Immunohistochemistry and Mutation Analysis of SDHx Genes in Carotid Paragangliomas

Carotid paragangliomas (CPGLs) are rare neuroendocrine tumors often associated with mutations in SDHx genes. The immunohistochemistry of succinate dehydrogenase (SDH) subunits has been considered a useful instrument for the prediction of SDHx mutations in paragangliomas/pheochromocytomas. We compared the mutation status of SDHx genes with the immunohistochemical (IHC) staining of SDH subunits in CPGLs. To identify pathogenic/likely pathogenic variants in SDHx genes, exome sequencing data analysis among 42 CPGL patients was performed. IHC staining of SDH subunits was carried out for all CPGLs studied. We encountered SDHx variants in 38% (16/42) of the cases in SDHx genes. IHC showed negative (5/15) or weak diffuse (10/15) SDHB staining in most tumors with variants in any of SDHx (94%, 15/16). In SDHA-mutated CPGL, SDHA expression was completely absent and weak diffuse SDHB staining was detected. Positive immunoreactivity for all SDH subunits was found in one case with a variant in SDHD. Notably, CPGL samples without variants in SDHx also demonstrated negative (2/11) or weak diffuse (9/11) SDHB staining (42%, 11/26). Obtained results indicate that SDH immunohistochemistry does not fully reflect the presence of mutations in the genes; diagnostic effectiveness of this method was 71%. However, given the high sensitivity of SDHB immunohistochemistry, it could be used for initial identifications of patients potentially carrying SDHx mutations for recommendation of genetic testing.


Introduction
Carotid paraganglioma (CPGL) is a rare neuroendocrine tumor that arises from the carotid body. CPGL represents more than half of all head and neck (HN) paragangliomas (PGLs) [1]. According to the WHO Classification of Head and Neck Tumors 2017, PGLs were reclassified from indeterminate to malignant tumors with variable potential of metastasis [2]. As CPGL associates with the carotid while several studies proposed the immunohistochemistry of the SDHB subunit as a useful instrument for the prediction of SDHx mutation status. These important results indicate the necessity of genetic testing of SDHx variants along with IHC study in CPGLs.

Correlation of SDHx Mutation Status with Their Immunostaining
Immunohistochemical staining was performed for SDHA, SDHB, SDHC, and SDHD subunits on 42 CPGL samples (Supplementary File S1). SDHB staining was assessed as follows: (+) Positive as granular cytoplasmic staining of tumor cells in parallel with the same intensity staining of internal positive control (endothelial cells); (-) negative as completely absent cytoplasmic staining together with staining of internal positive control; (*) weak diffuse as a cytoplasmic blush lacking definite granularity contrasting the strong granular staining of internal positive control ( Figure 2). Immunostaining of SDHA, SDHC, and SDHD was scored as positive or negative in the same manner as SDHB. In addition, we have analyzed the mutation status of RET, VHL, TMEM127, MAX, IDH1, IDH2, FH, and SLC25A11, as well as of genes belonging to the family of succinate dehydrogenase complex assembly factors (SDHAF1, SDHAF2, SDHAF3, and SDHAF4), which are the main susceptibility genes for CPGLs (Table 1). Pathogenic/likely pathogenic variants were found in RET and IDH1; no variants in VHL, TMEM127, MAX, IDH2, FH, SLC25A11, and SDHAF1-4 genes were identified.

Correlation of SDHx Mutation Status with Their Immunostaining
Immunohistochemical staining was performed for SDHA, SDHB, SDHC, and SDHD subunits on 42 CPGL samples (Supplementary File S1). SDHB staining was assessed as follows: (+) Positive as granular cytoplasmic staining of tumor cells in parallel with the same intensity staining of internal positive control (endothelial cells); (-) negative as completely absent cytoplasmic staining together with staining of internal positive control; (*) weak diffuse as a cytoplasmic blush lacking definite granularity contrasting the strong granular staining of internal positive control ( Figure 2). Immunostaining of SDHA, SDHC, and SDHD was scored as positive or negative in the same manner as SDHB. In the majority of SDHx-mutated tumors (94%, 15/16), we detected negative or weak diffuse staining of SDHB (Supplementary Table S1). The Spearman's rank correlation coefficient (rs) between the presence of mutations in any SDHx genes and negative or weak diffuse SDHB staining was 0.51, p ≤ 0.05.
Negative or weak diffuse SDHB staining was found in nine out of ten cases with In the majority of SDHx-mutated tumors (94%, 15/16), we detected negative or weak diffuse staining of SDHB (Supplementary Table S1). The Spearman's rank correlation coefficient (r s ) between the presence of mutations in any SDHx genes and negative or weak diffuse SDHB staining was 0.51, p ≤ 0.05.
Negative or weak diffuse SDHB staining was found in nine out of ten cases with pathogenic/likely pathogenic SDHD variants; one SDHD-mutated tumor showed positive staining of all SDH subunits. In all the samples, SDHD was positively stained.
In two out of three samples with pathogenic/likely pathogenic variants in the SDHC gene, we identified weak diffuse SDHB staining and simultaneous positive SDHC expression. In one case, negative immunohistochemical staining for both SDHB and SDHC was found.
All samples with pathogenic/likely pathogenic variants in the SDHB gene showed negative SDHB staining.
A pathogenic/likely pathogenic variant in the SDHA gene was identified only in one patient. We observed both weak diffuse SDHB staining and negative SDHA expression in this sample.
Among twenty-six CPGLs with no pathogenic/likely pathogenic variants in the SDHx genes, negative or weak diffuse SDHB staining was observed in eleven cases (42%). Fifteen samples were immunopositive for all SDH subunits.
Three tumors with pathogenic variants of the RET gene, which were corepresented with SDHA, SDHC, and SDHD variants, have been characterized by negative or weak diffuse SDHB staining. Positive immunoreactivity was found in one patient with a likely pathogenic variant in the IDH1 gene occurring with no variant in SDHx.

Calculation of Diagnostic Accuracy
To determine the ability of SDHB immunohistochemistry discriminating SDHx-mutation carriers, we measured the sensitivity, specificity, and diagnostic effectiveness (accuracy) according to the following formulas: Sensitivity = TP/TP + FN, Specificity = TN/TN + FP, where TP (true positives)-positively diagnosed subjects with the disease, FN (false negatives)-negatively diagnosed subjects with the disease, TN (true negatives)-negatively diagnosed subjects without the disease, and FP (false positives)-positively diagnosed subjects without the disease. In CPGLs, SDHB immunohistochemistry showed a sensitivity of 94% (15/16) and specificity of 58% (15/26). The diagnostic effectiveness of this method was 71% (30/42).

Discussion
Tumor cells are well-known to have alterations in energy metabolism that are exemplified by the Warburg effect [36,37]. A metabolic shift from mitochondrial respiration to glycolysis can be caused by mitochondrial dysfunction or by the reduction in its activity [38,39]. SDH has a critical role in mitochondrial metabolism; disruption of the SDH complex leads to abnormal accumulation of succinate in the cytosol, reprogramming of the energy metabolism, increased ROS production, stabilization of hypoxia-inducible factors (HIFs), and altered gene expression (in particular, for HIF targets) [40]. All of these changes can trigger neoplastic growth [38,41]. SDH abnormalities are associated with a tumorigenesis risk, including the development of PGLs/PCCs, renal and thyroid cancer, as well as composite PGLs/gastrointestinal stromal tumors (GISTs)/pulmonary chondromas (Carney triad) and PGLs/GISTs (Carney-Stratakis syndrome) [42].
SDHx are the most commonly mutated genes in PGLs/PCCs [43]. Variants in SDHD are more frequently observed in HNPGLs, followed by SDHB and SDHC mutations [44,45]. SDHA variants show extremely low penetrance in HNPGLs [10]. We obtained similar results; however, SDHC variants were found more often than SDHB variants. Previously, it has been reported that SDHC mutations are mainly associated with the development of CPGLs, explaining this difference [46].
Mutations in any of the SDHx genes can cause a destabilization of the SDH complex, loss of its enzymatic activity, and a disruption in the electron transport function [47][48][49][50]. Numerous studies have reported a changed expression pattern of SDHB presented as negative or weak diffuse immunostaining in tumors with SDHA-, SDHB-, SDHC-, and SDHD mutations [24,[51][52][53]. It was shown that negative SDHB staining is more commonly associated with mutations in SDHB, whereas weak diffuse staining often occurs in SDHD-mutated tumors [52]. We also detected loss of SDHB expression in all patients with SDHB variants and weak diffuse SDHB staining in the majority of SDHD mutation carriers that support this finding. Studied patients carrying variants in SDHC showed both negative and weak diffuse SDHB staining. Notably, a number of authors interpreted SDHB staining only as positive or negative and considered a weak diffuse expression pattern as negative. Generally, both patterns indicate SDH deficiency, which is a surrogate marker for SDHx germline mutations almost always causing the gene biallelic inactivation [23]. Somatic events leading to biallelic inactivation have been rarely reported for the SDHx genes [23]. In the study, we used an archival collection of CPGLs for which paired normal tissues were unavailable; therefore, germline and somatic mutation status could not be estimated. However, based on this conception, we can suppose that in the majority of studied patients with SDH deficiency, the mutations of SDHx genes are germline. In one patient with a novel likely pathogenic frameshift SDHD variant, we found retention of SDHB expression. Possibly, this variant does not have a high impact on the protein structure or it occurs in one allele of the gene.
In a patient with a pathogenic SDHA variant, we have seen completely absent SDHA expression and weak diffuse SDHB immunostaining that is in accordance with the literature. Direct correlation with the presence of the gene mutation and loss of the protein expression is observed only for SDHA. Negative SDHA expression is defined both when mutation leads to the truncated protein and owing to missense mutation [12,54]. SDHB expression at the same time also becomes negative in SDHA-mutated tumors, supported by almost all reported cases (including our results) [12,24,54,55]. Moreover, SDHA mutation is a rare event in PGLs; therefore, the use of SDHB immunohistochemistry seems to be more expedient than SDHA/SDHB immunohistochemistry for prediction of mutations in any SDHx genes.
The loss of SDHC expression was revealed in one out of three patients with variants in the SDHC gene. This variant, NM_003001.3: c.224G > A, p.(Gly75Asp) (chr1: 161310428бrs786205147;189841), was described in the ClinVar database as a germline likely pathogenic variant associated with the hereditary cancer-predisposing syndrome and Carney triad with no experimental evidence of its pathogenicity to date. In this patient, negative SDHB staining was also determined. Therefore, we can suggest that, except for SDHA, no evident correlations have been found between negative SDHC and SDHD immunohistochemistry and the presence of pathogenic variants in the corresponding genes.
Among 42 patients with CPGLs, we revealed pathogenic RET variants in three cases and a likely pathogenic IDH1 variant in one patient. RET variants were presented in SDHx-mutated tumors that showed negative or weak diffuse SDHB staining. SDHx-mutations seem to be the main drivers of SDH efficiency; therefore, we cannot correctly assess the correlation of identified RET variants with the SDHB immunohistochemistry in these samples. The presence of the IDH1 variant was not associated with the changed immunostaining of any SDH subunits. However, more cases are needed to assess the impact of IDHx mutations on the stability of the SDH complex.
Despite the great results showing a high correlation of SDHB immunohistochemistry with the presence of SDHx variants (94%), negative or weak diffuse SDHB staining has also been found in 42% of tumors without pathogenic/likely pathogenic variants in any SDHx genes. In this case, SDH deficiency can be caused by mutations in the DNA regions, which have not been screened, or epimutations.
Given these data, we presumed that SDHB immunohistochemistry could be used for the initial assessment of SDHx variants in CPGLs with genetic testing in parallel. Additional SDHA staining increases the cost of the IHC analysis, but among PGLs/PCCs, SDHA mutation frequency is extremely low, and in the majority of such cases, negative or weak diffuse SDHB staining is also observed.

Tumor Samples and Patients
A total of 42 carotid paraganglioma samples (archive material) were used in this study. The formalin-fixed paraffin-embedded (FFPE) tumor tissues were collected in the Vishnevsky Institute of Surgery, Ministry of Health of the Russian Federation. Tumors were obtained from patients who did not receive radiotherapy or chemotherapy before surgery. Samples have no less than 80% of tumor cells. All the patients provided written informed consent. The study was approved by the ethics committee from the Vishnevsky Institute of Surgery with ethics committee approval no. 007/18, 02.10.2018 and performed according to the Declaration of Helsinki (1964). The clinicopathologic characteristics of the patients with CPGLs are presented in Table 2.

DNA Extraction
DNA was extracted from tumor tissues using a High Pure FFPET DNA Isolation Kit (Roche, Basel, Switzerland) according to the manufacturer's instructions. The quantification of isolated DNA was performed with a Qubit 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). DNA quality was assessed by quantitative PCR (qPCR) using QuantumDNA Kit (Evrogen, Moscow, Russia).

Exome Sequencing
Exome libraries were prepared from DNA using a Rapid Capture Exome Kit (Illumina, San Diego, CA, USA) or TruSeq Exome Library Prep Kit (Illumina), according to the guidelines. Capture probes covered the same DNA regions in both kits (predominantly gene-coding regions). Library quantification was carried out using both Qubit 2.0 Fluorometer and qPCR. A quality assay of the libraries was performed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). High-throughput sequencing of the libraries was performed on a NextSeq 500 System (Illumina) in a paired-end mode of 76 × 2 bp. The average coverage for each sample was at least 300×. In this study, we used exome data of CPGL samples that were previously sequenced; raw sequence reads for Pat02-Pat51 are available at NCBI Sequence Read Archive (SRA) BioProject PRJNA411769 [56], and sequence data for Pat100-Pat104 are available at SRA BioProject PRJNA476932 [57]. Raw sequence data from an expanded set of CPGL samples (Pat53-Pat71) were added to the NCBI SRA BioProject PRJNA411769.

Immunohistochemistry
IHC staining was used to analyze SDHx gene expression at the protein level. Sections (3-5 µm) from FFPE samples were prepared on glass slides using an HM 355S Automatic Microtome (Thermo Fisher Scientific) and then stained with hematoxylin-eosin (H&E) for histomorphological analysis. Deparaffinization of the sections was performed with xylene, with further rehydration in decreasing alcohol concentrations (absolute, 90%, 70%, and 50%) and washing in distilled water. Immunoreactions were performed in a serial manner using primary antibodies for all four SDH subunits (SDHA, monoclonal, clone 2E3GC12FB2AE2; SDHB, monoclonal, clone 21A11AE7; SDHC, monoclonal, clone EPR11035(B); SDHD, polyclonal) from Abcam (Cambridge, United Kingdom) on a Lab Vision Autostainer 360-2D (Thermo Fisher Scientific), according to the manufacturer's instructions. Reactions continued in a ready-to-use visualization system Histofine DAB-2V (Nichirei Biosciences, Tokio, Japan) with universal chromogen-labeled (3,3 -diaminobenzidine, DAB) secondary antibodies. Additional Mayer's hematoxylin staining was performed. Samples incubated without primary antibodies were used as the negative controls (Supplementary Figure S1). Granular cytoplasmic staining of SDH subunits in endothelial cells was used as a positive internal control. The slides were visualized using an Axio Imager 2 (Carl Zeiss Microscopy, Jena, Germany).

Correlation Analysis
Correlation analysis between SDHB staining and the presence of mutations in any SDHx genes was performed using the Spearman's rank correlation test with STATISTICA 10 (StatSoft Inc., Tulsa, OK, USA).

Conclusions
This is the first study on the correlation between SDHx mutation status and their protein expression, respectively estimated with exome sequencing and IHC in a representative set of CPGLs. It has previously been reported that negative or weak diffuse SDHB staining has high sensitivity and specificity for the prediction of mutations of SDHx in PGLs/PCCs. However, our study showed that altered SDHB immunostaining widely occurs in tumors that do not carry pathogenic/likely pathogenic variants in the genes. These divergent results could be explained by the fact that earlier studies focused on PGLs/PCCs or all HNPGLs, but not only on CPGLs. Nevertheless, the sensitivity of the method remains high. Based on the collected data, we believe that SDHB immunohistochemistry could be used for primary identifications of patients potentially carrying SDHx variants who should be further referred for genetic testing. Acknowledgments: Authors thank the Vishnevsky Institute of Surgery for tissue samples, National Medical Research Center of Radiology and A. N. Severtsov Institute of Ecology and Evolution for assistance in the data analysis. This work was performed using the equipment of EIMB RAS "Genome" center (http://www.eimb.ru/ru1/ ckp/ccu_genome_c.php).

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.