Inc, Eden Prairie, MN) placed in 24-well plates (Falcon, BD Labware, NJ, USA). The donor HLA type with the highest measured serum antibody level was selected for the ELISPOT assay. 250,000 CD138+ cells are plated on HLA and tetanus ELISPOTs which were run simultaneously and in duplicate. Data were expressed as the number of antigen-specific PCs/ml of bone marrow aspirate.

Proteasome Inhibitor Dosing Schedule

The dosing schedule was based on schedules used for the treatment of myeloma. No dose adjustment was made for renal function. The first four patients received one 4-dose cycle of bortezomib (1.3 mg/m2 intravenously on days 1, 4, 8, and 11). After this initial experience that demonstrated tolerability and partial efficacy, the next five patients received a more prolonged four cycle 4-dose schedule (1.3 mg/m2 intravenously on days 1, 4, 8, and 11) with at least 10 days between doses. Antigen-specific PC numbers were evaluated 1 day before the first treatment (day 0) and 3 days after the last infusion.

Response to PE

By protocol, all patients received at least seven daily PE treatments to achieve a BFXM less than 300. Patient who did not achieve a BFXM less than 300 with seven PEs received a maximum of 14 treatments before being considered a treatment failure.

Statistical Analyses

Numerical data are expressed as mean±standard deviation and nominal data by counts and percents. The proportions of nominal data were tested using Fisher's exact test, two sided; numerical data, paired data (e.g., pretreatment and posttreatment numbers) were tested using a paired t test and unpaired data using the two-sample t test. A P value of less than or equal to 0.05 was considered significant.

For the assessment of the impact of bortezomib on serum DSA levels, patients in the comparative group were compared with those treated with bortezomib with respect to (a) mean decrease in BFXM channel shift after PE treatments and (b) number that achieved a BFXM less than 300 with treatments.


1. Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med 1969; 280: 735.
2. Lee P-C, Terasaki PI, Takemoto SK, et al. All chronic rejection failures of kidney transplants were preceded by the development of HLA antibodies. Transplantation 2003; 74: 1192.
3. Stegall MD, Gloor JM, Winters JL, et al. A comparison of plasmapheresis versus high-dose IVIG desensitization in renal allograft recipients with high levels of donor specific alloantibody. Am J Transplant 2006; 6: 346.
4. Manz RA, Thiel A, Radbruch A. Lifetime of plasma cells in the bone marrow. Nature 1997; 388: 133.
5. Slifka MK, Ahmed R. Long-lived plasma cells: A mechanism for maintaining persistent antibody production. Curr Opin Immunol 1998; 10: 252.
6. Ellyard JI, Avery DT, Phan TG, et al. Antigen-selected, immunoglobulin-secreting cells persist in human spleen and bone marrow. Blood 2004; 103: 3805.
7. Ramos EJ, Pollinger HS, Stegall MD, et al. The effect of desensitization protocols on human splenic B-cell populations in vivo. Am J Transplant 2007; 7: 402.
8. Perry DK, Pollinger HS, Burns JM, et al. Two novel assays of alloantibody-secreting cells demonstrating resistance to desensitization with IVIG and rATG. Am J Transplant 2008; 8: 133.
9. Perry DK, Burns JM, Pollinger HS, et al. Proteasome inhibition causes apoptosis of normal human plasma cells preventing alloantibody production. Am J Transplant 2009; 9: 201.
10. Neubert K, Meister S, Moser K, et al. The proteasome inhibitor bortezomib depletes plasma cells and protects mice with lupus-like disease from nephritis. Nat Med 2008; 14: 748.
11. Meister S, Schubert U, Neubert K, et al. Extensive immunoglobulin production sensitizes myeloma cells for proteasome inhibition. Cancer Res 2007; 67: 1783.
12. Lee A-H, Iwakashi NN, Anderson KC, et al. Proteasome inhibitors disrupt the unfolded protein response in myeloma cells. Proc Nat Acad Sci USA 2003; 100: 9946.
13. Sapaznikov A, Pewzner-Jung Y, Kalchenko V, et al. Perivascular clusters of dendritic cells provide critical survival signals to B cells in bone marrow niches. Nat Immunol 2008; 9: 388.
14. Nair RR, Tolentino J, Hazlhurst LA. The bone marrow microenvironment as a sanctuary for minimal residual disease in CML. Biochem Pharmacol 2010; 80: 602.
15. Fairfax KA, Kallies A, Nutt SL, et al. Plasma cell development: From B-cell subsets to long-term survival niches. Semin Immunol 2008; 20: 49.
16. Chtanova T, Tangye SG, Newton R, et al. T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J Immunol 2004; 173: 68.
17. Gelinck LB, Jol-vander Zijde CM, Jansen-Hoogendijk AM, et al. Restoration of the antibody response upon rabies vaccination in HIV-infected patients treated with HAART. AIDS 2009; 23: 2451.
18. Everly MJ, Everly JJ, Susskind B, et al. Bortezomib provides effective therapy for antibody and cell-mediated rejection. Transplantation 2008; 86: 1754.
19. Walsh RC, Everly JJ, Brailey P, et al. Proteasome inhibitor-based primary therapy for antibody-mediated renal allograft rejection. Transplantation 2010; 89: 277.
20. Trivedi HL, Terasaki PI, Feroz A, et al. Abrogation of anti-HLA antibodies via proteasome inhibition. Transplantation 2009; 87: 1555.
21. Sberro-Soussan R, Zuber J, Suberbielle-Boissel C, et al. Boretzomib as the sole post-renal transplantation desensitization agent does not decrease donor-specific anti-HLA antibodies. Am J Transplant 2010; 10: 681.
22. Adeyi OA, Girnita Al, Howe J, et al. Serum analysis after transplant nephrectomy reveals restricted antibody specificity patterns against structurally defined HLA class I mismatches. Transplant Immunol 2005; 14: 52.
23. Hardy MA, Suciu-Foca N, Reed E, et al. Immunomodulation of kidney and heart transplants by anti-idiotypic antibodies. Ann Surg 1991; 214: 522.
24. Burns JM, Cornell LD, Perry DK, et al. Alloantibody levels and acute humoral rejection early after positive crossmatch kidney transplantation. Am J Transplant 2008; 8: 2684.
25. Warhman M, Haidinger M, Körmöczi GF, et al. Effect of the proteasome inhibitor bortezomib on humoral immunity in two presensitized renal transplant candidates. Transplantation 2010; 89: 1385; epub ahead of print.

Proteasome inhibitor; Alloantibodies; Plasma cells; Clinical kidney transplantation; Sensitized patients

Supplemental Digital Content

© 2011 Lippincott Williams & Wilkins, Inc.