Germline loss-of-function PAM variants are enriched in subjects with pituitary hypersecretion

Abstract

Introduction: Pituitary adenomas (PAs) are common, usually benign tumors of the anterior pituitary gland which, for the most part, have no known genetic cause. PAs are associated with major clinical effects due to hormonal dysregulation and tumoral impingement on vital brain structures. PAM encodes a multifunctional protein responsible for the essential C-terminal amidation of secreted peptides.

Methods: Following the identification of a loss-of-function variant (p.Arg703Gln) in the peptidylglycine a-amidating monooxygenase (PAM) gene in a family with pituitary gigantism, we investigated 299 individuals with sporadic PAs and 17 familial isolated PA kindreds for PAM variants. Genetic screening was performed by germline and tumor sequencing and germline copy number variation (CNV) analysis.

Results: In germline DNA, we detected seven heterozygous, likely pathogenic missense, truncating, and regulatory SNVs. These SNVs were found in sporadic subjects with growth hormone excess (p.Gly552Arg and p.Phe759Ser), pediatric Cushing disease (c.-133T>C and p.His778fs), or different types of PAs (c.-361G>A, p.Ser539Trp, and p.Asp563Gly). The SNVs were functionally tested in vitro for protein expression and trafficking by Western blotting, splicing by minigene assays, and amidation activity in cell lysates and serum samples. These analyses confirmed a deleterious effect on protein expression and/or function. By interrogating 200,000 exomes from the UK Biobank, we confirmed a significant association of the PAM gene and rare PAM SNVs with diagnoses linked to pituitary gland hyperfunction.

Conclusion: The identification of PAM as a candidate gene associated with pituitary hypersecretion opens the possibility of developing novel therapeutics based on altering PAM function.

Keywords: Cushing disease; acromegaly; amidation; gigantism; peptidylglycine α-amidating monooxygenase; pituitary tumors.

Figure 1

Clinical findings in the index family with pituitary gigantism. (A) Pedigree. Generation numbers are represented by Roman numerals, and individual numbers are in Arabic numerals. The proband is II-2, indicated by the black arrow. Open square/circle, unaffected male/female; filled square, affected male. The peptidylglycine α-amidating monooxygenase (PAM) mutational status is shown under each screened individual. In (B), the white arrowheads point to a possible 4 mm lesion (left panel, sagittal plane; right panel, coronal plane) seen in II-2. (C) Growth chart for individual III-3 before and after medical interventions. (D) A sagittal T2-weighted magnetic resonance imaging of individual III-3 that was performed before medical therapy began did not reveal a pituitary lesion. (E) Time course of the effects of treatment modalities on the growth hormone (GH; left axis) and IGF-1 (right axis; squares) in individual III-3. The rapid decrease of IGF-1 after switching from lanreotide (LAN) to pegvisomant (PegV) is evident.

Figure 2

Location and evolutionary conservation of the missense and frameshift PAM SNVs that were functionally tested. (A) Schematic representation of the PAM gene (GenBank: NM_000919.3, PAM-1, 25 exons) and encoded protein (NP_000910.2, 974 amino acids), including functional domains, with 15 missense and one frameshift SNVs. Gene and protein structures were drawn with the Gene Structure Display Server (GSDS ver. 2.0) (12). Variants found to have deleterious effects on PAM function/expression (p < 0.01) are shown in yellow lettering, while those without major effects are shown in white lettering. Brackets identify the non-catalytic regions that precede PHMcc (N-terminus) and follow PALcc (C-terminus); CD, cytosolic domain; CDS, coding sequence; PALcc, catalytic core of peptidyl-α-hydroxyglycine α-amidating lyase; PHMcc, catalytic core of peptidylglycine α-hydroxylating monooxygenase; TMD, transmembrane domain; UTR, untranslated region. (B) Protein sequence alignments for five of the variants with deleterious effects. Conserved affected residues are shown in yellow. PAL activities, indicated in red, refer to functional experiments in PEAKrapid cells. (C) The crystal structure of rat PALcc (PDB entry 3FW0) was used to contextualize the missense variants categorized as likely pathogenic based on in silico analyses; the WT residue is shown on the left and the mutant residue on the right. PAL folds as a β-propeller, with six blades (numbered 1–6) positioned around a central cavity. The calcium and mercury ions are depicted as yellow and orange spheres, respectively. The mercury ion was used instead of zinc to capture the binding of a non-peptide substrate, α-hydroxyhippuric acid, depicted in blue. The affected residues are highlighted in purple (Ser or Trp 539), cyan (Gly or Arg 552), red (Asp or Gly 563), green (Arg or Gln 703), and orange (Phe or Ser 759), along the ribbon visualization of WT rat PALcc in gray. Arg703 is positioned at the active site and participates in substrate binding. Interestingly, p.Gly552Arg and p.Asp563Gly are located on the same face of the β-propeller. C, C-terminus; N, N-terminus.

Figure 3

Enzymatic activity, protein expression, and glycosylation pattern of PAM variants. (A) PHM and PAL activity. As described in the Methods, TMT-solubilized particulate fractions prepared from transiently transfected PEAKrapid cells were assayed for PHM and PAL activity. The data for the level of expression of WT PAM and each full-length variant were determined by quantifying the FLAG-tag signal. The levels of p.His778fs were assessed as described in the Methods. NT, not transfected; **p < 0.0001. (B) PAM protein expression. SNV expression was assessed using a FLAG tag antibody and an antibody to a peptide contained in PHMcc (JH246). The lines separating WT and p.His240Tyr from p.Arg703Gln and p.Phe759Ser indicate that the data for two intervening samples were removed. Molecular weight standards are indicated. (C) PAM protein diagram, indicating the location of the JH246 epitope (red horizontal line), the FLAG tag (gray diamond), and the expected location of N- and O-glycans (blue and yellow freeform shapes, respectively). (D) Glycosylation is altered in a subset of PAM variants. Cell lysates were treated with PNGase F or neuraminidase as described in the Methods. Proteins were visualized using the FLAG antibody. The samples treated with neuraminidase were analyzed on two separate gels, with the p.Arg703Gln samples appearing in part on both gels. Con, control; In, input; N-Gly, PNGase; Neur, neuraminidase; NT, not transfected.

Figure 4

Functional studies of the c.2332-2A>T (p.His778fs)-truncating variant. (A) RT-PCR analysis of PAM exon 21 splicing was performed using blood-extracted RNA from two family members carrying the c.2332-2A>T variant (NIH26 and her mother) and four WT control complementary DNAs (cDNAs). Primers were designed on exons 17 and 22. Both carriers and controls showed a normally spliced transcript (713 bp, upper band), while only the carriers showed an additional alternatively spliced transcript lacking exon 21 (the 613 bp band, identified by the arrow). MW, molecular weight marker. (B) The identity of the PCR products (panel A) was confirmed by Sanger sequencing. The arrows point to the variant-specific peaks present only in the carriers. MUT, mutated; WT, wild type. (C) Minigene assay. After transfection into HEK-293AD cells, mRNA synthesis from the plasmids using the cells’ own transcription and splicing machinery led to mRNA products containing (WT) or lacking (variant) exon 21 of PAM flanked by two exons from the pSPL3 vector. The RT-PCR analysis of the minigene transcripts was conducted using vector-specific primers. MOCK, cDNA from empty vector-transfected cells consisting of a 260 bp band made up of fragments of pSPL3 exons; –, negative control (RT-PCR without cDNA). The lines separating MOCK and negative control from WT and c.2332-2A>T indicate that the data for intervening samples were removed. (D) An expression vector lacking exon 21 of human PAM-1 (H778fs) was transiently expressed in PEAKrapid cells. Proteins were visualized using an antibody to PHMcc (JH246 PHM Ab). The expression of WT PAM produces a major band at 114 ± 1 kDa and a minor one at 105 kDa. The only band visible in the cells expressing p.His778fs migrated at 75 kDa; after signal peptide removal, the mass predicted for this protein—which includes only the first 777 residues of WT PAM-1 but extends 45 residues beyond residue 777 before reaching a stop codon—is 90.31 kDa. NT, not transfected.

Figure 5

WT and mutant PAM promoter activity. A 5 kb PAM promoter-5’UTR WT sequence was cloned upstream of a luciferase reporter. Six mutant constructs were created by site-directed mutagenesis. The constructs were transiently transfected into HEK-293AD cells together with a Renilla luciferase reporter for normalization. Firefly and Renilla luciferase activities were measured 24 h post-transfection. Two SNVs, c.-361G>A and c.-133T>C, have significantly lower transcriptional activity than the WT hybrid transcript. Mock, empty pRMT-Luc vector. Differences between experimental groups were analyzed by one-way ANOVA with Dunnett’s post hoc test, using WT as the control group. RLU, relative luciferase activity. *p = 0.0443; **p = 0.0019; ***p = 0.0005.

Figure 6

PHM and PAL activity in human sera. Assays for PHM and PAL activity were carried out on sera from individuals harboring variants showing deleterious effects in in vitro assays (p.Asp563Gly, p.Arg703Gln, and p.His778fs), subjects with more benign variants (p.Val27Ile and p.Val49Leu), and controls (WT). In NIH36, the p.Phe31Tyr variant occurs along with the 5’UTR c.-133T>C variant. The dashed line at 0.5 indicates the activity level expected with one completely inactive allele.

Figure 7

Gene- and variant-based association analyses for PAM in the UK Biobank (UKBB). (A) Results of the sequence kernel association test (SKAT) analysis of PAM variants for the diagnoses of hyperfunction of the pituitary gland (identified from UKBB fields 41270 and 41204—primary and secondary ICD-10 diagnoses from hospitalization records—and 20002—self-reported diagnosis) in 200,000 UKBB participants. The SKAT CommonRare algorithm was used for analysis. (B) Significant enrichment for PAM pathological missense PAM variants in subjects diagnosed with hyperfunction of the pituitary gland.

Figure 8

Biological processes and cellular components that might be affected by PAM haploinsufficiency in pituitary hormone-secreting cells. The trafficking of integral membrane PAM and pituitary hormones through the regulated secretory pathway is depicted. Immature secretory granules (iSGs) budding from the trans-Golgi network (TGN) contain prohormones and processing enzymes like the prohormone convertases (PCs), granins [chromogranins A and B (ChgA/B) and secretogranins II and II], and PAM. Mature secretory granules (mSGs) release their soluble content during regulated exocytosis (the yellow lightning bolt represents the external stimulus triggering secretion). Although PAM appears on the cell surface during exocytosis, its rapid endocytosis means that very little PAM remains on the plasma membrane at the steady state. Cleavage by secretase-like enzymes can generate a soluble, cytosolic fragment of PAM that can enter the nucleus, where it relays information about the status of the secretory granule pool. Several in vitro studies have shown that PAM misexpression can affect a variety of steps in the regulated secretory pathway; these are highlighted in red in the cartoon. ER, endoplasmic reticulum; GH, growth hormone; MVB, multivesicular body.