Predictive functional assay‐based classification of PMS2 variants in Lynch syndrome

Abstract The large majority of germline alterations identified in the DNA mismatch repair (MMR) gene PMS2, a low‐penetrance gene for the cancer predisposition Lynch syndrome, represent variants of uncertain significance (VUS). The inability to classify most VUS interferes with personalized healthcare. The complete in vitro MMR activity (CIMRA) assay, that only requires sequence information on the VUS, provides a functional analysis‐based quantitative tool to improve the classification of VUS in MMR proteins. To derive a formula that translates CIMRA assay results into the odds of pathogenicity (OddsPath) for VUS in PMS2 we used a set of clinically classified PMS2 variants supplemented by inactivating variants that were generated by an in cellulo genetic screen, as proxies for cancer‐predisposing variants. Validation of this OddsPath revealed high predictive values for benign and predisposing PMS2 VUS. We conclude that the OddsPath provides an integral metric that, following the other, higher penetrance, MMR proteins MSH2, MSH6 and MLH1 can be incorporated as strong evidence type into the upcoming criteria for MMR gene VUS classification of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP).

These criteria include allele frequency, cosegregation, age of tumor onset, microsatellite instability (MSI), and loss of protein staining in tumors, among others. Predisposing variants in PMS2 confer a low prevalence of cancer due to the protein's genetic redundancy (11%-20% lifetime colorectal cancer risk by age 70) (Senter et al., 2008;ten Broeke et al., 2015). The current clinically focused approach of variant classification, therefore, is often underpowered to classify PMS2 VUS, and 98% of PMS2 missense variants (1490/1521) remain unclassified (Class 3, VUS) in the ClinVar database (Accessed, August 19, 2021) (Landrum et al., 2018). Consequently, personalized healthcare for carriers and affected relatives cannot be implemented. For this reason, it is of the utmost importance to classify these VUS.
Since loss of biochemical MMR function is causative of cancer predisposition in LS, functional assays can be used to improve the classification of VUS. We have developed the complete in vitro mismatch repair activity (CIMRA) assay that rapidly quantifies the MMR activity of cDNA-encoded MMR protein VUS in vitro, based on sequence information of the VUS alone (Figure 1; Drost, Koppejan, et al., 2013;Drost, Lützen, et al., 2013). We have previously calibrated and validated the CIMRA assay for MSH2, MSH6, and MLH1 protein VUS (Drost et al., 2019. This was done by regression of the assay output against variants, independently classified using clinical data. In case there were insufficient clinically classified predisposing variants available we included MMR-inactivating variants produced by a genetic screen, as a proxy for such variants. The OddsPath may be combined with a probability of pathogenicity based on computational predictions (Prior-P) to calculate a posterior probability of pathogenicity (Posterior-P) between 0 and 1 (B. A. Thompson et al., 2013). This approach has been proven to have high sensitivities and specificities for both predisposing and benign variants of the MSH2, MSH6, and MLH1 MMR proteins (Drost et al., 2019. Here, we have used a set comprised of clinically classified benign and predisposing PMS2 substitution variants, supplemented with functionally disrupting (as a proxy for predisposing) Pms2 variants, generated using a genetic screen, to calibrate and validate the output from the CIMRA assay for PMS2 VUS into an OddsPath.
We demonstrate that the OddsPath is a robust metric that can be incorporated in the ACMG/AMP guidelines for PMS2 VUS classification.
2 | METHODS 2.1 | Cell culture and generation of Pms2-hemizygous mouse embryonic stem cells using CRISPR-Cas9 Hemizygous Pms2 mouse embryonic stem cells (mESCs) for use in the genetic screen were generated using CRISPR-Cas9. Briefly, two pairs of single-stranded oligonucleotides with BbsI overhangs targeting the C-terminal and N-terminal end of Pms2 were annealed and cloned into the Cas9 vector PX330-Puro (Cong et al., 2013). mESCs were transfected with the two PX330-Puro plasmids containing the gRNA's using Lipofectamine 2000 (Thermo scientific) and after 24 h, positively transfected cells were selected with 1 μg/ml Puromycin. After selection, cells were subcloned, and Pms2 hemizygosity was confirmed using allele-specific PCR. Pms2 protein expression was checked by immunoblotting.
F I G U R E 1 Flow diagram of the complete in vitro mismatch repair activity (CIMRA) assay. The variant of interest is produced by a PCR procedure, expressed in a coupled transcription-translation kit and then heterodimerized with MLH1 protein, expressed from wild-type MLH1 cDNA. Then the functional activity of the variant heterodimer is assessed in a complementation assay containing a substrate containing a T.G mismatch with a linked fluorescent group (star), and MLH1/PMS2-deficient cell extract. Repair of the mismatch recreates a Hin1II-cleavable restriction endonuclease site, resulting in the generation of a fluorescent diagnostic fragment (green arrow) that is visualized by capillary electrophoresis. Following calibration of the assay results against independently classified variants, the relative abundance of this diagnostic fragment is translated into an odds of pathogenicity for the variant PMS2/MLH1 heterodimer (Drost, Koppejan, et al., 2013;Drost, Lützen, et al., 2013).

| Generation of a set of pathogenic PMS2 missense variants using a genetic screen
To compile a set of inactivating amino acid substitutions in Pms2, as a proxy for cancer-predisposing variants, a genetic screen was performed, essentially as described for Msh2 and Msh6 (Drost, Lützen, et al., 2013). In brief, mESCs hemizygous for Pms2 were exposed to the strong mutagen N-ethyl-N-nitrosourea (ENU; Sigma-Aldrich) with the aim of introducing random missense substitutions in the genome (Figure 2). Cells that had become MMR-deficient, presumably by an inactivating mutation at the monoallelic Pms2 gene, were selected using the Guanine analog 6-Thioguanine (6-TG, Sigma-Aldrich). 6-TG is cytotoxic via the lethal processing of methyl-6-thioGuanine:Thymidine mismatches by MMR. Consequently, a defect in MMR results in survival to the drug and 6-TG is a widely used agent used to select for MMR-deficient cells and to test for functional inactivation and cancer predisposition caused by a variant Houlleberghs et al., 2016;Houlleberghs et al., 2020). Surviving clones were treated for 1 week in hypoxanthine-aminopterin-thymidine (HAT)-supplemented medium (Gibco) to kill inadvertent Hprt-mutant cells that also survive 6-TG treatment. Next, surviving clones were picked and expanded in 96-well plates. These clones were screened against loss of heterozygosity (LOH) at Pms2, an alternative way of losing the wild-type Pms2 allele, by intragenic allele-specific PCR. From remaining clones, Pms2 cDNA was generated and the critical, conserved, domains were sequenced using Sanger sequencing to identify the inactivating amino acid substitution. All primer sequences and PCR protocols are available upon request.

| Selection of missense substitutions for CIMRA assay calibration and validation
We derived a set of PMS2 missense variants from the ClinVar and  Table S1). A total of 31 unique inactivating missense substitutions were derived from the genetic screen (Supporting Information: Table S2). A part of the variants identified in the databases and screen were excluded from the calibration and validation set based on the following criteria: missense variants in the start codon of PMS2; proven spliceogenic variants; variants with an unavailable in silico prior probability (Prior-P); nonconserved amino acid variants (if derived from the murine genetic screen); one variant used as a predisposing control in the CIMRA assay (Supporting Information: Tables S1 and S2).
Briefly, the variant PMS2 cDNA is synthesized by mutagenic PCR followed by in vitro transcription and translation of the variant PMS2 protein as well as of the wild-type MLH1 dimerization partner. Nuclear extract from MLH1-deficient human HeLa cells (which is also deficient for PMS2 protein due to protein instability without its partner) generated by CRISPR-Cas9 gene targeting is complemented with in vitro produced human (variant) PMS2/ MLH1 protein and a mismatch-containing fluorescent substrate.
Repair of the mismatch recreates a Hin1II restriction site, which is T A B L E 1 CIMRA assay results and classification of substitution variants used for calibration and validation of the assay.

| Regression for CIMRA assay calibration and validation
Regression of CIMRA assay results against the known or assumed statistical probability of pathogenicity of the variants from the test set was performed following methods similar to those  Table S3.  Figure S1 and Table S2). Of these unique amino acid substitutions, two residues were not conserved between mice and humans and for one no Prior-P could be derived as two nucleotides were altered. The human equivalents of the remaining 28 Pms2-inactivating substitution variants were included, as proxies for predisposing PMS2 variants, in the calibration and validation set of the CIMRA assay ( Figure 3, Table 1). Of note, variant p.Asn749Lys, identified in the genetic screen, was also listed as an unclassified human VUS in the ClinVar database (Accessed, October 15, 2021), which supports its role in cancer predisposition.

| CIMRA assay calibration and validation
We used the CIMRA assay to determine the MMR activity of the set of 51 human PMS2 substitution variants, comprising the 20 class 1/2 variants and 31 (likely) predisposing or genetic screen-derived inactivating variants (   Table S3). Thus, the CIMRA assay-derived OddsPath displays a bimodal distribution, in support of its high predictive value for predisposing and for benign variants. The OddsPath for individual variants, and relevant ACMG/ AMP codes are shown in Table 1.
As summarized in Table 2, none of the "benign" reference set variants were incorrectly assigned a high OddsPath (in favor of pathogenicity), and none of the "predisposing" reference set variants were incorrectly assigned a low OddsPath.
F I G U R E 4 Mismatch repair activity in the complete in vitro mismatch repair activity (CIMRA) assay of PMS2 variants used for calibration and validation of the assay. Relative repair efficiencies for human PMS2 variants used in the calibration and validation of the CIMRA assay. Variants are colored according to their ClinVar classification and their identification as inactivating in the genetic screen. The PMS2 p.S46I variant was included in every assay as a mismatch repair-deficient pathogenic control and p.G197G as a wild-type control (Borràs et al., 2013) and are colored black. Bars represent ± Standarderror of the mean (SEM) of three independent experiments.
F I G U R E 5 Odds of pathogenicity of PMS2 variants The odds of pathogenicity (OddsPath) for PMS2 variants used in the calibration and validation of the complete in vitro mismatch repair activity (CIMRA) assay plotted against the average relative repair efficiency from the CIMRA assay. The odds of pathogenicity (OddsPath) for each PMS2 variant was derived using the leaveone-out procedure. Each variant is colored according to the computational prediction of pathogenicity (Prior-P) (see Table 1; B. A. Thompson et al., 2013). Four variants for which the bioinformatics-based Prior-P and the CIMRA assay-based OddsPath were strongly discordant are indicated separately in the figure.  Thompson et al., 2020). Nevertheless, to further improve the classification of predicted splice-impacting variants, additional functional evidence is warranted, as recommended by Brnich et al. (Brnich et al., 2019). An example of assays that can be used to assess functional consequences of putative splice, 5′ and 3′ UTR, and cellular compartmentalisation-perturbed variants are cellular assays that are based on the introduction of the VUS at its genomic locus and testing MSI and loss of toxicity of 6-TG or methylating agents (Houlleberghs et al., 2016(Houlleberghs et al., , 2020Rath et al., 2019).

| DISCUSSION
Of further relevance, the incidence of cancer in individuals carrying a fully inactivating PMS2 variant is lower than that in individuals carrying a fully inactivating MSH2 or MLH1 variant, which results from the partial genetic redundancy of PMS2 with minor MMR proteins (notably MLH3 or PMS1). Nevertheless, such a PMS2 variant will still be predisposing.
The high sensitivities and specificities of the CIMRA assay for the classification of VUS in MSH2, MSH6, and MLH1 (Drost et al., 2019, and now also in PMS2, has led to its inclusion in ACMG/AMP guidelines (https://www.insight-group.org/criteria) as a strong criterion for pathogenicity (PS3) or against pathogenicity (BS3). In these guidelines, the CIMRA assay-derived OddsPath is defined as strong evidence of pathogenicity (PS3-strong) when the OddsPath >18.7, a moderate evidence of pathogenicity (PS3moderate) with an OddsPath >4.32 and ≤18.7, while an OddsPath of >2.08 and ≤4.32 is considered supportive of pathogenicity (PS3supportive) (see also assay as a strong evidence type to classify VUS in all four MMR genes will not only significantly advance their classification worldwide, but also provide a paradigm for the translation of personalized cancer genetics into personalized healthcare.

CONFLICTS OF INTEREST
The authors declare no conflicts of interest.