A proteomic atlas of kidney amyloidosis provides insights into disease pathogenesis

Charalampous Charalampos; Dasari Surendra; McPhail Ellen; Theis D Jason; Vrana A Julie; Dispenzieri Angela; Leung Nelson; Muchtar Eli; Gertz Morie; Ramirez-Alvarado Marina; Kourelis Taxiarchis

doi:10.1016/j.kint.2023.11.023

. Author manuscript; available in PMC: 2025 Mar 1.

Published in final edited form as: Kidney Int. 2023 Dec 12;105(3):484–495. doi: 10.1016/j.kint.2023.11.023

A proteomic atlas of kidney amyloidosis provides insights into disease pathogenesis

Charalampous Charalampos ¹, Dasari Surendra ², McPhail Ellen ³, Theis D Jason ³, Vrana A Julie ³, Dispenzieri Angela ¹, Leung Nelson ¹, Muchtar Eli ¹, Gertz Morie ¹, Ramirez-Alvarado Marina ⁴, Kourelis Taxiarchis ¹

PMCID: PMC10922603 NIHMSID: NIHMS1961897 PMID: 38096952

Abstract

The mechanisms of tissue damage in kidney amyloidosis are not well described. To investigate this further, we used laser microdissection-mass spectrometry to identify proteins deposited in amyloid plaques (expanded proteome) and proteins over expressed in plaques compared to controls (plaque-specific proteome). This study encompassed 2650 cases of amyloidosis due to light chain (AL), heavy chain (AH), leukocyte chemotactic factor-2-type (ALECT2), secondary (AA), fibrinogen (AFib), apo AIV (AApoAIV), apo CII (AApoCII) and 14 normal/disease controls. We found that AFib, AA, and AApoCII have the most distinct proteomes predominantly driven by increased complement pathway proteins. Clustering of cases based on the expanded proteome identified two ALECT2 and seven AL subtypes. The main differences within the AL and ALECT2 subtypes were driven by complement proteins and, for AL only, 14–3-3 family proteins (a family of structurally similar phospho-binding proteins that regulate major cellular functions) widely implicated in kidney tissue dysfunction. The kidney AL plaque-specific proteome consisted of 24 proteins, including those implicated in kidney damage (α1 antitrypsin and heat shock protein β1). Hierarchical clustering of AL cases based on their plaque-specific proteome identified four clusters, of which one was associated with improved kidney survival and was characterized by higher overall proteomic content and 14–3-3 proteins but lower levels of light chains and most signature proteins. Thus, our results suggest that there is significant heterogeneity across and within amyloid types, driven predominantly by complement proteins, and that the plaque protein burden does not correlate with amyloid toxicity.

Keywords: Amyloidosis, Proteomics, Protein aggregation

Introduction

Renal amyloidosis is a heterogeneous disease caused by the deposition of several different types of amyloidogenic proteins¹. Immunoglobulin light chain (AL) is the most common type of renal amyloidosis, and the goal of therapy is to control the underlying plasma cell clone. For other amyloid types, such as AA, in which the precursor protein is serum amyloid A (SAA), treatment focuses on controlling the inflammatory disease driving amyloid deposition. Despite this, renal function deterioration can continue even after optimal control of the underlying cause. Finally, for other types, such as ALECT2 (leukocyte chemotactic factor-2-type), no disease-specific therapies exist. To develop novel therapies, we need to better understand the mechanisms of tissue damage in this disease.

The renal amyloid plaque proteome is an untapped source of information to provide insights into the mechanisms of tissue damage. Our center has extensive expertise in amyloid typing using laser microdissection (LMD) and mass spectrometry (MS) ^2–6, during which amyloid plaques are microdissected and analyzed by MS. In addition to the amyloidogenic protein, several other proteins are also identified that could inform disease physiology. For instance, the amyloid signature proteins (clusterin, vitronectin, ApoE, ApoA4, serum amyloid P-component-SAP), are seen with every amyloid type and across all tissues. Some (e.g., clusterin) can act as amyloid chaperones⁷, inhibiting amyloidogenicity, whereas others (e.g., SAP) can promote amyloidogenicity. In addition, several proteins of the involved tissue are represented within the plaque proteome. Although amyloid plaques do not consist of live cells, their proteome can provide information about amyloid burden, mechanisms of cellular toxicity, or physiologic tissue responses to amyloid deposition. This study aimed to comprehensively characterize the renal amyloid proteome and its correlation with patient characteristics and outcomes.

Methods

This study was approved by Mayo Clinic’s IRB (07–000988). Since most cases included were from outside institutions, the research could not have been completed otherwise, and the nature of the project posed minimal risk to the patients, a waiver of consent was granted. All newly diagnosed internal and external cases of the most common types submitted to Mayo Clinic for typing between 2008–2020 were included. A case had to be Congo red positive to be analyzed. We also included 9 disease controls (membranous nephropathy) with diagnostic biopsies and 5 normal controls (day zero renal transplant biopsy) between 2008–2013. Our methodology has abided by the Declaration of Helsinki and the Declaration of Istanbul guidelines. Peptide identification was performed using LMD/MS ³ of Congo red-positive amyloid deposits in the case of amyloidosis and glomeruli in control cases. All cases were analyzed by MS within approximately a week of receiving the biopsy sample. The amyloid typing assay is a CAP/CLIA-validated assay. There is stringent, periodic quality control to ensure consistent proteomic outputs from the instrument. There was a single change in MS instrumentation during the study period. This change was meticulously validated by comparing the old and new proteomes of hundreds of samples and proving that the new method change has not altered the proteomic output (a requirement for CAP/CLIA). Therefore, batch effects were not a concern despite the retrospective nature of the study. Protein spectral counts, normalized to the total number of spectral counts per LMD, were used as a semiquantitative measure of abundance. The existing renal staging system⁸ was used to risk-stratify patients with AL amyloidosis.

Statistical and bioinformatic analyses

Statistical analyses were performed using JMP (SAS Institute Inc., Cary, North Carolina) and the omiq.ai online platform. All identified proteins were considered part of the expanded amyloid proteome, and comparisons across groups considered a false detection rate (FDR) corrected p-value of <0.05 except protein comparisons and clinical parameters when considering the plaque-specific proteome where a p-value of <0.05 was considered significant, given the limited number of comparisons. Proteins were considered part of the plaque-specific proteome if their abundance in the amyloid plaque was increased by 50% compared to pooled normal and disease controls using an FDR-corrected p-value <0.05 for consistency with prior work ⁹. For dimensionality reduction and visualization, we used Uniform Manifold Approximation and Projection (UMAP)¹⁰, and to cluster the UMAP dataset, we used phenograph ¹¹ with the expanded proteome as input features for both methods and default parameters within omiq.ai. To cluster AL patients using their plaque-specific proteome, we used hierarchical clustering in JMP using ‘Ward’s minimum variance method to calculate distances between clusters. Overrepresentation analyses were performed in WebGestalt¹² and using the Reactome database and an FDR-corrected p < 0.05 when identifying significantly overrepresented pathways. Patient overall and renal survival (OS and RS, respectively) were calculated from the time of diagnosis to the time of death or initiation of dialysis, respectively, using the Kaplan-Meier method. For renal survival, the composite endpoint of transition to hemodialysis or patient death was considered. A Cox proportional hazards model was used to determine outcome differences. The Fisher exact test was used to compare categorical variables, and the Wilcoxon rank-sum was used for continuous variables.

Results

We included a total of 2650 renal amyloidosis cases. The distribution of the cases is shown in table 1. Since most cases were referred to our institution for amyloid typing, no other clinical information was available in our database except for a subset of AL patients.

Table 1.

Amyloid cases included in the study

Disease type	N =2650	(%)	Female sex, N (%)^a	Age, median (range)¹
Immunoglobulin	1618	35.52	640 (40%)	66 (25–93)
Lambda	1075	25.68	417 (39%)	66 (25–90)
Kappa	412	9.84	165 (40%)	66 (32–93)
IGH	73	1.74	36 (49%)	65 (34–92)
AH	58	1.39	22 (38%)	70 (41–88)
ALECT2	474	11.32	220 (46%)	67 (18–92)
AA	418	9.99	186 (44%)	60 (20–94)
AFib	80	1.91	26 (33%)	63 (34–81)
AApoAIV	38	0.91	9 (24%)	69 (47–86)
AApoCII	22	0.53	14 (64%)	70 (53–98)
Controls	N=14
Normal Controls	5	36%	2 (40%)	52 (51–67)
Membranous nephropathy controls	9	64%	2 (22%)	55 (48–86)

Open in a new tab

Fifty-seven patients (2%) without available gender/age information.

IGH: mix of heavy and light chain amyloid, AH: heavy chain amyloidosis, ALECT2: LECT2 amyloidosis, AA: secondary amyloidosis, Afib: fibrinogen amyloidosis, AApoAIV: apo A-IV amyloidosis, AApoCII: Apo C-II amyloidosis.

The total protein content varies across amyloidosis types

We identified 498 distinct proteins across all types. The most overrepresented pathway was the complement cascade (supplemental figure 1). AL and AApoAIV (apolipoprotein A-IV amyloidosis) had the lowest proteomic burden (defined as the total abundance of all identified proteins in normalized spectral counts), whereas AFib (fibrinogen amyloidosis) and AApoCII (apolipoprotein C-II amyloidosis) the highest (supplemental figure 1). Compared to the previously published cardiac proteome⁹, the renal proteome appeared more diverse (498 versus 161 distinct proteins). Pathway analyses of proteins unique to the heart/kidney or common across organs are shown in supplemental figure 2. Protein pathways increased in both organs reflected processing of lipoproteins (e.g., uptake of ligands by scavenger receptors)¹³, fibrosis pathways (e.g., “ECM proteoglycans”, “degradation of the extracellular matrix”), and keratinization pathways, which reflect patterns of organ damage shared across the 2 organs^14,15. Pathways unique to heart tissue were muscle contractility and keratinization pathways. Pathways unique to the kidneys included the complement cascade and pathways driven by high tubulin superfamily protein expression (e.g., all top 4 pathways for renal-specific proteins, supplemental figure 2), which reflect distinct physiologic responses to renal damage^16,17. Of note, complement proteins C3, C4B, and C9, as well as the complement regulatory proteins CFH, CFHR1 were present in both organs, suggesting that the complement pathway was activated from early to late components in amyloid plaques of hearts and kidneys across various amyloid types. The top 3% most abundant proteins in the renal proteome (excluding the amyloidogenic protein) across all types included amyloid signature proteins^18,19, serum, and structural proteins (supplemental figure 3). Signature proteins accounted for 19% of the proteomic abundance of renal plaques (IQR 12.6%−25%). APOAIV had the largest relative expression of signature proteins despite the lowest overall proteomic count, whereas AH/ALH had the lowest (supplemental figure 4). The abundance trends for each signature protein were similar across proteomic subtypes, with two exceptions. Kappa-restricted AL had the highest clusterin abundance, and APOAIV was nearly absent in ALECT2 (supplemental figure 4). Similar relative signature protein abundance (median 17%, IQR 2–55%) was seen in cardiac amyloid plaques⁹. These data suggest that the protein deposition dynamics for signature proteins are decoupled to that of other proteins in amyloid plaques but are preserved across tissues and that AL-K has high levels of protective chaperone clusterin. To understand the major drivers of variability across the entire dataset, we performed principal component (PC) analysis (PCA). PC 1 through 5 explained 20.5% of the variability in the data (not shown). The top 5% (25) of proteins with the highest absolute loading values for each of the first 5 principal components were involved in pathways relating to apoptotic cleavage of cell adhesion proteins (e.g., DSG1 and PKP1) and laminin family proteins, likely reflecting the differential breakdown of structural integrity caused by the various amyloid types.

AFib, AA, and AApoCII have distinct proteomes

To appreciate how the different amyloid types relate to one another based on their expanded proteomes, we used UMAP to visualize this dataset. This analysis positioned AFib, AA, and AApoCII separately, while the other types seemed to congregate in the center of the map (figure 1). These data suggest that AFib, AA, and AApoCII have very distinct proteomes.

Phenograph identifies 2 ALECT2 and 7 AL proteomic subtypes

We used phenograph to identify proteomic subtypes (figure 1). As expected, AFib, AA, and AApoCII formed distinct clusters in the periphery of the map, as did AApoAIV in the center. However, AL was characterized by significant heterogeneity, forming 7 distinct clusters. The expected distribution of lambda and kappa cases in AL is approximately 70:30 for lambda:kappa. In 4 of the 7 identified AL clusters, this distribution was largely maintained. However, 2 clusters showed lambda predominance (lambda-like 1,2 with 95% and 89% lambda cases, respectively), and one cluster showed kappa predominance (kappa-like, with 99.5% kappa cases). AH and, in some cases, ALECT2 and AA cases were included in some of the AL-rich clusters (supplemental figure 5), which suggests that some ALECT2 and AL/AH cases are characterized by some proteomic overlap.

Complement and collagens are the main proteins differentially expressed across various types

We evaluated what drives the AA, AFib, and AApoCII proteomes to be so distinct (supplemental spreadsheets 1-3 and supplemental figures 6-8). Several complement proteins were increased across all three types, and for AA, fibrinogen proteins were also increased, consistent with a more inflammatory phenotype. Among the most differentially expressed proteins, some structural proteins, such as fibulin (FBLN1), were increased in AA, which could reflect the different amyloid deposition pattern (interstitial/vascular versus glomerular). Others, such as midkine (MDK) in AA, could reflect its more inflammatory phenotype. Collagen proteins were decreased compared to the central clusters. Proteins relating to proteostasis (ubiquitins) were increased in AApoCII only.

We then compared ALECT2 to AL/AH and identified that complement proteins were increased along with various collagen proteins (supplemental figure 9, supplemental spreadsheet 4), whereas various structural proteins, fibrinogen, and 14–3-3 family proteins (gene names YWHA) were decreased. However, the differences in structural/collagen proteins seen here and in the case of AA, AFib, and AApoCII could be explained by the different deposition pattern in ALECT2 (predominantly interstitial). Finally, we compared AApoAIV to AL/AH and identified several apolipoproteins increased in the former, whereas fibrinogen and several collagen proteins were increased in the latter (supplemental figure 10, supplemental spreadsheet 5).

When considering AL/AH amyloidosis, AL had decreased complement pathway proteins and increased collagen proteins (supplemental figure 11, supplemental spreadsheet 6). When comparing kappa to lambda cases, complement pathway proteins were increased, and 14–3-3 family proteins, well-described markers of renal damage ²⁰, decreased (figure 2, table 2). Of note, the top increased protein in kappa cases was DnaJ heat shock protein (Hsp40) family member B9 (DNAJB9), which has been implicated in the development of fibrillary glomerulonephritis ^21–23, suggesting that this protein may be implicated in other diseases where immunoglobulin-based fibrils are present. After lambda immunoglobulin genes, plasminogen (PLG) was the most increased protein in lambda cases and has been implicated in renal fibrosis by promoting epithelial-to-mesenchymal transition ²⁴.

Figure 2. — For kappa, we found increased complement activation and decreased collagen deposition, as well as markers of renal damage

Table 2.

List of proteins differentially expressed between Kappa vs. Lambda AL amyloidosis^a

Protein (gene symbol)	Log2 fold change	Protein (gene symbol)	Log2 fold change	Protein (gene symbol)	Log2 fold change
DNAJB9	7.766106	ALB	0.42499	COL4A3	−0.51984
IGKC	4.801048	H2BC1	0.412236	YWHAQ	−0.50284
APOB	3.837124	HSPG2	0.326402	PRDX1	−0.50087
C7	2.186334	APOE	0.297083	FLNA	−0.48534
C8A	2.141322	VTN	0.296005	SERPINA1	−0.4827
C5	1.999357	C3	0.251701	TPM2	−0.45821
C8B	1.596573	IGLC7	−7.45081	APCS	−0.45001
SPP2	1.23396	IGLC1	−4.83142	YWHAB	−0.44653
IGHG4	1.230476	IGLC3	−4.29387	COL3A1	−0.39656
CXCL14	1.189888	IGLC3	−3.94362	EEF1A1	−0.39427
C6	1.161839	IGLC2	−3.8245	HSPB1	−0.38764
APOA2	1.158891	PLG	−1.58155	COL1A2	−0.34061
TTR	1.110925	ATP5F1A	−1.17429	EEF1A1P5	−0.32894
AMBP	1.107204	GATM	−1.12204	COL1A1	−0.32412
CLU	1.027796	MLRN	−1.03058	MYL6	−0.28277
FN1	0.863618	LAC3	−1.02334	APOA4	−0.26459
IGHG1	0.842007	ATP5F1B	−0.99087	LMNA	−0.2545
CFHR5	0.817226	SFN	−0.74861	H3-4	−0.23036
NPNT	0.785676	TPM1	−0.66684	H4C1	−0.22728
TF	0.757747	MYH11	−0.65196	H1-4	−0.21333
MFGE8	0.749248	TAGLN	−0.64567	H3C13	−0.21215
C9	0.669253	ACTA2	−0.57452	H3C1	−0.2106
MDK	0.65542	YWHAG	−0.57195	H2BC13	−0.16476
FGA	0.623313	COL4A4	−0.52514	VIM	−0.10963

Open in a new tab

Proteins in green are increased and in red decreased in kappa compared to lambda cases. A false detection rate (FDR) corrected p value of <0.05 was considered significant for all proteins.

We then focused on the difference between the 2 ALECT2 subtypes (cluster #9 versus #10 in figure 1). Pathways overrepresented within the increased proteins were related to the complement pathway and collagen formation/degradation (supplemental figure 12, supplemental spreadsheet 7). Pathways overrepresented within the decreased proteins included common serum proteins such as albumin, hemoglobin, haptoglobin, transferrin, serum amyloid A-4 protein (SAA4), protein S100-A8, which may reflect a loss of these proteins in the renal interstitium in cluster 9 or an acute phase reaction in situ in cluster 10. Of note, alpha-1-acid glycoprotein 1 (ORM1) was the second most increased protein in cluster 10 after SAA4; it has been associated with protection from renal inflammation and fibrosis ^25,26 and could represent a compensatory mechanism. DNAJB9 was again noted as one of the most increased in cluster 9.

Having described major differences between kappa and lambda AL cases, we evaluated potential differences between the mixed AL-like subclusters. We considered the mixed AL-like clusters 4,5,7 together since they congregated closer in the UMAP (figure 1) and compared them with the more distal, mixed AL-like cluster 6. Complement, collagen, and keratin proteins, well-characterized markers of renal epithelial damage, were overrepresented in clusters 4,5,7 ²⁷. Laminin family and cytoskeletal proteins, heat shock proteins (HSPs), which are thought to have a protective role in various forms of renal injury ²⁸, and tubulin family proteins (supplemental figure 13, supplemental spreadsheet 8) were decreased. Of note, Clusters 4, 5, and 7 were twice as likely to have renal stage 3 compared to cluster 6 (34% versus 15%, respectively, p=0.03), which suggests that HSP \ activation may have protective effects in a subset of AL cases.

Finally, to evaluate the impact of deposition patterns (glomerular versus interstitial versus perivascular), we evaluated AL cases that had pathology information available and identified 108 cases with glomerular but no interstitial involvement, as well as 32 cases with glomerular but no perivascular involvement and 11 cases with perivascular but no glomerular involvement and compared their proteomes. Only 12 proteins were significantly increased in cases with interstitial involvement (supplemental spreadsheet 9). Reassuringly, only collagen alpha-1,2(I) overlapped with some of the proteins seen differentially expressed in the comparisons above. No significant differences were identified between glomerular and perivascular cases, possibly due to the limited number of samples.

The plaque-specific proteomes of renal AL and AH are highly restricted

Since many of the proteins identified in amyloid plaques could be part of normal tissue background or non-specific nephrotic syndrome markers, we compared the expanded proteomes of AL, AH, AApoCII, and AFib to pooled cases from normal controls and from patients with membranous nephropathy. We limited ourselves to these subtypes because of their predominantly glomerular distribution, where LMD samples were taken from controls. The plaque-specific proteome of AL consisted of 24 proteins (Table 3). The AH plaque proteome was identical to that of AL except for AHNAK, Immunoglobulin heavy constant mu (IGHM), and Secreted Phosphoprotein 2 (SPP2), which were unique to AH, and the absence of apolipoprotein A-I (APOA1), Myosin Heavy Chain 11 (MYH11), H1.0 Linker Histone (H1–0), immunoglobulin lambda constant 3 (IGLC3) and Metallopeptidase Inhibitor 3 (TIMP3) which were unique to AL (supplemental table 1). AFib and AApoCII had a less restricted plaque proteome with a total of 51 and 43 identified proteins, respectively (supplemental spreadsheets 10,11), with a predominance of complement pathway proteins (supplemental figures 14,15). When considering proteins decreased in AL or AH compared to controls (supplemental figure 16), ubiquitin-related proteins were among the most decreased in AL compared to controls, suggesting impaired protein processing in AL (supplemental spreadsheet 12). For AFib and AApoCII, heat shock and collagen proteins were decreased (supplemental spreadsheet 10,11 supplemental figure 14,15).

Table 3.

The plaque-specific proteome of AL amyloidosis (proteins increased in AL versus normal/disease controls)

Protein	Function in kidneys	Log2 Fold Change	FDR corrected P value
*COL1A2* Collagen alpha-2(I) chain	Collagen protein	22.74	<0.0001
*COL1A1* Collagen alpha-1(I) chain	Collagen protein	22.43	<0.0001
*APOA4* Apolipoprotein A-IV	Signature protein	22.35	<0.0001
*TAGLN* Transgelin	Implicated in several types of kidney disease.⁵⁹	21.92	0.009
*COL3A1* Collagen alpha-1(III) chain	Collagen protein	21.82	<0.0001
TIMP3 Metallopeptidase Inhibitor 3	Implicated in several types of kidney disease.^60,61	21.59	0.008
*H1-0* Histone H1.0	Unclear	21.52	0.02
C6 Complement component 6	Complement activation	21.5	0.03
*MYH11* Myosin heavy chain 11	Structural	21.45	0.02
*HSPB1* Heat shock protein beta-1	Increased in response to renal insults. Chaperone activity. Protects from complement-mediated renal epithelial injury.^48,62	21.23	0.005
*USH2A* Usherin	Unclear	20.94	0.0007
*TMSB4X* Thymosin beta 4	Implicated in several types of kidney disease.⁶³	20.6	0.01
*MSN* Moesin	Implicated in several types of kidney disease.^64,65	20.4	0.02
*IGLC3* Immunoglobulin lambda constant 3	Amyloidogenic protein	5.45	0.0003
*SERPINA1* Alpha-1 antitrypsin	Marker of renal damage in other nephrotic syndromes⁶⁶	5.24	0.02
*COL6A1* Collagen alpha-1(VI) chain	Collagen protein	4.64	0.04
*LAC3* Laccase-3	Unclear.	4.61	0.007
*APCS* Serum amyloid P-component	Signature protein	4.54	<0.0001
*APOE* Apolipoprotein E	Signature protein	4.39	<0.0001
*VTN* Vitronectin	Signature protein	4.06	<0.0001
*APOA1* Apolipoprotein A-I	Unclear	3.79	0.02
*CLU* Clusterin	Signature protein	3.56	<0.0001
*S100A6* Protein S100-A6	Renal damage marker.⁴⁹	3.22	<0.0001
*YWHAB* Tyrosine 3-Monooxygenase/T ryptophan 5-Monooxygenase Activation Protein Beta	Regulates large spectrum of signaling pathways, including several implicated in kidney disease.²⁰	2.9	0.04

Open in a new tab

The plaque-specific proteome of kappa AL is different from that of lambda AL

Kappa cases had a higher proteomic content than lambda (supplemental figure 17), including higher (p<0.001 for all) amyloidogenic, complement C6, and signature proteins except for serum amyloid P-component and ApoA4. Plaque-specific proteins significantly (p<0.001 for all) lower in kappa included: Tyrosine 3-Monooxygenase/Tryptophan 5-Monooxygenase Activation Protein Beta (YWHAB), Alpha-1 antitrypsin, Laccase-3, IGLC3, MYH11, transgelin, and all the collagen proteins except for COL6A1 (not shown).

The plaque-specific and expanded proteomes of AL differ according to renal stage and gender

Of 2650 cases, only 353 were evaluated at our institution and had some clinical information available. Of these, we analyzed the 228 renal AL patients with complete clinical follow-up and adequate demographic and clinical data available (supplemental table 2). Regarding the plaque-specific proteome, and despite similar overall proteomic content, we found that YWHAB and H1–0, and moesin (MSN) were higher in patients with stage 1/2 disease versus 3, proteins associated with amyloid fibrils in other types of amyloid ^29,30 (FDR p<0.05 for all, not shown). Clusterin was higher in patients with stage 3. In the expanded proteome, patients with renal stage 3 had higher levels of fibrinogen and alpha-2-microglobulin (figure 3, supplemental spreadsheet 13), all markers of progressive renal dysfunction^31–33, and higher levels of complement pathway proteins. Despite a similar protein burden, several proteins were increased in female versus male patients, including several histone proteins, complement C8b, and clusterin (supplemental spreadsheet 14). Except for complement, no other pathways were significantly over/under-represented (not shown). It is unclear if these are amyloid-specific differences or generally observed differences between males and females, although complement proteins are generally higher in males ³⁴.

Figure 3. — For renal stage ½, we found increased cytoskeletal proteins and decreased collagen activation, as well as markers of renal damage

A plaque-specific proteomic signature is associated with improved renal survival

To identify patterns of plaque-specific deposition, we used hierarchical clustering to group AL cases with available clinical data based on their plaque-specific proteome. We identified 4 major clusters (figure 4). The distribution of the expanded proteome clusters (figure 1) within the amyloid-specific proteome clusters (figure 4) is shown in supplemental figure 18. AL-like-3 cluster (expanded proteome) and Lambda-like-2 cluster (expanded proteome) were overrepresented in cluster 1 (33.1% versus 16.6%, p<0.001, and 44.8% versus 8%, p<0.001, respectively). The major differences in (amyloid-specific) protein abundance are shown in supplemental figure 19a-c. Clusters 1 and 4 had the highest protein abundance but the lowest light chain deposition, suggesting that the dynamics of the light chain deposition differ from those of the total proteome, similar to what we have shown in the heart⁹. We explored whether any of these groups correlated with different renal outcomes and found that cluster 1 was associated with improved renal survival compared to the rest (Figure 5). In a multivariable analysis (not shown) that includes cluster 1 status and renal stage, this effect was not independent (p>0.05) of renal stage, and indeed cluster 1 was more likely to have a lower renal stage (97% renal stage 1 or 2 versus 72%, p=0.001). Cluster 1 had the highest levels of YWHAB, H1–0, MSN, S100A6, Thymosin beta-4 (TMSB4X), and Usherin (USH2A) and the lowest levels of TIMP3 and vitronectin. Collagen proteins and TIMP3 clustered predominantly in clusters 3 and 4 and transgelin (TAGLN) in cluster 4, whereas complement 6 was deposited at very low levels in cluster 4 (figure 4, supplemental figure 19b, c). When considering the expanded proteome of cluster 1 compared to the other clusters, 14–3-3 family proteins (YWHA) were increased, whereas collagen proteins were decreased (supplemental figure 20, supplemental spreadsheet 15). TMSB4X was the most increased protein in cluster 1, and its loss has been implicated in accelerated renal fibrosis. ³⁵ Galectin-1 (LGALS1) was among the highest expressed proteins and has anti-ischemic and anti-inflammatory effects in the kidney ^36
37. These data suggest that proteins relating to matrix remodeling (collagens, TIMP3) deposit in different patterns and that lower levels of matrix fibrosis may be responsible for the improved renal survival seen in cluster 1.

Figure 4. — We identified 4 major clusters. Cluster one had increased proteomic burden and increased levels of 14–3-3 family proteins

Figure 5. — Cluster 1 had improved renal survival, compared to other proteomic clusters. **RS:** Renal survival

Discussion

This is the first study to comprehensively examine the proteomic content of the amyloidosis plaque in the kidney. Key findings include that the more indolent (non-AL) amyloid types have a higher proteomic content than AL. The complement cascade was the most overrepresented pathway across all types, and responsible for most of the differences across various disease types. We show that AA, AFib, and AApoCII are distinct proteomically and that AL and ALECT2 are the most heterogeneous. Finally, we show that a subset of AL patients with high proteomic content but low amounts of light chains and proteins involved in renal fibrosis has the best renal survival.

Similar to what we have shown for cardiac ATTR compared to AL, the more rapidly evolving AL had the lowest plaque proteomic content, which suggests that the proteomic “density” of amyloid plaques may be a surrogate of more indolent disease progression ⁹. In addition, patients with AL and worse renal outcomes had higher levels of amyloidogenic light chains but lower overall proteomic content, suggesting that the deposition dynamics of amyloidogenic proteins are decoupled from that of other proteins. Kappa AL cases, associated with improved survival compared to lambda ³⁸, also had a higher proteomic burden than lambda and higher levels of the protective chaperone clusterin.

Complement pathway proteins were major sources of variability across various types and their proteomic subgroups. They were also increased in all amyloid cases when compared to normal and disease controls. These data suggest that complement deposition is an amyloid-specific process seen in most amyloid types. In addition to systemic deposition, our group has previously shown upregulation of complement genes after exposure to AL fibrils in cultured cardiomyocytes in vitro. ³⁹. This suggests that complement may be produced in situ and represent a form of an atypical response to pathogen. It is unclear to what extent complement activation is responsible for renal tissue damage.

We noted a relative decrease of ubiquitins and heat shock proteins in AL, AH, AApoCII, and AFib patients compared to controls. Indeed, amyloid proteins cause a reduction of free ubiquitin levels in vitro models of ‘Alzheimer’s disease. ⁴⁰ This may reflect the overutilization of these proteins by the ubiquitin-proteasome system. The same may be true for heat shock proteins, a well-described group of “anti-amyloid” chaperones that could be amenable to pharmacologic interventions.⁴¹ Ubiquitins and heat shock proteins can be directly sequestered within misfolded protein aggregates, further contributing to the cell’s proteostatic collapse^42–44.

14–3-3 family proteins were increased in AL cases overall and, in lambda cases in particular, and in cluster 1 of renal AL, which had improved renal outcomes. 14–3-3 proteins are increased in animal models of ureteric obstruction, IgA, and membranous nephropathy and act as molecular chaperones for several other proteins involved in a variety of cellular processes such as signal transduction, proliferation, differentiation, apoptosis, and autophagy ⁴⁵. They have been shown to directly interact with other amyloid types where they promote amyloid clearance ^46,47.

The AL-specific proteome was enriched in some proteins of interest to renal pathologies, such as HSPB1, S100A6, TAGLN, and TIMP3. HSPB1 increases and induces autophagy in acute kidney injury ⁴⁸, which is also induced in response to misfolded proteins. S100A6, a calcium-binding protein, is increased in acute kidney injury and may decrease beta-amyloid deposition in animal models of Alzheimer’s disease ^49,50. TAGLN interacts with cytoskeletal proteins and is induced by amyloid precursor protein in Alzheimer’s disease ⁵¹. Finally, TIMP3, also a component of the cardiac amyloid proteome, could be related to matrix remodeling and fibrosis in AL ⁹.

We found that proteins associated with the protection from oxidative stress were increased in the early stages. Glutathione-S Transferase (GST) and phosphatidylethanolamine-binding protein 1 (PEB1) were the most upregulated proteins in the proteome of early-stage amyloidosis. Both proteins act as antioxidants, with PEB1 being a major regulator of the ferroptosis pathway ^52,53. Studies from cardiomyocytes demonstrate that AL fibrils induce cell death through oxidative stress^54,55, which suggests an organ-wide mechanism of cellular toxicity. On the other hand, we did not note proteins associated with autophagy, unlike what we have noted in the heart⁹ and in vitro models⁵⁶, which suggests this is a cardiac-specific mechanism of tissue damage.

Our study had limitations. Our definition of amyloid-specific proteome was conservative. In addition, although LMD/MS has high accuracy and has been extensively validated, the risk of contamination by normal tissue or serum proteins was possible but unlikely. We generally expect renal tissue to have several thousands of proteins identified^57,58, and not the less than 500 proteins identified here. Therefore, if any samples were significantly contaminated by adjacent normal tissue, this would have been reflected in the identified proteomic diversity. Similarly, the lack of proteomic information from adjacent renal tissue containing live cellular material is another limitation. Indeed, the renal proteome presents us only with remnant proteins of dying renal cells that adhere to amyloid plaques. Therefore, they are not fully representative of the complex physiologic responses occurring in the tissue. Many of the identified differences in structural proteins could be attributed to differences in amyloid deposition patterns across the various types (e.g., interstitial, perivascular, glomerular). However, only a limited number of AL cases had complete pathology information and pure glomerular/interstitial/perivascular involvement to compare these compartments proteomically. Finally, we did not have enough clinical data for comparisons for all the amyloid types, and thus, clinical correlations were restricted to AL. Even within this subtype, only a small number of patients had enough clinical information to perform additional analyses based on proteomic subtypes, which limited the statistical power of this study. Our analyses in relationship to some of these outcomes should, therefore be considered exploratory and in need of additional validation.

Supplementary Material

Supplemental spreadsheet. This contains all the proccessed, anonymized proteomic data and clinicolaboratory information required to recreate the study. The unique ID column is a unique identifier linked to the raw proteomic data that have been deposited to zenodo.

NIHMS1961897-supplement-1.xlsx^{(5.3MB, xlsx)}

Supplemental figure 1. Total proteomic burden across types and pathways associated with identified proteins

Supplemental figure 2. Overlap of renal and cardiac amyloid proteomes

Supplemental figure 3. Top 3% (N=15) most abundant proteins (excluding amyloidogenic protein)

Supplemental figure 4. Relative abundance of signature proteins compared to total proteomic count by amyloid type

Supplemental figure 5. Relative proportions of various amyloid types identified within phenograph clusters

Supplemental figure 6. Pathway analyses (reactome) of differentially expressed proteins between the AA cluster and all the central clusters (non-AFib/AApoCII)

Supplemental figure 7. Pathway analyses (reactome) of differentially expressed proteins between the AFib cluster and all the central clusters (non-AA/AApoCII)

Supplemental figure 8. Pathway analyses (reactome) of differentially expressed proteins between the AApoCII cluster and all the central clusters (non-AA/AFib)

Supplemental figure 9. Pathway analyses (reactome) of differentially expressed proteins between the ALECT2 cluster and AL/AH

Supplemental figure 10. Pathway analyses (reactome) of differentially expressed proteins between AApoAIV cluster and AL/AH

Supplemental figure 11. Pathway analyses (reactome) of differentially expressed proteins between the AL and AH amyloidosis

Supplemental figure 12. Pathway analyses (reactome) of differentially expressed proteins between the ALECT2 clusters

Supplemental figure 13. Pathway analyses (reactome) of differentially expressed proteins between the clusters 4,5,7 and cluster 6 in AL

Supplemental Figure 14. Pathway analyses (reactome) of differentially expressed proteins between the AFib vs. normal and disease controls

Supplemental Figure 15. Pathway analyses (reactome) of differentially expressed proteins between the AApoCII vs. normal and disease controls

Supplemental figure 16. Pathway analyses (reactome) of differentially decreased proteins between the AL/AH vs. normal and disease controls

Supplemental figure 17. Total protein abundance, amyloidogenic, and signature protein abundance of kappa AL vs. lambda AL

Supplemental figure 18. Relative proportions of expanded amyloid proteome clusters identified within amyloid-specific clusters

Supplemental figure 19a. Major differences in protein abundance between the four clusters

Supplemental figure 19b. Major differences in protein abundance between the four clusters

Supplemental figure 19c. Major differences in protein abundance between the four clusters

Supplemental figure 20. Pathway analyses (reactome) of differentially expressed proteins between cluster 1 and the remaining AL clusters

Supplemental table 1. The plaque-specific proteome of AH (proteins increased in AH compared to normal/disease controls)

Supplemental table 2. Baseline demographic and clinical data of 228 renal AL patients with clinical information

NIHMS1961897-supplement-2.pptx^{(7MB, pptx)}

Translational Statement.

We present the amyloid proteome of 2650 renal amyloid cases from 8 different types of renal amyloidosis. We find that complement pathway proteins are deposited across all amyloid types. We identify that AA, Afib, and AApoCII are the most distinct proteomically compared to other types and that AL and ALECT2 have significant proteomic heterogeneity. A subset of AL patients with high amyloid plaque proteomic content but low amounts of light chains and proteins involved in renal fibrosis have improved renal survival. Our results suggest that the deposition dynamics of the amyloidogenic are decoupled to that of other proteins and that the renal toxicity in this disease may not be explained merely by amyloid abundance. Exploring the role of key proteins identified herein in the role of disease pathogenesis may serve as a starting point for the development of novel therapies.

Acknowledgments

The authors would like to acknowledge the patients for allowing them to use their biopsy data for research purposes.

Funding

This work was supported by the Paul Calabresi K12 Career Development Award (CA90628–21), the Mayo Clinic Myeloma SPORE, and an Amyloidosis Research Foundation research grant.

Footnotes

Disclosures

The authors have nothing to disclose and no conflicts of interest pertaining to this study.

Publisher's Disclaimer: 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.

Data sharing statement

All raw and proccessed, anonymized proteomic data used for this manuscript have been deposited to zenodo in 9 separate batches:

https://zenodo.org/record/8347058

https://zenodo.org/record/8350517

https://zenodo.org/record/8350765

https://zenodo.org/record/8351623

https://zenodo.org/record/8352367

https://zenodo.org/record/8352939

https://zenodo.org/record/8353759

https://zenodo.org/record/8355053

https://zenodo.org/record/8357105

References

1.Benson MD, Buxbaum JN, Eisenberg DS, et al. Amyloid nomenclature 2020: update and recommendations by the International Society of Amyloidosis (ISA) nomenclature committee. Amyloid. Dec 2020;27(4):217–222. doi: 10.1080/13506129.2020.1835263 [DOI] [PubMed] [Google Scholar]
2.Dasari S, Theis JD, Vrana JA, et al. Amyloid Typing by Mass Spectrometry in Clinical Practice: a Comprehensive Review of 16,175 Samples. Mayo Clin Proc. Sep 2020;95(9):1852–1864. doi: 10.1016/j.mayocp.2020.06.029 [DOI] [PubMed] [Google Scholar]
3.Vrana JA, Gamez JD, Madden BJ, Theis JD, Bergen HR 3rd, Dogan A. Classification of amyloidosis by laser microdissection and mass spectrometry-based proteomic analysis in clinical biopsy specimens. Blood. Dec 3 2009;114(24):4957–9. doi: 10.1182/blood-2009-07-230722 [DOI] [PubMed] [Google Scholar]
4.Sethi S, Theis JD, Leung N, et al. Mass spectrometry-based proteomic diagnosis of renal immunoglobulin heavy chain amyloidosis. Clin J Am Soc Nephrol. Dec 2010;5(12):2180–7. doi: 10.2215/cjn.02890310 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Vrana JA, Theis JD, Dasari S, et al. Clinical diagnosis and typing of systemic amyloidosis in subcutaneous fat aspirates by mass spectrometry-based proteomics. Haematologica. Jul 2014;99(7):1239–47. doi: 10.3324/haematol.2013.102764 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Sethi S, Vrana JA, Theis JD, et al. Laser microdissection and mass spectrometry-based proteomics aids the diagnosis and typing of renal amyloidosis. Kidney Int. Jul 2012;82(2):226–34. doi: 10.1038/ki.2012.108 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Poon S, Easterbrook-Smith SB, Rybchyn MS, Carver JA, Wilson MR. Clusterin is an ATP-independent chaperone with very broad substrate specificity that stabilizes stressed proteins in a folding-competent state. Biochemistry. Dec 26 2000;39(51):15953–60. doi: 10.1021/bi002189x [DOI] [PubMed] [Google Scholar]
8.Palladini G, Hegenbart U, Milani P, et al. A staging system for renal outcome and early markers of renal response to chemotherapy in AL amyloidosis. Blood. Oct 9 2014;124(15):2325–32. doi: 10.1182/blood-2014-04-570010 [DOI] [PubMed] [Google Scholar]
9.Kourelis TV, Dasari SS, Dispenzieri A, et al. A Proteomic Atlas of Cardiac Amyloid Plaques. JACC CardioOncol. Nov 2020;2(4):632–643. doi: 10.1016/j.jaccao.2020.08.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Becht E, McInnes L, Healy J, et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol. Dec 3 2018;doi: 10.1038/nbt.4314 [DOI] [PubMed] [Google Scholar]
11.Levine JH, Simonds EF, Bendall SC, et al. Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis. Cell. Jul 2 2015;162(1):184–97. doi: 10.1016/j.cell.2015.05.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Liao Y, Wang J, Jaehnig EJ, Shi Z, Zhang B. WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs. Nucleic Acids Res. Jul 2 2019;47(W1):W199–w205. doi: 10.1093/nar/gkz401 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Canton J, Neculai D, Grinstein S. Scavenger receptors in homeostasis and immunity. Nat Rev Immunol. Sep 2013;13(9):621–34. doi: 10.1038/nri3515 [DOI] [PubMed] [Google Scholar]
14.Papathanasiou S, Rickelt S, Soriano ME, et al. Tumor necrosis factor-alpha confers cardioprotection through ectopic expression of keratins K8 and K18. Nat Med. Sep 2015;21(9):1076–84. doi: 10.1038/nm.3925 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Djudjaj S, Papasotiriou M, Bulow RD, et al. Keratins are novel markers of renal epithelial cell injury. Kidney Int. Apr 2016;89(4):792–808. doi: 10.1016/j.kint.2015.10.015 [DOI] [PubMed] [Google Scholar]
16.Manissorn J, Khamchun S, Vinaiphat A, Thongboonkerd V. Alpha-tubulin enhanced renal tubular cell proliferation and tissue repair but reduced cell death and cell-crystal adhesion. Sci Rep. Jul 1 2016;6:28808. doi: 10.1038/srep28808 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Han SJ, Kim JH, Kim JI, Park KM. Inhibition of microtubule dynamics impedes repair of kidney ischemia/reperfusion injury and increases fibrosis. Sci Rep. Jun 8 2016;6:27775. doi: 10.1038/srep27775 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Rognoni P, Mazzini G, Caminito S, Palladini G, Lavatelli F. Dissecting the Molecular Features of Systemic Light Chain (AL) Amyloidosis: Contributions from Proteomics. Medicina (Kaunas). 2021;57(9):916. doi: 10.3390/medicina57090916 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Mollee P, Boros S, Loo D, et al. Implementation and evaluation of amyloidosis subtyping by laser-capture microdissection and tandem mass spectrometry. Clin Proteomics. 2016;13:30–30. doi: 10.1186/s12014-016-9133-x [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Rizou M, Frangou EA, Marineli F, et al. The family of 14–3-3 proteins and specifically 14–3-3σ are up-regulated during the development of renal pathologies. J Cell Mol Med. Sep 2018;22(9):4139–4149. doi: 10.1111/jcmm.13691 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Dasari S, Alexander MP, Vrana JA, et al. DnaJ Heat Shock Protein Family B Member 9 Is a Novel Biomarker for Fibrillary GN. J Am Soc Nephrol. Jan 2018;29(1):51–56. doi: 10.1681/asn.2017030306 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Nasr SH, Vrana JA, Dasari S, et al. DNAJB9 Is a Specific Immunohistochemical Marker for Fibrillary Glomerulonephritis. Kidney Int Rep. 2017;3(1):56–64. doi: 10.1016/j.ekir.2017.07.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Andeen NK, Yang HY, Dai DF, MacCoss MJ, Smith KD. DnaJ Homolog Subfamily B Member 9 Is a Putative Autoantigen in Fibrillary GN. J Am Soc Nephrol. Jan 2018;29(1):231–239. doi: 10.1681/asn.2017050566 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Zhang G, Kernan KA, Collins SJ, et al. Plasmin(ogen) promotes renal interstitial fibrosis by promoting epithelial-to-mesenchymal transition: role of plasmin-activated signals. J Am Soc Nephrol. Mar 2007;18(3):846–59. doi: 10.1681/asn.2006080886 [DOI] [PubMed] [Google Scholar]
25.Bi J, Watanabe H, Fujimura R, et al. A downstream molecule of 1,25-dihydroxyvitamin D3, alpha-1-acid glycoprotein, protects against mouse model of renal fibrosis. Sci Rep. Nov 26 2018;8(1):17329. doi: 10.1038/s41598-018-35339-x [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Watanabe H, Fujimura R, Hiramoto Y, et al. An acute phase protein α(1)-acid glycoprotein mitigates AKI and its progression to CKD through its anti-inflammatory action. Sci Rep. Apr 12 2021;11(1):7953. doi: 10.1038/s41598-021-87217-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Djudjaj S, Papasotiriou M, Bülow RD, et al. Keratins are novel markers of renal epithelial cell injury. Kidney Int. Apr 2016;89(4):792–808. doi: 10.1016/j.kint.2015.10.015 [DOI] [PubMed] [Google Scholar]
28.Nayak Rao S The role of heat shock proteins in kidney disease. J Transl Int Med. Sep 1 2016;4(3):114–117. doi: 10.1515/jtim-2016-0034 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Duce JA, Smith DP, Blake RE, et al. Linker histone H1 binds to disease associated amyloid-like fibrils. J Mol Biol. Aug 18 2006;361(3):493–505. doi: 10.1016/j.jmb.2006.06.038 [DOI] [PubMed] [Google Scholar]
30.Darmellah A, Rayah A, Auger R, et al. Ezrin/radixin/moesin are required for the purinergic P2X7 receptor (P2X7R)-dependent processing of the amyloid precursor protein. J Biol Chem. Oct 5 2012;287(41):34583–95. doi: 10.1074/jbc.M112.400010 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Randles M, Lausecker F, Kong Q, et al. Identification of an Altered Matrix Signature in Kidney Aging and Disease. J Am Soc Nephrol. May 28 2021;32(7):1713–32. doi: 10.1681/asn.2020101442 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Yang AH, Chen JY. Glomerular deposition of alpha 2-macroglobulin in glomerular diseases. Nephrol Dial Transplant. Mar 1997;12(3):465–9. doi: 10.1093/ndt/12.3.465 [DOI] [PubMed] [Google Scholar]
33.Motojima M, Matsusaka T, Kon V, Ichikawa I. Fibrinogen that appears in Bowman’s space of proteinuric kidneys in vivo activates podocyte Toll-like receptors 2 and 4 in vitro. Nephron Exp Nephrol. 2010;114(2):e39–47. doi: 10.1159/000254390 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Gaya da Costa M, Poppelaars F, van Kooten C, et al. Age and Sex-Associated Changes of Complement Activity and Complement Levels in a Healthy Caucasian Population. Front Immunol. 2018;9:2664. doi: 10.3389/fimmu.2018.02664 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Vasilopoulou E, Kolatsi-Joannou M, Lindenmeyer MT, et al. Loss of endogenous thymosin β(4) accelerates glomerular disease. Kidney Int. Nov 2016;90(5):1056–1070. doi: 10.1016/j.kint.2016.06.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Gu M, Mei X, Zhao Y. Galectins as potential pharmacological targets in renal injuries of diverse etiology. European Journal of Pharmacology. 2020/August/15/ 2020;881:173213. doi: 10.1016/j.ejphar.2020.173213 [DOI] [PubMed] [Google Scholar]
37.Carlos CP, Silva AA, Gil CD, Oliani SM. Pharmacological treatment with galectin-1 protects against renal ischaemia-reperfusion injury. Sci Rep. Jun 22 2018;8(1):9568. doi: 10.1038/s41598-018-27907-y [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Sidiqi MH, Aljama MA, Muchtar E, et al. Light chain type predicts organ involvement and survival in AL amyloidosis patients receiving stem cell transplantation. Blood Adv. Apr 10 2018;2(7):769–776. doi: 10.1182/bloodadvances.2018016782 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Jordan TL, Maar K, Redhage KR, et al. Light chain amyloidosis induced inflammatory changes in cardiomyocytes and adipose-derived mesenchymal stromal cells. Leukemia. 2020;34(5):1383–1393. doi: 10.1038/s41375-019-0640-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Park CW, Jung BK, Ryu KY. Reduced free ubiquitin levels and proteasome activity in cultured neurons and brain tissues treated with amyloid beta aggregates. Mol Brain. Jun 8 2020;13(1):89. doi: 10.1186/s13041-020-00632-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Balana AT, Levine PM, Craven TW, et al. O-GlcNAc modification of small heat shock proteins enhances their anti-amyloid chaperone activity. Nat Chem. May 2021;13(5):441–450. doi: 10.1038/s41557-021-00648-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Park SH, Kukushkin Y, Gupta R, et al. PolyQ proteins interfere with nuclear degradation of cytosolic proteins by sequestering the Sis1p chaperone. Cell. Jul 3 2013;154(1):134–45. doi: 10.1016/j.cell.2013.06.003 [DOI] [PubMed] [Google Scholar]
43.Boronat S, Cabrera M, Hidalgo E. Spatial sequestration of misfolded proteins as an active chaperone-mediated process during heat stress. Curr Genet. Apr 2021;67(2):237–243. doi: 10.1007/s00294-020-01135-2 [DOI] [PubMed] [Google Scholar]
44.Eisele F, Eisele-Burger AM, Hao X, et al. An Hsp90 co-chaperone links protein folding and degradation and is part of a conserved protein quality control. Cell Rep. Jun 29 2021;35(13):109328. doi: 10.1016/j.celrep.2021.109328 [DOI] [PubMed] [Google Scholar]
45.Aghazadeh Y, Papadopoulos V. The role of the 14–3-3 protein family in health, disease, and drug development. Drug Discov Today. Feb 2016;21(2):278–87. doi: 10.1016/j.drudis.2015.09.012 [DOI] [PubMed] [Google Scholar]
46.Williams DM, Thorn DC, Dobson CM, et al. The Amyloid Fibril-Forming β-Sheet Regions of Amyloid β and α-Synuclein Preferentially Interact with the Molecular Chaperone 14–3-3ζ. Molecules. Oct 11 2021;26(20)doi: 10.3390/molecules26206120 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Herod SG, Dyatel A, Hodapp S, Jovanovic M, Berchowitz LE. Clearance of an amyloid-like translational repressor is governed by 14–3-3 proteins. Cell Rep. May 3 2022;39(5):110753. doi: 10.1016/j.celrep.2022.110753 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Matsumoto T, Urushido M, Ide H, et al. Small Heat Shock Protein Beta-1 (HSPB1) Is Upregulated and Regulates Autophagy and Apoptosis of Renal Tubular Cells in Acute Kidney Injury. PLoS One. 2015;10(5):e0126229. doi: 10.1371/journal.pone.0126229 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Cheng CW, Rifai A, Ka SM, et al. Calcium-binding proteins annexin A2 and S100A6 are sensors of tubular injury and recovery in acute renal failure. Kidney Int. Dec 2005;68(6):2694–703. doi: 10.1111/j.1523-1755.2005.00740.x [DOI] [PubMed] [Google Scholar]
50.Tian ZY, Wang CY, Wang T, Li YC, Wang ZY. Glial S100A6 Degrades β-amyloid Aggregation through Targeting Competition with Zinc Ions. Aging Dis. Aug 2019;10(4):756–769. doi: 10.14336/ad.2018.0912 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Müller T, Concannon CG, Ward MW, et al. Modulation of gene expression and cytoskeletal dynamics by the amyloid precursor protein intracellular domain (AICD). Mol Biol Cell. Jan 2007;18(1):201–10. doi: 10.1091/mbc.e06-04-0283 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Wenzel SE, Tyurina YY, Zhao J, et al. PEBP1 Wardens Ferroptosis by Enabling Lipoxygenase Generation of Lipid Death Signals. Cell. Oct 19 2017;171(3):628–641.e26. doi: 10.1016/j.cell.2017.09.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Pasten C, Herrera-Luna Y, Lozano M, et al. Glutathione S-Transferase and Clusterin, New Players in the Ischemic Preconditioning Renal Protection in a Murine Model of Ischemia and Reperfusion. Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology. Oct 27 2021;55(5):635–650. doi: 10.33594/000000442 [DOI] [PubMed] [Google Scholar]
54.Imperlini E, Gnecchi M, Rognoni P, et al. Proteotoxicity in cardiac amyloidosis: amyloidogenic light chains affect the levels of intracellular proteins in human heart cells. Sci Rep. Nov 15 2017;7(1):15661. doi: 10.1038/s41598-017-15424-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Shi J, Guan J, Jiang B, et al. Amyloidogenic light chains induce cardiomyocyte contractile dysfunction and apoptosis via a non-canonical p38alpha MAPK pathway. Proc Natl Acad Sci U S A. Mar 2 2010;107(9):4188–93. doi: 10.1073/pnas.0912263107 [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Zhang Y, Yu W, Chang W, Wang M, Zhang L, Yu F. Light Chain Amyloidosis-Induced Autophagy Is Mediated by the Foxo3a/Beclin-1 Pathway in Cardiomyocytes. Lab Invest. Feb 2023;103(2):100001. doi: 10.1016/j.labinv.2022.100001 [DOI] [PubMed] [Google Scholar]
57.Kim MS, Pinto SM, Getnet D, et al. A draft map of the human proteome. Nature. May 29 2014;509(7502):575–81. doi: 10.1038/nature13302 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Wilhelm M, Schlegl J, Hahne H, et al. Mass-spectrometry-based draft of the human proteome. Nature. May 29 2014;509(7502):582–7. doi: 10.1038/nature13319 [DOI] [PubMed] [Google Scholar]
59.Marshall CB, Krofft RD, Blonski MJ, et al. Role of smooth muscle protein SM22alpha in glomerular epithelial cell injury. Am J Physiol Renal Physiol. Apr 2011;300(4):F1026–42. doi: 10.1152/ajprenal.00187.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Fiorentino L, Cavalera M, Menini S, et al. Loss of TIMP3 underlies diabetic nephropathy via FoxO1/STAT1 interplay. EMBO Mol Med. Mar 2013;5(3):441–55. doi: 10.1002/emmm.201201475 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Wang Z, Famulski K, Lee J, et al. TIMP2 and TIMP3 have divergent roles in early renal tubulointerstitial injury. Kidney Int. Jan 2014;85(1):82–93. doi: 10.1038/ki.2013.225 [DOI] [PubMed] [Google Scholar]
62.Sanchez-Nino MD, Sanz AB, Sanchez-Lopez E, et al. HSP27/HSPB1 as an adaptive podocyte antiapoptotic protein activated by high glucose and angiotensin II. Lab Invest. Jan 2012;92(1):32–45. doi: 10.1038/labinvest.2011.138 [DOI] [PubMed] [Google Scholar]
63.Mason WJ, Vasilopoulou E. The Pathophysiological Role of Thymosin beta4 in the Kidney Glomerulus. Int J Mol Sci. Apr 22 2023;24(9)doi: 10.3390/ijms24097684 [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Caster DJ, Korte EA, Merchant ML, et al. Patients with Proliferative Lupus Nephritis Have Autoantibodies That React to Moesin and Demonstrate Increased Glomerular Moesin Expression. J Clin Med. Feb 16 2021;10(4)doi: 10.3390/jcm10040793 [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Chen YX, Zhang W, Wang WM, et al. Role of moesin in renal fibrosis. PLoS One. 2014;9(11):e112936. doi: 10.1371/journal.pone.0112936 [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Candiano G, Musante L, Bruschi M, et al. Repetitive fragmentation products of albumin and alpha1-antitrypsin in glomerular diseases associated with nephrotic syndrome. J Am Soc Nephrol. Nov 2006;17(11):3139–48. doi: 10.1681/ASN.2006050486 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials