Complement-fixing antibodies against denatured HLA and MICA antigens are associated with antibody mediated rejection

    Complement-fixing antibodies against denatured HLA and MICA antigens are associated with antibody mediated rejection Junchao Cai, Paul I. Terasaki, Dong Zhu, Nils Lachmann, Constanze Sch¨onemann, Matthew J. Everly, Xin Qing PII: DOI: Reference:

S0014-4800(15)00234-8 doi: 10.1016/j.yexmp.2015.11.023 YEXMP 3866

To appear in:

Experimental and Molecular Pathology

Received date: Accepted date:

19 November 2015 20 November 2015

Please cite this article as: Cai, Junchao, Terasaki, Paul I., Zhu, Dong, Lachmann, Nils, Sch¨ onemann, Constanze, Everly, Matthew J., Qing, Xin, Complement-fixing antibodies against denatured HLA and MICA antigens are associated with antibody mediated rejection, Experimental and Molecular Pathology (2015), doi: 10.1016/j.yexmp.2015.11.023

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Complement-fixing antibodies against denatured HLA and MICA antigens

PT

are associated with antibody mediated rejection

Everly1, and Xin Qing4 1

SC

Terasaki Research Institute, Los Angeles, CA

RI

Junchao Cai1*, Paul I. Terasaki1, Dong Zhu2, Nils Lachmann3, Constanze Schönemann3, Matthew, J.

2

NU

Department of Urology, Zhongshan Hospital, Fudan University, Shanghai, China

3

MA

Charité-Universitätsmedizin Berlin, Center for Tumor Medicine, HLA-Laboratory CampusVirchow-

Klinikum, Berlin, Germany 4

TE

*

D

Department of Pathology, Harbor-UCLA Medical Center, Torrance, CA 90502

AC CE P

Corresponding author: Junchao Cai, [email protected]

Contributions of authors to the manuscript: J Cai: study concept and design, statistical analysis and interpretation of data, drafting of the manuscript P.I. Terasaki: obtained funding, acquisition of data, and study supervision D Zhu, Performance of the experiments, data analysis, critical revision of the manuscript N. Lachmannand C. Schönemann: patient samples and data collection, critical revision of the manuscript M.J. Everly: critical revision of the manuscript X Qing: proposal on potential mechanisms of c1q-fixing denatured HLA/MICA antibodies development, critical revision of the manuscript

ACCEPTED MANUSCRIPT 2

PT

Funding source: Terasaki Family Foundation Disclosures: All authors have no conflict of interest

SC

RI

Running Title: C1q-fixing denatured antigen-specific antibody and graft outcome

Keywords: c1q-fixing antibody; HLA antibody; MICA antibody; denatured antigen; antibody epitope;

NU

renal transplant; antibody-mediated rejection.

MA

Word Count Abstract: 280

TE

Figures: 5

AC CE P

Table: 0

D

Text: 3755

ACCEPTED MANUSCRIPT 3

Abbreviations:

PT

HLA: human leukocyte antigen

RI

MICA: MHC class I chain-related gene A

SC

AMR: antibody mediated rejection

AC CE P

TE

D

MA

NU

CMR: cell mediated rejection

ACCEPTED MANUSCRIPT 4 Abstract: Background: We have found antibodies against denatured HLA class I antigens in the serum of allograft

PT

recipients which were not significantly associated with graft failure. It is unknown whether transplant

RI

recipients also have denatured HLA class II and MICA antibodies. The effects of denatured HLA class I, class II, and MICA antibodies on long-term graft outcome were further investigated based on their ability

SC

to fix complement c1q.

NU

Materials and Methods: In this 4-year retrospective cohort study, post-transplant sera from 975 kidney transplant recipients were tested for antibodies against denatured HLA/MICA antigens and these

MA

antibodies were further classified based on their ability to fix c1q.

D

Results: Thirty percent of patients had antibodies against denatured HLA class I, II, or MICA antigens.

TE

Among them, 8.5% and 21.5% of all patients had c1q-fixing and non c1q-fixing antibodies respectively. There was no significant difference on graft survival between patients with or without antibodies against

AC CE P

denatured HLA/MICA. However, when these antibodies were further classified according to their ability to fix c1q, patients with c1q-fixing antibodies had a significantly lower graft survival rate than patients without antibodies or patients with non c1q-fixing antibodies (p=0.008). In 169 patients who lost renal grafts, 44% of them had c1q-fixing antibodies against denatured HLA/MICA antigens, which was significantly higher than that in patients with functioning renal transplants (25%, p<0.0001). C1q-fixing antibodies were more significantly associated with graft failure caused by AMR (72.73%) or mixed AMR/CMR (61.9%) as compared to failure due to CMR (35.3%) or other causes (39.2%) (p=0.026). Conclusions: Transplant recipients had antibodies against denatured HLA class I, II, and MICA antigens. However, only c1q-fixing antibodies were associated with graft failure which was related to antibody mediated rejection.

ACCEPTED MANUSCRIPT 5 Introduction Accumulating evidence has shown that HLA and MICA antibodies are major causes of allograft rejection

PT

[1-7]. Sequence-based antibody-epitope mapping revealed that alloantibodies detected in transplant

RI

recipients recognized not only surface residues but also non-surface residues of target antigens [8-13]. This finding suggested that there are two types of antibodies in transplant recipients, one targets intact

SC

antigen and the other targets denatured antigen. To investigate the clinical importance of both intact and

NU

denatured antigen specific antibodies, we did a cohort study to compare the long-term graft outcome of patients who developed either intact or denatured HLA class I antibodies. We demonstrated that only

MA

intact but not denatured HLA class I antibodies are predictive of graft failure [9]. Complement dependent cytotoxicity (CDC) assay was introduced into transplant clinic in 1964to identify

D

complement-fixing antibody which is cytotoxic against (potential) donor cell target [14, 15]. Cytotoxicity

TE

of alloantibodies has been proved to be significantly associated with graft failure [16, 17]. As we

AC CE P

discussed in our previous review, even though CDC assay was proved to be more significantly associated with graft failure, one of the major disadvantages of CDC assay is its low sensitivity[18]. In recent year, due to the development of solid phase HLA antibodies detection technology using luminex, a new technique to detect c1q-fixing antibodies based on Luminex technology has been introduced correspondingly and proved to be more effective to identify a subgroup of HLA/MICA antibodies which have significant association with graft outcome [19-21]. The questions we try to address in this study include: 1) What are the frequencies of denatured HLA class I, class II, and MICA specific antibodies in transplant recipients? 2) What are the frequencies of c1qfixing and non c1q-fixing antibodies specific for denatured HLA and MICA antigens? 3) Do denatured HLA class II and MICA antibodies have a similar effect on graft outcome as denatured HLA class I antibodies as reported previously? 4) Will they have different effects on graft outcome if denatured HLA/MICA antibodies are further classified into c1q-fixing and non c1q-fixing antibodies based on their

ACCEPTED MANUSCRIPT 6 ability to bind complement fragment c1q? 5) Is there any association between c1q-fixing antibodies and antibody mediated rejection in patients who lost allografts during 4-year follow-up post antibody testing?

PT

Patients and Methods

RI

Patients and characteristics

SC

A total of 975 kidney recipients with follow-up information transplanted between 1984 and 2004 at the Charite-Universitätsmedizin in Berlin, Germany were enrolled in this study. Among them, 59% were

NU

female and 41% were male patients. Mean age of all recipients was 43.6 with a standard deviation of 13.2. Immunosuppressive drug therapy information was not available in 10% of patients. The other patients

MA

were on either double or triple therapy. The percentages of patients using PRD, CSA, FK, AZA, MMF, and RAPA were 81%, 46%, 44%, 25%, 39%, and 7% respectively. Eighty two percent of all patients

D

received deceased donor transplants and 86.2% of all recipients were primary renal graft recipients.

TE

Patient sera were tested once at least 6 months post-transplant and graft function was monitored 4 year

AC CE P

post antibody testing. The average time that the sera were tested was 6.8±5.5 years post-transplant. Sixteen percent of all patients had pre-transplant antibodies detected by ELISA. Detection of alloantibodies against denatured HLA and MICA antigens LABScreen® Mixed antigen beads (One Lambda Inc., Canoga Park, CA) were spun down, washed once with PBS containing 0.1% BSA, then incubated in an appropriate volume of PBS containing 0.1% BSA at 90°C for 5 minutes, then cooled down to 4°C in a PCR machine (Gene Amp® PCR System 9700; Applied Biosystems, Foster City, CA). The beads were then washed with PBS containing 0.1% BSA 3 times in a 1.5ml Eppendorf tube. The beads were then spun down and re-suspended in an appropriate volume of PBS with 0.1% BSA. Alloantibodies recognizing denatured HLA and MICA antigens were identified using heat-treated antigen beads using Luminex assay. An MFI of 1,000 and test-to-control ratio of 2 was used as cutoff to define an antibody positive serum.

ACCEPTED MANUSCRIPT 7 Detection of complement c1q binding HLA and MICA antibodies Patient serum was heat inactivated at 560C for 90 minutes and then mixed with c1q (One Lambda Inc.,

PT

Canoga Park). Mixed antigen beads which contain HLA class I, class II, and MICA antigens were mixed with patient serum with c1q and incubated for 1 hour. Beads were washed three times and incubated with

RI

PE-conjugated anti-C1q antibodies. Then, beads were read with LabScanTM 100 after three time washes

SC

with wash buffer. A cutoff value of 200 MFI level was used to define a c1q binding antibody positive

NU

serum. Statistical analysis

MA

All study analyses were conducted using STATA version 9.0 (StataCorp LP, College Station, Texas). Survival rates were calculated using Kaplan-Meier methods. Statistical significances were determined by

D

the log-rank test for comparison of survival curves. Hazard ratios and statistical significance of variables

AC CE P

TE

were calculated using Cox proportional hazard model.

ACCEPTED MANUSCRIPT 8 Results Frequencies of c1q-fixing and non c1q-fixing antibodies against denatured HLA class I, class II, and

PT

MICA antigens Post-transplant sera from 975 renal transplant recipients enrolled in this study were tested for c1q-fxing

RI

antibodies against denatured HLA class I, class II, and MICA antigens. As shown in Figure 1, 24%, 15%,

SC

and 17% of patients had detectable antibodies against denatured HLA class I, II, and MICA antigens. In combination, approximately 30% of all patients had antibodies specific for either HLA class I, II, or

NU

MICA antigens. Among all patients, 5.85%, 4.10%, and 4.21% of them had c1q-fxing antibodies, while

MA

18.05%, 10.87%, and 12.62%had non c1q-fixing antibodies against denatured class I, II, and MICA antigens respectively. Overall, 8.51% and 21.54% of all patients had c1q-fixing and non c1q-fixing antibodies against denatured HLA and/or denatured MICA antigens.

TE

D

Antibodies against denatured HLA/MICA are not predictive of graft failure In this retrospective cohort study, one post-transplant serum from each patient was tested for antibodies

AC CE P

against denatured HLA class I, II, and MICA antigens. Four-year graft survival rates post antibody testing were compared between patients with and without denatured HLA and/or MICA antibodies. Figure 2 demonstrates that similar to denatured HLA class I specific antibodies (Fig 2a, p=0.8594), patients with antibodies specific for denatured HLA class II (Fig 2b, p=0.8256) and MICA (Fig 2c, p=0.3492) had comparable 4 year graft survival rates when compared to antibody negative patient groups. Graft survival difference remained statistically insignificant when all HLA class I, II, and MICA antibodies were considered ((Fig 2d, p=0.4958). C1q-fixingantibodies against denatured HLA/MICA are predictive of graft failure When denatured HLA and MICA specific antibodies were further classified into c1q-fixing and non c1qfixing antibodies based on their ability to bind complement fragment c1q, long-term graft survival rates were compared between these two groups and to the patient group without antibodies. Figure 3 demonstrates that patients with c1q-fixing HLA class I (Fig. 3a, p=0.0193)), class II (Fig. 3b, p=0.7892), and MICA (Fig. 3c, p=0.0227) antibodies had lower graft survival rated as compared to patients without

ACCEPTED MANUSCRIPT 9 denatured antigen-specific antibodies or patients with non c1q-fixing antibodies. Difference reached statistical significance in HLA class I and MICA groups. When HLA class I, class II, and MICA

PT

antibodies were considered together, the graft survival difference within three comparison groups became more significant (Fig. 3d, p=0.0084).

RI

C1q-fixing antibodies against denatured HLA/MICA are associated with graft loss due to AMR

SC

In a total of 975 patients, 169 patients lost their renal grafts, while the rest 806 patients had functional renal grafts within 4-year follow-up post antibody testing. Figure 4 shows that 43.79% of patients with

NU

failed allografts had c1q-fixing antibodies, while only 25.31% of patients with functioning grafts

MA

developed c1q-fixing antibodies against denatured HLA/MICA (p<0.0001). In 169 graft failure patients, the causes of graft failure include AMR (N=11), AMR/CMR (N=21), CMR (N=17), CSA cytotoxicity (N=3), and recurrence of original diseases (N=7), while biopsy information were not available in the rest

TE

D

of 110 patients. Figure 5 shows thatc1q-fixing antibodies were detected at significantly higher frequencies in patients who lost renal grafts due to AMR(72.73%)and AM/CMR (61.90%) when

AC CE P

compared to patients who lost grafts due to CMR (35.29%) or other causes (39.17%, p=0.026).

ACCEPTED MANUSCRIPT 10 Discussion Right after flow cytometry and Luminex microbead-based HLA antibody detection kits became

PT

commercially available by One Lambda Inc., we found that purified HLA antigens coated on the surface

RI

of microbeads contained both intact and denatured antigens. This finding was reported at the International HLA and Transplantation Conference held Jan 30th-Feb 1st, 2004 in Los Angeles. Thereafter, in an

SC

International HLA and Immunogenetics Workshop study, we reported that there are two types of HLA

NU

class I antibodies in post-transplant sera of renal graft recipients. One is specific for intact and the other is specific for denatured (β2m-free heavy chain) HLA class I antigens. Only intact but not denatured HLA

MA

class I antibodies were associated with graft failure [9].

Our current study not only confirms the previous findings but also demonstrates for the first time that

D

HLA class II and MICA antibodies have the similar features to that of HLA class I antibodies. After

TE

further investigation on denatured HLA class I, class II, and MICA antibodies, we are the first to show

AC CE P

that not all the denatured HLA/MICA antibodies are clinically unimportant. Complement fixing ability can be used to differentiate the clinically important antibodies from the unimportant ones. Cell-based complement-dependent (CDC) cytotoxicity assay detects cytotoxic antibody against alloantigen on target cell, which has been introduced into transplant clinic for more than five decades [15, 16]. It has been widely used in cross-match test for selection of a low risk donor and in PRA test for the screening of pre-existing alloantibodies [18]. However, it has been rarely used for monitoring patients for the development of anti-donor antibodies post-transplant due to its technical complexity and low sensitivity[18]. Since flow cytometry and Luminex bead technology were introduced, purified HLA and MICA antigen-based technique has been widely used to detect HLA and MICA antibodies in pre- and post-transplant sera of allograft recipients [22, 23].These commercially available HLA/MICA antibody testing kits mainly detect HLA and MICA binding IgG antibodies regardless of their isotypes and cytotoxic activity[18]. Based on the hypothesis that cytotoxic antibodies are clinically more important

ACCEPTED MANUSCRIPT 11 than non-cytotoxic antibodies, method in detecting c1q-fixing HLA and MICA antibody was developed and c1q-fixing HLA/MICA antibodies detected using this technique were proved to be a better predictor

PT

of graft outcome in some studies[19-21].

RI

The most significant finding of this study is to identify a subgroup of patients who have c1q-fixing antibodies against denatured HL/MICA antigens and are under the risk of potential graft loss. Without

SC

further classification of denatured HLA/MICA antibodies based on their c1q-fixing ability, denatured

NU

HLA/MICA antibodies have been previously considered clinically unimportant [9]. These findings raise some important questions to be addressed which include: 1) “Are these c1q-fixing denatured HLA/MICA

MA

antibodies preexisting natural antibodies or alloantibodies induced by transplantation or pre-transplant sensitization?”2) What are the corresponding potential mechanisms of the development of these

D

antibodies?” 3) Will identification of the donor specificities of these antibodies better predict graft

TE

outcome?

AC CE P

This is a cohort study with only post-transplant sera tested for c1q-fixing antibodies against denatured HLA/MICA. Since pre-transplant sera were not available for most of the patients, we could not determine whether these antibodies were natural antibodies not relating to allo-sensitization, or alloantibodies induced by pre-or post-transplant allo-sensitization. Different from intact HLA/MICA antibodies which seemed to be associated with variables such as previously transplantation, pre-transplant PRA, patient age etc, the positivity of denatured HLA/MICA antibodies were not associated with any patient, donor or transplant related variables in multivariate logistic regression analysis(data not shown). These data seemed to suggest that these antibodies might be of natural origin and exist regardless of allo-sensitization. However, a definite answer requires further investigation. To identify antibody epitopes in HLA antibody positive transplant recipients, we are the first to report that allograft recipients had HLA antibodies specific for not only surface but also cryptic epitopes [12, 13]. Later, Morales demonstrated that 63% of 424 healthy males with no known sensitization history had

ACCEPTED MANUSCRIPT 12 naturally occurring HLA antibodies [24]. Further follow-up study on natural antibodies revealed that the majority of HLA class I natural antibodies recognized denatured HLA class I antigen (data not published).

PT

El-Awar’s epitope analysis of HLA class I antibodies found in sera of normal healthy males and cord blood reported that 60% of identified potential epitopes involve hidden amino acid residues[25]. These

RI

reports implied that denatured HLA/MICA antibodies were not necessarily triggered by transplantation

SC

since un-sensitized healthy males already have these antibodies. However, for a transplant recipient who had denatured HLA/MICA antibodies, it is more complicated to determine the origin of these antibodies

NU

since both naturally occurring and alloantigen-induced denatured HLA/MICA antibodies may co-exist in

MA

patient periphery.

In regard to the potential mechanism of induction of natural antibodies against denatured HLA/MICA, it

D

is hypothesize that these antibodies are produced against cross-reactive epitopes found in microorganisms,

TE

allergens, vaccines, and animal source foods, etc. Alberú J et al. demonstrated that hepatitis B vaccination induced HLA antibodies production in 9 out of 20 PRA negative healthy donors[26]. Due to the fact that

AC CE P

we are extensively exposed to an environment which is full of a great variety of xeno-antigens, it is relatively more difficult to prove that exposure to a specific antigen, such as microorganism, allergen, or especially animal source food intake can induce HLA antibody production; however, these xeno-antigens are theoretical potential inducers of specific antibody responses which cross-react with HLA antigens, since these antigens may have shared epitopes with HLAs. Protein sequence alignment results revealed that MHC antigens of animals have different level of sequence homology with human HLA antigens. For instance, sequence homology of MHC class I heavy chains between chicken and human or pig/cattle and human vary between 40 to 70%.Since dietary proteins of animal origin are mainly absorbed in amino acid, dipeptide, and tripeptide forms, it is a question regarding how these animal source food antigens trigger antibody responses in humans. According to Erickson’s review published in 1990[27], some peptides can also be absorbed intact and remain un-hydrolyzed after absorption. Roberts also reported that large (10-51 amino acids) peptides generated in the diet can be absorbed intact through the intestines and produce

ACCEPTED MANUSCRIPT 13 biologic effects at the tissue level [28]. The above evidence seems to support the hypothesis that exposure to ingested proteins, such as animal source foods, might be a major player in causing HLA/MICA

PT

antibody positivity especially denatured HLA/MICA antibodies, since foods of animal origin are usually cooked and MHC antigens become denatured before consumption. Certainly, it needs further

RI

investigation to see whether vegetarians or vegans (better candidates) have less denatured HLA/MICA

SC

antibodies.

NU

Besides the possibility of naturally occurring antibodies with c1q-fixing ability against denatured HLA/MICA antigens, allo-sensitization against denatured HLA/MICA antigens on donor tissue cells may

MA

also be a potential mechanism to induce c1q-binding antibodies in allograft recipients. It has been reported that denatured HLA antigens are present on the surface of human cell surface, even though the

D

amount of the denatured form of the antigen is much less than its intact form [29, 30]. Allo-sensitization

TE

will certainly trigger antibody responses against not only the intact but also the denatured form of alloantigen. Moreover, in the process of allograft rejection in which more graft cells are under apoptosis

AC CE P

or necrosis due to cell- or antibody-mediated rejection, more denatured HLA/MICA antigens will appear on the cell surface and are released into circulation [30-33]. This pathological process of graft rejection will further stimulate the production of denatured HLA/MICA antibodies, and some of which may fix complement.

It has been previously reported that antibody detected by antibody screening assay using Luminex mixed antigen beads predict graft failure [6, 34]. In our current study, due to the fact that limited information were available on HLA and MICA typing for both patients and recipients, HLA and MICA antibodies were tested using Luminex mixed antigen beads without identification of donor specificities. Given the fact that further identification of donor specificities of alloantibodies using Luminex single antigen beads better predict graft outcome [9, 35], we predict that c1q-fixing denatured donor HLA/MICA antibodies may better correlate with graft failure. However it requires further investigation to confirm.

ACCEPTED MANUSCRIPT 14 In conclusion, we demonstrated that transplant recipients had antibodies against not only denatured HLA class I as previously reported but also denatured HLA class II, and MICA antigens. These antibodies

PT

seemed clinically unimportant. However, when these denatured HLA/MICA antibodies were further classified according to their ability to fix complement, c1q-fixing antibodies seemed to be associated with

AC CE P

TE

D

MA

NU

SC

RI

graft failure which was possibly due to antibody mediated rejection.

ACCEPTED MANUSCRIPT 15 Figure legends: Figure 1. Frequencies of c1q-fixing and non c1q-fixing antibodies against denatured HLA class I, class II,

PT

and MICA antigens.

RI

Figure 2. Antibodies against denatured HLA and MICA are not predictive of graft failure. A) Denatured

SC

HLA class I specific antibodies; B) Denatured HLA class II specific antibodies; C) Denatured MICA

NU

specific antibodies; D. Denatured HLA and/or MICA specific antibodies. Figure 3. C1q-fixing antibodies against denatured HLA class I, II and/or MICA are predictive of graft

MA

failure. A) C1q-fixing and non c1q-fixing antibodies against denatured HLA class I antigens; B) C1qfixing and non c1q-fixing antibodies against denatured HLA class II antigens; C) C1q-fixing and non c1q-

D

fixing antibodies against denatured MICA antigens; D) C1q-fixing and non c1q-fixing antibodies against

TE

denatured HLA class I, II, and/or MICA antigens.

AC CE P

Figure 4. Percentages of c1q-fixing antibodies against denatured HLA and/or MICA antigens in patients with functioning or failed renal grafts. Figure 5. C1q-fixing antibody is associated with antibody mediated rejection. AMR: antibody mediated rejection; AMR/CMR: mixed antibody- and cell-mediated rejection; CMR: cell-mediated rejection. Others include CSA cytotoxicity (N=3), recurrence of original diseases (N=7), and patients with no biopsies (N=110).

ACCEPTED MANUSCRIPT 16

AC CE P

TE

D

MA

NU

SC

RI

PT

ACKNOWLEDGMENTS We thank Naomi Anderson for technical assistance. This work was supported by Terasaki Family Foundation.

ACCEPTED MANUSCRIPT 17 References:

8. 9. 10. 11. 12.

13. 14. 15. 16. 17. 18. 19.

20. 21.

22.

PT

RI

SC

6. 7.

NU

5.

MA

4.

D

3.

TE

2.

Cai, J., et al., Humoral theory of transplantation: some hot topics. Br Med Bull, 2013. 105: p. 13955. Terasaki, P.I. and J. Cai, Human leukocyte antigen antibodies and chronic rejection: from association to causation. Transplantation, 2008. 86(3): p. 377-83. Ozawa, M., et al., 14th International HLA and Immunogenetics Workshop: report on the Prospective Chronic Rejection Project. Tissue Antigens, 2007. 69 Suppl 1: p. 174-9. Terasaki, P.I. and J. Cai, Humoral theory of transplantation: further evidence. Curr Opin Immunol, 2005. 17(5): p. 541-5. Cai, J. and P.I. Terasaki, Humoral theory of transplantation: mechanism, prevention, and treatment. Hum Immunol, 2005. 66(4): p. 334-42. Terasaki, P.I., Humoral theory of transplantation. Am J Transplant, 2003. 3(6): p. 665-73. Lee, P.C., et al., All chronic rejection failures of kidney transplants were preceded by the development of HLA antibodies. Transplantation, 2002. 74(8): p. 1192-4. El-Awar, N., et al., Epitopes of HLA-A, B, C, DR, DQ, DP and MICA antigens. Clin Transpl, 2009: p. 295-321. Cai, J., et al., Intact HLA not beta2m-free heavy chain-specific HLA class I antibodies are predictive of graft failure. Transplantation, 2009. 88(2): p. 226-30. Cai, J. and P.I. Terasaki, Post-transplantation antibody monitoring and HLA antibody epitope identification. Curr Opin Immunol, 2008. 20(5): p. 602-6. El-Awar, N., et al., Epitopes of the HLA-A, B, C, DR, DQ and MICA antigens. Clin Transpl, 2007: p. 175-94. Cai, J., et al., Development of nondonor-specific HLA-DR antibodies in allograft recipients is associated with shared epitopes with mismatched donor DR antigens. Am J Transplant, 2006. 6(12): p. 2947-54. Cai, J., S. Kohanof, and P.I. Terasaki, HLA-DR antibody epitopes. Clin Transpl, 2006: p. 103-14. Terasaki, P.I. and N.E. Rich, Quantitative Determination of Antibody and Complement Directed against Lymphocytes. J Immunol, 1964. 92: p. 128-38. Terasaki, P.I. and J.D. McClelland, Microdroplet Assay of Human Serum Cytotoxins. Nature, 1964. 204: p. 998-1000. Patel, R. and P.I. Terasaki, Significance of the positive crossmatch test in kidney transplantation. N Engl J Med, 1969. 280(14): p. 735-9. Cook, D.J., et al., The flow cytometry crossmatch in kidney transplantation. Clin Transpl, 1987: p. 409-14. Cai, J. and P.I. Terasaki, Human leukocyte antigen antibodies for monitoring transplant patients. Surg Today, 2005. 35(8): p. 605-12. Chin, C., et al., Clinical usefulness of a novel C1q assay to detect immunoglobulin G antibodies capable of fixing complement in sensitized pediatric heart transplant patients. J Heart Lung Transplant, 2011. 30(2): p. 158-63. Piazza, A., et al., Post-transplant development of C1q-positive HLA antibodies and kidney graft survival. Clin Transpl, 2013: p. 367-75. Yabu, J.M., et al., C1q-fixing human leukocyte antigen antibodies are specific for predicting transplant glomerulopathy and late graft failure after kidney transplantation. Transplantation, 2011. 91(3): p. 342-7. Pei, R., et al., Flow cytometric detection of HLA antibodies using a spectrum of microbeads. Hum Immunol, 1999. 60(12): p. 1293-302.

AC CE P

1.

ACCEPTED MANUSCRIPT 18

29. 30.

31.

32. 33.

34. 35.

PT

RI

SC

28.

NU

27.

MA

26.

D

25.

TE

24.

Pei, R., et al., Single human leukocyte antigen flow cytometry beads for accurate identification of human leukocyte antigen antibody specificities. Transplantation, 2003. 75(1): p. 43-9. Morales-Buenrostro, L.E., et al., "Natural" human leukocyte antigen antibodies found in nonalloimmunized healthy males. Transplantation, 2008. 86(8): p. 1111-5. El-Awar, N., D. Cook, and P.I. Terasaki, HLA class I epitopes: A and B loci. Clin Transpl, 2006: p. 79-94. Alberu, J., et al., A non-allogeneic stimulus triggers the production of de novo HLA antibodies in healthy adults. Transpl Immunol, 2007. 18(2): p. 166-71. Erickson, R.H. and Y.S. Kim, Digestion and absorption of dietary protein. Annu Rev Med, 1990. 41: p. 133-9. Roberts, P.R., et al., Effect of chain length on absorption of biologically active peptides from the gastrointestinal tract. Digestion, 1999. 60(4): p. 332-7. Burlingham, W.J. and F.H. Claas, Soluble MHC in allograft recipients. Hum Immunol, 1999. 60(5): p. 401-2. DeVito-Haynes, L.D., et al., The metalloproteinase-mediated pathway is essential for generation of soluble HLA class I proteins by activated cells in vitro: proposed mechanism for soluble HLA release in transplant rejection. Hum Immunol, 1998. 59(7): p. 426-34. DeVito-Haynes, L.D., et al., Monitoring of kidney and simultaneous pancreas-kidney transplantation rejection by release of donor-specific, soluble HLA class I. Hum Immunol, 1994. 40(3): p. 191-201. Sell, K.W., et al., Studies of major histocompatibility complex class I/II expression on sequential human heart biopsy specimens after transplantation. J Heart Transplant, 1988. 7(6): p. 407-18. Wang, Y.C., et al., Influence of cytokines and immunosuppressive drugs on major histocompatibility complex class I/II expression by human cardiac myocytes in vitro. Hum Immunol, 1991. 31(2): p. 123-33. Terasaki, P.I. and M. Ozawa, Predicting kidney graft failure by HLA antibodies: a prospective trial. Am J Transplant, 2004. 4(3): p. 438-43. Lachmann, N., et al., Anti-human leukocyte antigen and donor-specific antibodies detected by luminex posttransplant serve as biomarkers for chronic rejection of renal allografts. Transplantation, 2009. 87(10): p. 1505-13.

AC CE P

23.

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

19

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

20

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

21

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

22

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

23

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

24

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

25

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

26

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

27

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

28

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

29