The objective of the present study was to explore the correlation between the BAFF signal and HCMV-TLR activation in RTx recipients complicated by HCMV. Peripheral blood (anticoagulated by EDTA-Na2) and urine of 113 RTx recipients were collected; healthy volunteers were controlled. Urine HCMV-DNA was detected by real-time PCR. Recipients were classified into a positive group (>10,000 copies/mL urine) and a negative group (<10,000 copies/mL urine). ELISA results showed that sBAFF, sera anti-HCMV pp65 immunoglobulin (Ig)G antibody, and total IgG all significantly increased in recipients with positive HCMV-DNA (>10,000 copies/mL urine) (P < 0.05) compared with negative recipients (<10,000 copies/mL urine). In the positive group, HCMV-DNA copies and total IgG positively correlated with sBAFF (r = 0.988 and 0.625, respectively) (P < 0.05). Luminex assay results suggested that the incidence of anti-HLA I and II and MICA antibody obviously increased in positive recipients. The expression level of BAFF and BAFF-R increased in positive recipients. A total of 88 particular genes—involved in TLR signaling pathways, NF-κB signaling pathways, and cytokine-cytokine receptor signaling pathways—were detected in real-time PCR chip assay. A total of 46 genes were differentially expressed greater than two-fold, and the expression characteristic of BAFF-R was concordant with FACS results. Our findings are that activation of HCMV would induce or enhance the activation of BAFF code in RTx recipients, which may independently or cooperatively participate in renal allograft injury and decrease the long-term outcome of renal allografts.
activation-induced cytidine deaminase
absent in melanoma 2
B-cell-activating factor belonging to the tumor necrosis factor family
indirect fluorescent assay
mean fluorescence intensity
major histocompatibility complex class I-related chain A
NOD-, LRR-, and CARD-containing 5
nucleotide-binding domain and leucine-rich repeat-containing receptor
stimulator of interferon genes
transmembrane activator, calcium modulator, and cyclophilin ligand interactor
Rejection remains the primary reason for the loss of renal allografts. New immunosuppressive agents can effectively inhibit T-cell-mediated rejection and acute AMR, yet there is no satisfactory treatment for chronic AMR . Opportunistic infection, neoplasms, and drug poisoning occur in recipients because they must attain immunosuppressive status to achieve allograft tolerance. All of these immunological and non-immunological factors may have adverse effects on the long-term outcome of RTx . Therefore, it is necessary to further explore the injury mechanism of renal allografts.
HCMV is a herpesvirus that persistently infects a majority of the world’s human population. HCMV infection is the most common complication for RTx recipients . People with a functional immune system who are infected by HCMV are usually asymptomatic. However, infection of immunodeficient individuals or transplant recipients often leads to life-threatening disease. Clinical manifestation of CMV disease is diverse, from asymptomatic to dangerous consequences. Several studies have indicated that HCMV infection induces renal allograft rejection and decreases the survival time of renal allografts [4, 5]. However, the detailed mechanism is not clear.
During the initial stage of HCMV infection, large amounts of immunoglobulin (Ig)M-type anti-HCMV antibody are produced. B lymphocytes in hosts will then be constantly activated to produce IgG-type anti-HCMV antibodies that will exist over a host’s entire lifetime . HCMV-specific antibody, as well as alloantigen-specific antibody, is produced by activated B lymphocytes. It is accepted that cross-talk occurs between different signal pathways . However, will the activation of alloantigen-specific B lymphocytes be encouraged or prompted during the production of IgG-type anti-HCMV antibody? The answer to this question would help elucidate the mechanism of chronic rejection induced by HCMV infection in RTx recipients.
BAFF (also known as BLys, TALL-1, zTNF4, THANK, or TNFSF13b) is produced mainly by macrophages, monocytes, dendritic cells, and other non-myeloid cells. BAFF is a crucial homeostatic cytokine for B lymphocytes through BAFF-R, TACI, and BCMA . Excessive production of BAFF has been associated with the development of autoimmune diseases and autoimmune disorders, principally by acting on the maturation of splenic B lymphocytes [9, 10]. Some studies have also reported on BAFF-BAFF-R interactions in transplantation. BAFF was identified in RTx recipients with HLA immunization assessed by the measurement of PRA . BAFF has also been shown to be increased in rejected renal allografts in which B-lymphocyte tolerance breakdown occurs .
The first barrier for hosts in the face of danger signals is the germline-encoded PRR . There are three PRR: TLR, RLR, and NLR. Pathological DNA is recognized by TLR9 (TLR family), AIM2 (NLR family), and STING (RLR family). Once ligated, HCMV induces innate and adaptive immune responses as extraneous antigens, including HCMV-DNA and its packaged proteins .
Danger signal transduction has been reported to closely associate with BAFF-BAFF-R interaction. CpG-ODN, ligand for TLR9, and induced increasing expression of TACI are likely to play important roles in Ig secretion, following activation of B lymphocytes . Response to TLR3, and to a lesser extent TLR4 stimulation, was enhanced in a BAFF-dependent manner by rheumatoid synovial fibroblasts to induce prompt activation-induced cytidine deaminase (AID) and Ig class switching in B lymphocytes .
To some extent, CpG-ODN can mimic the effect of a DNA virus. However, CMV is a large, complex, enveloped virus constituting viral and cellular proteins other than viral DNA. Whether HCMV infection will enhance BAFF-BAFF-R interaction in RTx recipients remains unexplored.
MATERIALS AND METHODS
In the present study, a total of 113 follow-up RTx recipients were enrolled at Third Affiliated Hospital of Soochow University, Changzhou, Jiangsu, China, from October 2011 to June 2013. Of the 113 patients, 71 were male and 42 were female, ranging in age from 18 to 71 years; the average age was 47 years. Serum creatinine values ranged from 50 to 861 µmol/L (normal values: <144 µmol/L). Twenty-four healthy volunteers worked as controls, and their examination indexes were all within the normal range.
All subjects were first-time RTx with compatible ABO type, and they routinely took immunosuppressive agents, methylprednisolone, mycophenolate mofetil and cyclosporin A/tacrolimus (FK506). None of them took anti-CD20 antibodies or anti-CD52 antibodies.
All subjects committed to informed consent, and the study was approved by the Ethics Committee of the Third Affiliated Hospital of Soochow University based on the Declaration of Helsinki.
Peripheral blood (anticoagulated by edetate disodium salt dihydrate) of all subjects was collected. After centrifugation at 382 g for 10 min, serum was stored at −20°C. Total RNA in peripheral blood mononuclear cells was extracted immediately according to instructions provided by the manufacturer (QIAamp RNA Blood Mini Kit; QIAGEN, Hilden, Germany). Extracted total RNA was stored at −80°C, or reversed into cDNA and stored at −20 °C for future use.
Urine of all subjects was also collected. After centrifugation at 2483 g for 10 min, urine sediment cells were stored at −20 °C for future use.
HCMV-DNA quantification by real-time PCR
Urine HCMV-DNA in the RTx recipients was detected by HCMV-DNA detection kits (Diagnostic Kit for Quantification of Human Cytomegalovirus DNA; DaAn Gene, Guangzhou, China; CMV Detection Kit, BIOER, Hangzhou, China) according to instructions provided by the manufacturer. Extracted DNA from urine sediment cells worked as templates to run quantitative PCR using the provided virus-specific primers under the proper conditions.
Serum anti-HCMV IgG/IgM antibodies detection by IFA
HCMV pp65 antigen slides (Cytomegalovirus Antigen Slides; Virusys, Taneytown, MD, USA) were used to observe the titer of sera anti-HCMV IgG/IgM antibody, according to instructions provided by the manufacturer. In brief: (i) diluted (for IgG-type antibody) or undiluted (for IgM-type antibody detection) serum was added to slides for 30 min incubation at 37 °C; (ii) goat-anti-human IgG-FITC or goat-anti-human IgM-FITC (goat-anti-human IgG/M[H&L]-FITC [Fab’2] fragment with counter stain; Virusys) was added, respectively, for another 30 min incubation at 37 °C; and (iii) Staybright Mounting Media (for non-quenching results; Virusys) was added to slides to stop the incubation. Fluorescence signals were then observed by fluorescent microscope (Olympus IX71; Olympus, Tokyo, Japan). Apple-green indicated positivity on a background of reddish color of uninfected cells. Analysis of images and intensity of fluorescence were determined by Leica LAS v3.6 software (Leica Microsystems, Wetzlar, Germany).
Serum anti-HLA I and II and MICA antibody detection by Luminex xMAP technology
Sera anti-HLA I and II and MICA antibodies were analyzed by using a multiplexed microsphere-based suspension array from Luminex xMAP technology (Luminex Co., Austin, TX, USA). Detailed procedures were carried out according to instructions provided by the manufacturer (One Lambda, Canoga Park, CA, USA). In brief: (i) all sera were centrifuged at 2292 g for 5 min to remove polymers; (ii) 7 µL serum of each sample was added to the reactive plate (One Lambda); 2 µL beads (One Lambda) and 14 µL 1× PBS were added to each well, followed by incubation for 30 min at room temperature, avoiding light, and mixing for several seconds per 10 min; (iii) 50 µL goat-anti-human IgG-PE (One Lambda) was added to each well, with incubation for 30 min at room temperature, again avoiding light; and (iv) all samples were transferred onto a reading plate (One Lambda).
Determination of MFI was done with HLA Fusion 3.0 (One Lambda). Sera reactivity was assessed based on the fluorescent signal for each HLA-coated or MICA-coated microbead, following correction for non-specific binding to the negative control microbead. The normalized fluorescent signal is equal to the value of the antigen-coated microbead minus the value of the negative control microbead. If any one microbead in the mixed assay is positive, the result is considered positive.
Differentially expressed genes tested by quantitative real-time PCR chip assay
In this part of the experiment, three groups were assigned: HCMV-DNA-positive recipients, -negative recipients, and healthy controls—four cases/group, respectively. A total of 88 specific genes were selected according to Gene Ontology functional classification and KEGG Pathway analysis, considering the research background simultaneously. The specific primers of each specific gene were designed by Changzhou Chutian Bio-Tech Co. (Changzhou, China) and synthesized by Huada Gene Company (Guangzhou, China). Real-time PCR was run and analyzed by Changzhou Chutian.
The relative quantification of each objective gene was expressed by 2−ΔCt (average Ct of objective gene minus average Ct of reference gene). These data were also analyzed by Changzhou Chutian.
BAFF and its receptors detected on peripheral blood mononuclear cells by FACS
Of anticoagulated blood from each subject, 100 µL was placed in a tube. Then, carboxyfluorescein-conjugated mouse monoclonal anti-human BAFF (IC1241F; R&D Systems, Minneapolis, MN, USA), PE-conjugated mouse monoclonal anti-human TACI (FAB1741P; R&D Systems, Minneapolis, MN, USA), or antigen-affinity-purified goat polyclonal anti-human BAFF-R (AF1162; R&D Systems) and BCMA (BAF193; R&D Systems) were added, respectively. After incubation for 30 min at 4 °C, cell-lysing solution (BD Biosciences, San Jose, CA, USA) was added directly, or donkey anti-goat IgG (H + L)-PE (F0107; R&D Systems) was added for another 30-min incubation at 4 °C, and then cell-lysing solution was added. It was mixed adequately and stilled for 15 min. Then, all tubes were centrifuged at 250 g for 10 min; the supernatant was discarded and PBS was added, mixed, and examined by BD FACS Canto™ II (Becton, Dickinson and Company, San Jose, CA, USA). The results were analyzed by BD FACSDiva™ software (Becton, Dickinson and Company).
sBAFF and sera total IgG detection by ELISA
sBAFF was tested by Human BAFF/BLys/TNFSF13B Immunoassay ELISA Kit (R&D Systems), and serum total IgG was measured by Human IgG ELISA Kit (eBioScience, San Diego, CA, USA) according to the manufacturer’s instructions.
All statistical analysis was carried out using a statistical software package (SPSS version 17.0; SPSS Inc., Chicago, IL, USA). Measurement data were expressed as mean ± SD and analyzed by dependent t-test. Comparison among groups was analyzed by multiple analyses of variance. Correlation analysis between two variables was analyzed by Pearson correlation analysis. Statistical significance was defined as a P-value <0.05. All P-values were two-sided.
Urine HCMV-DNA levels in RTx recipients
According to instructions, when the Ct value of the reference gene was <36 and the Ct value of HCMV-DNA was not shown, the amount of HCMV-DNA in the sample was considered as 0 copy. Ct value > 10,000 copy/mL urine was set as the cut-off value . Thus, if the Ct value of HCMV-DNA was >10,000 copy/mL urine, the case was considered HCMV positive.
Positive HCMV-DNA copies were found in 12 RTx recipients from 113 cases, and their mean value was 28 594 ± 19, 419 copy/mL. Basic clinical details of these 12 RTx recipients are listed in Table 1. The other 101 cases of RTx recipients were all negative. Among them, the mean value of HCMV-DNA in 83 cases was between 0 and 10,000 copy/mL, and no HCMV-DNA was amplified in 18 cases.
|Patient no.||Gender||Age (years)||Postoperative period (months)||Creatinine (µmol/L)||HCMV-DNA (copies/mL)|
Sera anti-HCMV IgG/IgM antibody detection
Apple-green fluorescence was observed on the slides when serum anti-HCMV IgG/IgM antibodies existed, and the density of fluorescence paralleled the concentration of antibodies. Reddish fluorescence was observed if the cells were uninfected by HCMV, and as the background of images (Fig. 1a).
Leica v4 analysis software (Leica Microsystems) was applied for image analysis. During the analysis, colors except for green fluorescence were filtered from the images, then the green fluorescence was transformed into gray images, and the total gray values were calculated. Eight visual fields were randomly selected in each slide, and the mean values were calculated. The mean value of anti-HCMV IgG antibody of the positive group was 2,558 ± 604, and the mean value of the negative group was 910 ± 265. There was statistical significance between these two groups (P < 0.05) (Fig. 1b).
sBAFF and sera total IgG concentration in HCMV-DNA-positive RTx recipients significantly increased
ELISA results showed that the mean concentration of sBAFF in the positive group was 928 ± 360 pg/mL, whereas the mean concentration in the negative group was 511 ± 298 pg/mL. There was statistical difference between these two groups (P < 0.05) (Fig. 2).
The results also showed that sera total IgG in HCMV-positive RTx recipients significantly increased, compared with that in HCMV-negative recipients (26.26 ± 16 µg/mL vs 9.94 ± 7.37 µg/mL, P = 0.004) (Fig. 3).
Incidence of sera anti-HLA I and II and MICA antibody in HCMV-DNA-positive and -negative RTx groups
Data obtained from the Luminex xMAP technology was analyzed by Fusion 3.0 software (One Lambda). It was considered to be positive when MFI of anti-HLA antibody and anti-MICA antibody was higher than 1000, respectively. Results indicated that the incidence of anti-HLA and MICA antibody in the positive group was 46.2%; the incidence of anti-HLA and MICA antibody in the negative group was 12.5% (Table 2).
|Incidence of anti-HLA I and II antibody (%)||Incidence of anti-HLA I and II and MICA antibody (%)|
|Urine HCMV-DNA positive group||23.1||46.2|
|Urine HCMV-DNA negative group||8.75||12.5|
Differential mRNA levels of specific genes by real-time PCR chip assay
For the real-time PCR chip assay, four HCMV-DNA-positive cases were assigned to the positive group and four cases without HCMV-DNA amplification were assigned to the negative group. The other healthy volunteers (without HCMV-DNA amplification) were healthy controls.
In this experiment, 88 genes were examined in all, of which 54 genes involved TLR signaling pathways, 42 genes involved NF-κB signaling pathways, and 56 genes involved cytokine-cytokine receptor signaling pathways. Many specific genes simultaneously participated in several different signaling pathways. Analysis results suggested that the differential expression of total 46 genes (e.g. BAFF-R, IFN-γ) exceeded >two-fold. Compared with the negative group, two genes (BAFF-R and ICAM1) were significantly up-regulated, and 22 genes (e.g. TLR-9, NLRC5, LIGHT) were significantly down-regulated in the positive group. Compared with healthy controls, two genes (IFN-γ and ICAM1) were significantly up-regulated, and eight genes (e.g. CCL3L1, IL12B, ILIR1) were significantly down-regulated in the positive group. Compared with the healthy controls, 11 genes (e.g. IFN-γ, IRAK4, NF-κB2) were significantly up-regulated, and BAFF-R were significantly down-regulated in negative recipients. Analysis results of some particular genes are shown in Table 3.
|Analysis results between two groups|
|Gene name||Positive group vs negative group||P-value||Positive group vs healthy control group||P-value||Negative group vs healthy control group||P-value|
Expression of BAFF and BAFF-R significantly increased in peripheral blood mononuclear cells in HCMV-positive RTx recipients
FACS results showed that, compared with HCMV-negative RTx recipients, the incidence of BAFF and BAFF-R expression in peripheral blood mononuclear cells significantly increased (45.32 ± 9.67% and 50.43 ± 10.32% vs 10.81 ± 8.70% and 7.90 ± 10.40%, respectively, P < 0.05) in positive RTx recipients. However, compared with negative recipients, there was no significant difference in the incidence of BCMA and TACI between positive recipients and negative recipients (22.10 ± 10.98% and 25.29 ± 11.24% vs 12.10 ± 8.90% and 9.87 ± 10.20%, respectively, P > 0.05). Typical FACS images are shown in Figure 4.
Correlation analysis results of HCMV-DNA copies, sBAFF, and sera total IgG
Pearson correlation analysis indicated that HCMV-DNA copies significantly correlated positively with sBAFF concentration (r = 0.988, P = 0.012); sBAFF also significantly correlated positively with sera total IgG in RTx recipients (r = 0.625, P < 0.05).
In our earlier study, viral loads >104 copies/mL plasma continuing for 3 weeks served as a cut-off for predicting CMV interstitial pneumonia because of its high sensitivity and specificity . In the present study, urine HCMV-DNA was monitored because the kits were more sensitive for urine HCMV-DNA detection. To better explore the association of HCMV infection and BAFF code, 104 copies/mL urine was set as the cut-off value. In the present study, patients with viral loads ≧104 copies/mL urine were classified into the positive group, and patients with viral loads <104 copies/mL urine were classified into the negative group.
The IFA assay results suggested that not only RTx recipients, but also healthy volunteers, were infected by HCMV, because sera anti-HCMV IgG antibody could be found in both recipients and volunteers, which was concordant with previous reports . It was also observed that titers of sera anti-HCMV IgG antibody in the positive group were significantly higher than that in the negative group, a much smaller amount of sera anti-HCMV IgM antibody was detected in the positive group, and no sera anti-HCMV IgM was found in the negative group and healthy volunteers. These results implied that HCMV was activated in the positive group.
Results also showed that BAFF (including soluble and membrane-bound pattern) significantly increased in HCMV-DNA-positive recipients, and sBAFF concentration statistically correlated with the loading amount of CMV-DNA, which indicated that the active status of HCMV correlated with the expression level of BAFF.
It has been reported that excessive BAFF is always accompanied by abnormally high production of antibodies as a result of disturbance of B lymphocytes [13, 19, 20]. In the present study, BAFF-R, specific receptor for BAFF, was significantly high in the positive recipients. Also, total serum IgG protein significantly increased in the positive recipients, compared with negative recipients. Combined with these data, it could be considered that BAFF-BAFF-R interactions were enhanced in positive recipients.
It is known that elevated BAFF-R expression increases the incidence of renal allograft dysfunction in RTx recipients with stable normal transplanted renal function, and elevated sBAFF concentration increases the incidence of alloantibody production . Our earlier studies also suggested that abnormal expression of BAFF in RTx recipients was correlated with renal allograft AMR [22, 23]. In the present study, the incidence of anti-HLA I and II and MICA antibody in positive recipients was obviously higher than that in negative recipients, which implied that those recipients complicated with active HCMV infection will be prone to produce much more anti-HLA I and II and MICA antibody, which generally is closely correlated with allograft rejection and dysfunction .
In the present study, the transplanted renal function of all the HCMV-positive recipients was within the normal range (serum creatinine <144 µmol/L), which meant that these recipients could not be diagnosed with chronic rejection in the clinic. However, the data on increasing total IgG, sBAFF, and anti-HLA antibody levels suggested that chronic rejection existed in these recipients. Only powerful renal compensation can explain the phenomenon. In general, when more than 50% of renal function has been damaged, creatinine levels rise. As for RTx recipients, damage to the renal allografts may be irreversible and the long-term outcome may be reduced [4, 5]. These data revealed the necessity of further exploring the potential mechanism of renal allograft rejection and dysfunction induced by HCMV infection.
Activation of virus was accompanied by massive viral replication that would be recognized by its potential receptors, TLR9 and NLRC5 (NLR family) [25, 26]. Thus, in theory, expression of TLR9 and NLRC5 would be up-regulated. Moreover, TLR9 activation would induce expression of membrane-bound BAFF on human B lymphocytes and increase proliferation . In the present study, elevated levels of membrane-bound BAFF and serum total IgG protein were detected in HCMV-positive recipients. However, the chip results showed that TLR9 and NLRC5 mRNA were all down-regulated in positive recipients, to 6.54- and two-fold, respectively, compared with negative recipients. The following potential reasons may explain the differential results.
Phase of detection —these two receptors may exhaust because of usage.
TLR9 and NLRC5 can also recognize other pathogens, except for HCMV. Therefore, the expected results were not obtained on the basis of HCMV loads.
One of the antiviral immune responses for the patients .
Strategy to escape a host defense pathway for the virus .
MyD88 and TRAF6 are key adaptors during HCMV/TLR9, HCMV/NLRC5 signal transduction, and they are also involved in BAFF-BAFF-R interaction . IRAK4 was necessary for signal transduction between MyD88 and TRAF6 . The potential regulation mode is shown in Figure 5.
Nevertheless, we did not obtain significant results for MyD88 and TRAF6 mRNA, perhaps because of their specific phase of secretion. IRAK4 was a double-edged sword [31, 32], and its sophisticated expression characteristics in positive and negative recipients are worth investigating in depth.
In the present study, sBAFF and membrane-bound BAFF were all elevated, but chip assays showed a decreasing mRNA level of BAFF in positive recipients. These expression characteristics of BAFF were distinct from previous reports [33, 34] and need to be further explored.
In conclusion, our results show that BAFF code is enhanced in HCMV-positive RTx recipients, and cross-talk occurs between HCMV-TLR9/NLRC5 signal transduction and BAFF-BAFF-R interaction. Cross-talk results not only in amplification of inflammatory responses, but also in an increase of alloantibody production, thereby inducing deterioration of renal allografts and reducing the long-term outcome of RTx. These are the novel injury mechanisms of HCMV infection in RTx recipients, providing new clues for monitoring and therapy of HCMV infection in RTx recipients, and providing new targets for improving long-term outcome of renal allografts.
This study was supported by the National Science Foundation of Jiangsu Province (BK2011248).
The authors of this manuscript have no commercial or financial conflict of interest to disclose.