MALT1 serves as a biomarker for estimating disease risk of lupus nephritis: a prospective case-control study.

Background: Mucosa-associated lymphoid tissue lymphoma translocation protein 1 (MALT1) regulates immune/inflammation response and kidney injury, indicating that it might participate in the lupus nephritis (LN) pathology. The present study sought to investigate the utility of MALT1 serving as an indicator of severity and inflammation biomarker in LN.
Methods: In total, 30 LN patients and 30 non-LN systemic lupus erythematosus (SLE) patients were recruited, and blood MALT1 expression was detected using reverse transcription-quantitative polymerase. Besides, the SLE disease activity index was calculated for all subjects to assess disease activity.
Results: MALT1 was more increased in LN patients than non-LN SLE patients {2.26 [interquartile range (IQR): 1.38-3.54] vs. 1.04 (IQR, 0.60-1.62); P<0.001}. Additionally, the multivariate logistic regression model analysis indicated that higher MALT1 [odds ratio (OR): 3.097, 95% CI: 1.292-7.425, P=0.011] and higher serum creatinine (Scr) (OR: 1.055, 95% CI: 1.012-1.099, P=0.011) were independently related to an elevated risk of LN. The MALT1 and Scr-based nomogram exhibited a good value in estimating LN risk with an area under the curve (AUC) of 0.873 (95%CI: 0.787-0.960). MALT1 was highest in LN patients with class IV (accompanied by V or not), followed by class V and class III (accompanied by V or not) patients, and lowest in class II patients (P=0.046). MALT1 was positively linked to the LN activity index (rs=0.370, P=0.044); however, while it was positively associated with the LN chronicity index, the level did not reach statistical significance (rs=0.341, P=0.066). MALT1 was positively linked to 24 h proteinuria (rs=0.473, P=0.008), Scr (rs=0.378, P=0.039), and C-reactive protein (CRP) (rs=0.410, P=0.025) in LN patients. Conclusions: MALT1 might serve as a useful biomarker for estimating the disease risk in LN patients; however, future large-sample size studies need to be conducted to validate our findings. 2022 Annals of Translational Medicine. All rights reserved.

Introduction

Lupus nephritis (LN) is a common inflammatory kidney disease caused by the deposition of immune complexes (ICs) in the glomeruli, whose clinical features include hematuria and proteinuria (1-3). As a major organ involved in systemic lupus erythematosus (SLE), about 40–50% of SLE is complicated with LN; additionally, LN is more prevalent in females than males with a female-to-male ratio ranging from 6.1:1 to 15:1 (4,5). Notably, once LN occurs, end-stage renal disease occurrence risk is greatly increased in SLE patients, and it might eventually cause death in these patients (6-8). Thus, identifying some feasible biomarkers that will assist clinicians to monitor the disease risk and severity of LN is meaningful.

Mucosa-associated lymphoid tissue lymphoma translocation protein 1 (MALT1) is an important intracellular signaling protein with both scaffold and proteolytic functions, which activates immune and inflammation-related signaling pathways (9,10). On account of the above functions, previous studies observe that MALT1 participates in the development of several autoimmune/inflammatory diseases [e.g., psoriasis and rheumatoid arthritis (RA)] (11-13). For example, a previous study showed that MALT1 triggers a glutaminase 1-mediated glutaminolysis and immune imbalance, which further promotes the skin lesions of psoriasis (11). Another study showed that MALT1 is linked to disease activity in RA patients, and its decrement is related to the treatment response and remission following the administration of a conventional disease‐modifying antirheumatic drug or a biologics‐based regimen (13). In relation to kidney injury, MALT1 binds to B cell lymphoma 10 (BCL-10) and further facilitates renal fibrosis in angiotensin II (Ang II)-induced renal damage (14). Consequently, MALT1 might serve as a useful biomarker in LN management. The previous studies find some biomarkers for monitoring the disease severity and inflammation level of LN, such as urinary galectin-3 binding protein, CD163, etc. (15,16). Nonetheless, the utility of MALT1 serving as an indicator of severity and inflammation biomarker in LN remains unknown in LN.

Hence, the current study sought to investigate the clinical value of MALT1 in indicating LN risk and its link to disease severity and inflammation in LN patients. We present the following article in accordance with the TRIPOD reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-22-3442/rc).

Methods

Subjects

Thirty LN patients treated between January 2019 and June 2021 in The First People’s Hospital of Jingzhou were sequentially included in this study. Patients had to meet the undermentioned inclusion criteria to be eligible for this study: (I) have a diagnosis of LN in accordance with the American College of Rheumatology criteria (17); (II) be aged 16–80 years; and (III) be willing to comply with the protocol of data collection, sample collection, and disease evaluation. Patients who met any of the following exclusion criteria were excluded: (I) had LN complicated with a hematologic malignancy; (II) had LN concomitant with a solid tumor; and/or (III) were a female patient who was pregnant or breast feeding. Additionally, during the same period, 30 SLE patients without LN (non-LN SLE patients) were also recruited as controls. The same exclusion criteria as that for the LN patients also applied to the non-LN SLE patients. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of The First People’s Hospital of Jingzhou [No. 22 (2013)]. Written informed consent was obtained from the patients or their parental/legal guardians.

Data collection

After enrollment, the clinical data were obtained from all subjects, including demographics, clinical manifestations, biochemical indexes, SLE features, and histological features. The SLE disease activity index (SLEDAI) was also calculated for all subjects to assess disease activity (18).

Assessment

For LN patients, LN classification was obtained and categorized as per the classification criteria of the International Society of Nephrology/Renal Pathology Society 2003 (19). The LN activity index and LN chronicity index were determined using the method proposed by Austin et al. (20).

Sample collection and detection

Peripheral blood samples of all subjects were collected after enrollment, and the peripheral blood mononuclear cell samples were then isolated to detect MALT1 expression by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). Total ribonucleic acid (RNA) was obtained by RNeasy Protect Mini Kit (Qiagen, Germany); after which, PrimeScript™ RT reagent Kit (Takara, China) was used for RT. Next, the qPCR was finished by KOD SYBR® qPCR Mix (Toyobo, Japan). The 2−ΔΔCt method was used for calculation of MALT1 relative expression, for which glyceraldehyde 3-phosphate dehydrogenase was used as the internal reference. The previous study were referred for the qPCR primer design (13).

Statistical analysis

The data processing (analysis) and graph construction were conducted via SPSS version 22.0 (IBM Corp., USA) and GraphPad Prism version 7.02 (GraphPad Software Inc., USA), respectively. The clinical characteristics and MALT1 expression were compared between the LN and non-LN SLE patients using the Chi-square test, Student’s t-test, Fisher’s exact test, or Mann-Whitney U test. The profile of MALT1 expression in differentiating subjects was estimated using the receiver-operating characteristic (ROC) curve. Factors affecting the risk of LN were investigated using a multivariate logistic regression model analysis and nomogram (all the collected factors were included in the analysis by the forward stepwise method). The correlation of MALT1 expression with clinical features in LN patients was determined through the Kruskal-Wallis H rank-sum test, Mann-Whitney U-test or Spearman r test. All tests were two-sided, and a P value <0.05 was considered statistically significant.

Results

Clinical features

A total of 37 LN patients were invited, then 5 of them were excluded because of being ineligible for inclusion and 2 of them disagreement to participate. The remaining 30 patients were recruited. In total, 30 LN patients and 30 non-LN SLE patients were recruited with mean ages of 42.1±19.4 and 51.0±15.7 years, respectively (see ). The LN group comprised 4 (13.3%) males and 26 (86.7%) females, while the non-LN SLE group comprised 3 (10.0%) males and 27 (90.0%) females. The median SLE duration of the LN and non-LN SLE patients was 48.0 [interquartile range (IQR), 2.0–140.0] months and 24.0 (IQR, 2.2–48.0) months, and their median SLEDAI scores were 9.0 (IQR, 6.0–14.0) and 9.5 (IQR, 7.0–11.0), respectively. Notably, there was no difference in most of the clinical characteristics of patients between the 2 groups (all P>0.050), except for arthritis manifestation (P=0.010), albumin (P<0.001), 24 h proteinuria (P<0.001), serum creatinine (Scr) (P<0.001), blood urea nitrogen (BUN) (P=0.005), and immunoglobulin G (P=0.006). Additionally, 3 (10.0%), 3 (10.0%), 14 (46.7%), 3 (10.0%), 2 (6.7%), and 5 (16.7%) LN patients were assessed as class II, class III, class IV, class V, class V + III, and class V + IV, correspondingly. The median LN activity index and chronicity index of the LN patients were 7.5 (IQR, 4.0–10.0) and 3.0 (IQR, 2.0–4.0), respectively. The specific clinical features were set out in .

Table 1

Clinical characteristics

Items	Non-LN SLE (N=30)	LN (N=30)	P value
Demographics
Age (years), mean ± SD	51.0±15.7	42.1±19.4	0.057
Gender, n (%)			0.688
Male	3 (10.0)	4 (13.3)
Female	27 (90.0)	26 (86.7)
Clinical manifestations, n (%)
Affected nervous system	3 (10.0)	6 (20.0)	0.472
Affected cardiovascular system	3 (10.0)	5 (16.7)	0.706
Affected blood system	15 (50.0)	15 (50.0)	1.000
Photo allergy	2 (6.7)	3 (10.0)	1.000
Acropachia	1 (3.3)	0 (0.0)	1.000
Arthritis	21 (70.0)	11 (36.7)	0.010
Reynold’s phenomenon	4 (13.3)	0 (0.0)	0.112
Alopecia	10 (33.3)	11 (36.7)	0.787
Rash	15 (50.0)	14 (46.7)	0.796
Oral ulcer	3 (10.0)	3 (10.0)	1.000
Velcro rales	3 (10.0)	1 (3.3)	0.612
Chest tightness	2 (6.7)	1 (3.3)	1.000
Tussiculation	4 (13.3)	0 (0.0)	0.112
Chest pain	0 (0.0)	0 (0.0)	–
Fever	6 (20.0)	6 (20.0)	1.000
Biochemical indexes
WBC (×109/L), median (IQR)	3.92 (3.14–5.78)	5.56 (4.02–8.53)	0.051
Hb (g/L), median (IQR)	108.5 (97.2–117.8)	99.0 (86.5–109.8)	0.099
Platelet (×109/L), mean ± SD	221.6±125.5	197.8±108.0	0.433
ALB (g/L), mean ± SD	34.5±4.6	27.7±6.9	<0.001
ALT (U/L), median (IQR)	18.0 (11.2–31.2)	18.5 (11.0–30.2)	0.923
24 h proteinuria (g), median (IQR)	0.078 (0.052–0.144)	2.045 (0.180–3.622)	<0.001
Scr (µmol/L), median (IQR)	57.2 (53.4–68.3)	84.7 (68.2–114.4)	<0.001
BUN (mmol/L), median (IQR)	4.18 (3.34–5.39)	5.77 (4.26–10.59)	0.005
CRP (mg/L), median (IQR)	3.67 (1.14–11.09)	7.66 (3.35–21.46)	0.099
ESR (mm/h), median (IQR)	44.0 (30.0–62.8)	44.5 (30.9–65.8)	0.923
C3 (g/L), median (IQR)	0.70 (0.45–0.85)	0.48 (0.38–0.68)	0.099
C4 (g/L), median (IQR)	0.14 (0.09–0.18)	0.12 (0.08–0.18)	0.668
IgA (g/L), median (IQR)	2.48 (2.14–3.05)	2.08 (1.53–2.91)	0.107
IgG (g/L), median (IQR)	18.81 (16.01–20.82)	14.08 (9.13–18.66)	0.006
IgM (g/L), median (IQR)	0.94 (0.72–1.33)	0.80 (0.60–1.21)	0.325
ANA positive, n (%)	30 (100.0)	29 (96.7)	1.000
Anti-dsDNA positive, n (%)	14 (46.7)	16 (53.3)	0.606
Anti-Sm positive, n (%)	9 (30.0)	8 (26.7)	0.774
Anti-SSA positive, n (%)	20 (66.7)	22 (73.3)	0.573
Anti-SSB positive, n (%)	6 (20.0)	4 (13.3)	0.488
Anti-nRNP positive, n (%)	19 (63.3)	14 (46.7)	0.194
SLE features, median (IQR)
Disease duration (months)	24.0 (2.2–48.0)	48.0 (2.0–140.0)	0.213
SLEDAI score	9.5 (7.0–11.0)	9.0 (6.0–14.0)	0.976
Histological features
LN classification, n (%)			–
Class II	–	3 (10.0)
Class III	–	3 (10.0)
Class IV	–	14 (46.7)
Class V	–	3 (10.0)
Class V + III	–	2 (6.7)
Class V + IV	–	5 (16.7)
LN activity index, median (IQR)	–	7.5 (4.0–10.0)	–
LN chronicity index, median (IQR)	–	3.0 (2.0–4.0)	–

LN, lupus nephritis; SLE, systemic lupus erythematosus; SD, standard deviation; IQR, interquartile range; WBC, white blood cell; Hb, hemoglobin; ALB, albumin; ALT, alanine aminotransferase; Scr, serum creatinine; BUN, blood urea nitrogen; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; C3, complement C3; C4, complement C4; IgA, immunoglobulin A; IgG, immunoglobulin G; IgM, immunoglobulin M; ANA, Anti-nuclear antibody; Anti-dsDNA, anti-double-stranded DNA antibody; Anti-Sm, anti-Smith antibody; Anti-SSA, Anti-Sjögren’s-syndrome-related antigen A antibody; Anti-SSB, Anti-Sjögren’s-syndrome-related antigen B antibody; Anti-nRNP, anti-nuclear ribonucleoprotein antibody; SLEDAI, systemic lupus erythematosus disease activity index.

MALT1 expression and its link to LN risk

MALT1 in LN patients was more increased than that in non-LN SLE patients [2.26 (IQR, 1.38–3.54) vs. 1.04 (IQR, 0.60–1.62); P<0.001; see ]. Additionally, MALT1 was positively correlated with LN risk and had an AUC of 0.811 (95% CI: 0.705–0.917), while MALT1 expression was 1.860 at the best cut-off point and had a sensitivity of 63.3% and a specificity of 83.3% (see ).

Figure 1

MALT1 was more upregulated in LN patients than non-LN SLE patients. Comparison of MALT1 between LN and non-LN SLE patients (A). The value of MALT1 in distinguishing LN patients from non-LN SLE patients (B). LN, lupus nephritis; SLE, systemic lupus erythematosus; MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; AUC, area under the curve.

To further validate the correlation between MALT1 and LN risk, a multivariate logistic regression model analysis was conducted, which showed that higher MALT1 was independently related to high LN risk [OR: 3.097, 95% CI: 1.292–7.425; P=0.011; see ]. Additionally, higher Scr was also independently linked to elevated LN risk (OR: 1.055, 95% CI: 1.012–1.099; P=0.011). Subsequently, MALT1 and Scr were used to develop a LN risk estimation nomogram (see ), which had a good value for estimating LN risk and had an AUC of 0.873 (95% CI: 0.787–0.960; see ). The calibration plots also disclosed good consistency between the observed probabilities and the nomogram’s predictions in relation to LN risk (see ).

Table 2

Factors affecting the risk of LN in all LN patients by a multivariate logistic regression model analysis

Items	P value	OR	95% CI
			Lower	Upper
Higher MALT1	0.011	3.097	1.292	7.425
Higher Scr	0.011	1.055	1.012	1.099

LN, lupus nephritis; OR, odds ratio; CI, confidence interval; MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; Scr, serum creatinine.

Figure 2

MALT1-based nomogram for predicting LN risk. The proposed nomogram (A) and its ROC curve in predicting LN risk (B). The calibration plot of the proposed nomogram (C). MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; LN, lupus nephritis; ROC, receiver-operating characteristic; AUC, area under the curve.

Comparison of MALT1 with different LN classifications

MALT1 did not differ among LN patients with class II, class III, class IV, class V, class V + III, and class V + IV (P=0.131; see ). Further, considering that class V LN might occur in combination with class III or IV, we mixed the class III patients with the class V + III patients, and the class IV patients with the class V + IV patients, and found that MALT1 varied among class II, III (accompanied by V or not), IV (accompanied by V or not), and V patients (P=0.046; see ). Further, MALT1 was highest in class IV (accompanied by V or not) patients, followed by class V and class III (accompanied by V or not) patients, and lowest in class II patients.

Figure 3

MALT1 varied among patients with different LN classifications. Comparison of MALT1 among classes II, III, IV, V, V + III, and V + IV LN patients (A). Comparison of MALT1 among LN patients with classes II, III (accompanied by V or not), IV (accompanied by V or not), and V (B). MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; LN, lupus nephritis.

Association of MALT1 with renal injury and inflammation in LN patients

MALT1 was positively linked to the LN activity index in LN patients (rs=0.370, P=0.044; see ). However, while MALT1 was positively associated with the LN chronicity index, this did not reach statistical significance in LN patients (rs=0.341, P=0.066; see ).

Figure 4

MALT1 was positively correlated with the LN activity index in all LN patients. Association of MALT1 with the LN activity index (A) and the LN chronicity index (B). MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; LN, lupus nephritis.

In relation to the renal function indexes, MALT1 was positively linked to 24 h proteinuria (rs=0.473, P=0.008; see ) and Scr (rs=0.378, P=0.039; see ) in LN patients. It also showed a positive correlating trend (which lacked statistical significance) with BUN (rs=0.350, P=0.058; see ). In relation to the inflammation indexes, MALT1 was positively associated with C-reactive protein (CRP) (rs=0.410, P=0.025; see ) and showed a positive association trend (without statistical significance) with the erythrocyte sedimentation rate (ESR) (rs=0.299, P=0.108; see ).

Figure 5

MALT1 was positively correlated with 24 h proteinuria, Scr, and CRP in all LN patients, but not correlated with BUN or ESR. Association of MALT1 with 24 h proteinuria (A), Scr (B), BUN (C), CRP (D), and ESR (E) in LN patients. MALT1, mucosa-associated lymphoid tissue lymphoma translocation protein 1; Scr, serum creatinine; BUN, blood urea nitrogen; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; LN, lupus nephritis.

Discussion

MALT1, which is known for its proteolytic activity and scaffold function, is an important signal-transmitting protein that modulates immune response and inflammation in some autoimmune diseases (21-23). For example, a previous study showed that MALT1 binds to BCL-10 and they form a complex, whose aberrant expression leads to psoriasis or inflammatory bowel disease (IBD) (21). Another study reported that MALT1 is more upregulated in IBD patients than healthy controls, and that MALT1 is positively correlated with the disease risk of IBD (22). The current study showed that MALT1 in LN patients was more elevated than that in non-LN SLE patients, and higher MALT1 was independently related to an increased risk of LN. Some possible explanations include that: (I) MALT1 formed the CARD9-BCL10-MALT1 (CBM) complex, which induces the over-production of ICs; and IC deposition is the main etiology of LN (24-26); thus, higher MALT1 is independently associated with elevated LN risk; and (II) MALT1 induces Ang II, which serves as an important renin-angiotensin and accelerates kidney damage (14,27). Thus, higher MALT1 is independently linked to an increased risk of LN.

In addition to the correlation of MALT1 with LN risk, this study also observed that MALT1 is positively linked to the LN activity index, 24 h proteinuria, and Scr in LN patients, which implies that MALT1 is positively associated with LN severity. It could be that MALT1 facilitates the secretion of pro-inflammatory cytokines and interferon production by triggering the nuclear factor-κB (NF-κB) signaling pathway, which aggravates kidney injury and elevates the disease severity of LN (28,29). Thus, upregulated MALT1 is related to increased LN severity. We also found that MALT1 is highest in class IV (accompanied by V or not) patients, followed by class V and class III (accompanied by V or not) patients, and lowest in class II patients. This may be because the ICs in the class IV or V + IV patients involved ≥50% of the glomerulus [with or without glomerular basement membrane (GBM) involvement] (30). Conversely, in class V patients, the GBM was involved by ICs; and ICs in class III or V + III patients involved <50% of the glomerulus (with or without GBM involvement) (31). Further, the ICs were purely involved in the mesangial matrix in class II patients (32). Additionally, as mentioned above, MALT1 is positively linked to ICs in LN patients. Thus, MALT1 is highest in class IV (accompanied by V or not) patients, followed by class V and class III (accompanied by V or not) patients, and lowest in class II patients.

As to the association of MALT1 with inflammation, a previous study showed that MALT1 could develop multi-organ inflammation due to its dual scaffold and proteolytic functions (33). Moreover, another study revealed that MALT1 serves as a pro-inflammation factor in psoriasis, which is positively correlated with interleukin (IL)-1β, interferon-gamma, etc. (12). Similarly, this study showed that upregulated MALT1 is associated with increased CRP in LN patients, which implies that MALT1 is linked to elevated inflammation in LN patients. This may be because MALT1 promotes pro-inflammatory cytokine production, which leads to inflammation flare in LN; thus, MALT1 is correlated with upregulated inflammation in LN patients (34,35).

The current study had some limitations. First, the sample size was relatively small. Thus, large-scale studies need to be conducted to validate these findings. Second, according to previous studies, MALT1 is closely linked to T cells, natural killer cells, and B cells in some autoimmune diseases, but their relationship to LN remains unknown (36-38). In future studies, health subjects should be enrolled as health controls. The value of MALT1 as a prognostic factor in LN patients is also worthy of further study.

In conclusion, MALT1 might serve as a useful biomarker to estimate the disease risk of LN; however, in the future, large-sample size studies need to be conducted to validate our findings.

The article’s supplementary files as