Colorectal Cancer Is a Paracrine Deficiency Syndrome Amenable to Oral Hormone Replacement Therapy

Abstract

The most commonly lost gene products in colorectal carcinogenesis include the paracrine hormones guanylin and uroguanylin, the endogenous ligands for guanylyl cyclase C (GCC), the intestinal receptor for diarrheagenic bacterial enterotoxins. Recently, GCC‐cyclic guanosine monophosphate (GMP) signaling has emerged as a principal regulator of proliferation, genetic integrity, and metabolic programming in normal human enterocytes and colon cancer cells. Elimination of GCC in mice produced hyperplasia of the proliferating compartment associated with increases in the rapidly cycling progenitor cells and reprogrammed enterocyte metabolism, with a shift from oxidative phosphorylation to glycolysis. In addition, in the colons of mice carrying mutations in adenomatous polyposis coli gene Apc (Apc^Min/+) or exposed to the carcinogen azoxymethane, elimination of GCC increased tumor initiation and promotion by disrupting genomic integrity and releasing cell cycle restriction. These previously unrecognized roles for GCC as a fundamental regulator of intestinal homeostasis and as an intestinal tumor suppressor suggest that receptor dysregulation reflecting paracrine hormone insuffciency is a key event during the initial stages of colorectal tumorigenesis. Together with the uniform overexpression of GCC in human tumors, these novel roles for GCC underscore the potential of oral replacement with GCC ligands for a targeted prevention and therapy of colorectal cancer.

Introduction

Colorectal cancer, arising as a malignant transformation of the intestinal epithelial cells, is the third leading cause of cancer and second leading cause of cancer mortality in the United States and the world. ¹ Conventionally, colon cancer is defined as a genetic disease, reflecting sequential accumulation of mutations in oncogenes, tumor susceptibility genes, or tumor suppressors, ² most frequently (>80%) including sporadic or inherited alterations in the gene encoding adenomatous polyposis coli (APC). ³ , ⁴ However, a novel paradigm is emerging, suggesting that at inception, intestinal neoplasia evolves from a state of hormone insufficiency. ⁵ , ⁶ , ⁷ Indeed, the most commonly lost gene products early in colorectal carcinogenesis include the paracrine hormones guanylin and uroguanylin, ⁸ , ⁹ , ¹⁰ , ¹¹ the endogenous ligands for guanylyl cyclase C (GCC), the intestinal receptor for diarrheagenic bacterial enterotoxins. ¹² In that context, oral administration of uroguanylin reduced both tumor number and size in the intestines from Apc^Min/+ mice. ⁵ Moreover, activation of GCC by the exogenous ligand, bacterial heat‐stable enterotoxin (ST), restricted the proliferation of human colorectal cancer cells by inducing a calcium influx and cytostasis‐antagonizing cell growth. ⁶ , ¹³ , ¹⁴ These observations suggest that sporadic colorectal tumorigenesis is a process initiated by the loss of paracrine hormone expression, inducing a state of guanylinopenia and uroguanylinopenia, resulting in dysregulation of GCC and its downstream signaling. ⁷

The intestinal epithelium is dynamically renewing, continuously undergoing homeostatic cycles of proliferation, migration, differentiation, apoptosis, and shedding, which maintains organ‐specific functions, including digestion, absorption, secretion, and barrier function. ¹⁵ Transit cells originate from slowly proliferating stem cells near the base of crypts and undergo sequential cycles of division. Proliferating transit cells migrate along the vertical axis to the differentiated compartment where homo‐ and heterotypic interactions induce a systems‐level reprogramming of nuclear and cytoplasmic circuits, coordinating proliferative restriction, lineage commitment, genomic integrity, and metabolic reprogramming. Disruption of homeostatic renewal, especially the transition from proliferating progenitor to terminally differentiated cell, results in mucosal hyperplasia and a maturational shif that may be a principal mechanism contributing to the development of colon cancer. ¹⁵

GCC, exclusively expressed in apical brush border membranes of intestinal epithelial cells, is the only identified receptor for diarrheagenic bacterial STs, the principal pathobiological agents mediating endemic secretory diarrhea in animals and humans worldwide. ¹⁶ STs reflect molecular mimicry coupled with convergent evolution, whereby microorganisms co‐opt a normal physiological mechanism that confers adaptive survival advantages. Indeed, STs are structurally and functionally homologous to two endogenous intestinal paracrine hormones, guanylin and uroguanylin, which are the endogenous ligands for GCC. ¹⁶ , ¹⁷ Ligation of the extracellular receptor domain of GCC by endogenous or exogenous ligands activates the canonical cytoplasmic catalytic domain, resulting in the accumulation of intracellular cyclic GMP (cGMP). This cyclic nucleotide activates cGMP‐dependent protein kinase (PKG) that phosphorylates the cystic fibrosis transmembrane conductance regulator (CFTR), resulting in an efflux of salt and water that, in the case of STs, manifests as secretory diarrhea. ¹⁶ GCC and its endogenous ligands were considered the physiological regulators of the intestinal fluid and electrolyte homeostasis. ¹⁷ However, beyond the regulation of fluid and electrolyte secretion, studies have revealed key roles for GCC and cGMP in maintaining intestinal crypt–villus homeostasis and suppressing intestinal tumorigenesis. ⁵ , ⁶ , ⁷ , ¹³ , ¹⁴ , ¹⁸ , ¹⁹

GCC Signaling Regulates Integrated Dynamic Homeostatic Systems Along the Intestinal Crypt‐Surface Axis

The intestinal mucosa is lined by a single layer of epithelial cells organized in vertical anatomical units underlying specialized organ functions. In small intestine, villi projecting into the lumen and flask‐like crypts embedded in the mesenchyme expand the secretory and absorptive surface and provide the structure supporting digestion, absorption, and secretion, in part, by increasing the surface area. ²⁰ In contrast, the large intestine exhibits a comparatively smooth surface, with tubular crypts embedded in the colonic mesenchyme. ¹⁵ In the small intestine, the crypts form the proliferating zone populated with regenerative stem cells and rapidly proliferating transit cells. ¹⁵ The proliferating compartment is the source of cells contributing to homeostatic epithelial renewal and is tightly controlled by both pro‐ and antiproliferative signaling. Although the mechanisms remain incompletely defined, the signaling pathways contributing to organizing and maintaining the crypt–villus axis, include Wnt/β‐catenin/Tcf‐4 pathway, ²¹ , ²² Notch pathway, ²³ , ²⁴ transfoming growth factor signaling, ²⁵ Ca²⁺, ²⁶ , ²⁷ the transactivation factors CDX1 and 2, ²⁸ , ²⁹ , ³⁰ and the family of integrins. ³¹ Disruption of these signaling pathways results in crypt hyperplasia, altering the organization of the crypt‐surface axis and contributing to intestinal tumorigenesis. ³²

In contrast to the transit cells in the proliferating compartment, the main characteristics of mature enterocytes include well‐developed microvillus brush border membranes containing key functional proteins mediating cognate digestive and absorptive functions. Goblet cells are mucin‐secreting cells protecting the intestinal lumen and facilitating enterocytes nutrient absorption. ³³ Enteroendocrine cells produce autacoids, peptides, and hormones and are part of the enteral neuroendocrine system, with paracrine and autocrine functions locally and endocrine functions supporting parenteral systems. ³³ Finally, Paneth cells protect the mucosa by establishing a functional barrier, secreting antimicrobial peptides, digestive enzymes, and growth factors into the lumen. ³⁴ Paneth cells, absent in the colon, are one of the mechanisms defending the small intestine against tumorigenesis through innate immune responses. ³⁴

Beyond proliferative restriction, the signaling pathways coordinately regulate the requisite metabolic reprogramming associated with cell cycle exit and lineage specification. ³⁵ , ³⁶ , ³⁷ There is an established gradient of metabolism along the vertical crypt‐surface axis. Glycolysis prevails in the crypts, exploiting the capacity for ATP generation, reflecting an accelerated metabolic substrate throughput to support anabolic bioenergetic demands in the rapidly proliferating cells. Conversely, mitochondrial metabolism predominates in the villi, capitalizing on the efficiency of adenosine‐5′‐triphosphate (ATP) production through oxidative phosphorylation to support catabolic demands in well‐differentiated cells. ³⁵ , ³⁶ , ³⁷

Elimination of GCC (GCC^−/−) ¹⁹ or guanylin ¹⁸ in mice produced hyperplastic crypts along the rostral–caudal axis, which contained more proliferating cells, quantified by immunohistochemistry (IHC) for proliferating cell nuclear antigen (PCNA) and Ki67, associated with an accelerated cell cycle, quantified by 5′‐bromo‐2′‐deoxyuridine (BrdU) kinetics. Hyperproliferation in the absence of GCC signaling reflected an increased expression of key mediators driving the cell cycle, including β‐catenin, an established downstream target of cGMP ³⁸ that enhances the proliferation through T‐cell factor (TCF)4‐dependent transcription of cell cycle intermediates. In that context, elimination of GCC increased the expression of cyclin D1 and phosphorylated retinoblastoma (pRb), the downstream targets of β‐catenin/TCF4 transcriptional regulation mediating G1–S transition. Moreover, elimination of GCC signaling decreased the expression of critical cell cycle suppressors, including the p27 cyclin‐dependent kinase inhibitor. ¹⁹ , ³⁶

Beyond proliferation, elimination of GCC signaling specifically altered the proliferating transit cell commitment to the differentiated secretory lineage, and GCC^−/− mice exhibited fewer Goblet cells, quantified by alcian blue staining and immunoblot for intestine trefoil factor, and Paneth cells, quantified by enumerating mucosal cells stained with antilysozyme antibody. ¹⁹ In contrast, elimination of GCC was without effect on the commitment of proliferating transit cells to the enteroendocrine or absorptive enterocyte lineages. ¹⁹ This specific impairment of lineage commitment suggests that GCC selectively regulates differentiation along the crypt‐surface axis through discreet molecular mechanisms, rather than a passive consequence of dysregulation of the proliferating compartment. ¹⁹

In conjunction with transitioning cells from the cell cycle to lineage commitment, GCC coordinates the associated requisite metabolic reprogramming by balancing mitochondrial biogenesis and glycolysis along the crypt‐surface axis. ³⁶ Elimination of GCC disrupted the metabolic switch from glycolysis to oxidative phosphorylation, characterizing the transition from proliferation to differentiation along the crypt–villus axis. ³⁶ GCC^−/− mice exhibited a reduction in mitochondrial biogenesis, quantified by decreases in the principal mitochondrial proteins including cytochrome oxidase, core II, and ATP synthase. A reduction in the mitochondrial proteins was associated with a loss of mitochondrial DNA and diminished mitochondrial function in GCC^−/− mice, characterized by reduced mitochondria‐specific oxygen consumption, dehydrogenase activity, and steady‐state ATP concentrations. ³⁶ Conversely, disruption of GCC signaling reciprocally increased glycolytic metabolism, characterized by the induction of key rate‐limiting proteins including glucose transporter 1 (Glut 1), hexokinase II, phosphofructokinase, and lactate dehydrogenase. Moreover, the induction of glycolysis in GCC^−/− mice was associated with an increased lactate production and glucose transport. ³⁵ , ³⁶ , ³⁷ Taken together, these observations suggest that GCC signaling critically regulates the dynamic homeostasis of the crypt‐surface axis by coordinating the cell cycle, differentiation, and metabolic programming and reciprocally controlling the statutory balance between proliferation and lineage specification.

GCC Suppresses Initiation and Promotion of Intestinal Tumorigenesis

Disruption of tightly organized homeostatic mechanisms organizing the crypt‐surface axis through the inactivation of antiproliferating signaling produces intestinal tumorigenesis. ³⁰ , ³⁹ Indeed, hyperplasia of the intestinal epithelial compartment is one of the greatest risk factors, and an immediate precursor lesion, of colorectal cancer. ³ , ³⁰ Unrestricted proliferation not only produces overgrowth of the epithelial cells, but also promotes DNA damage and accelerates chromosomal instability by inducing premature entry into, and accumulation in, S phase ⁴⁰ , ⁴¹ , ⁴² without a coordinated activation of apoptosis to maintain a dynamic balance across compartments or eliminate damaged cells.

Elimination of GCC signaling increases the size of the proliferating crypt compartment, the number of proliferating cells in that compartment, and accelerates their cell cycle. ¹⁸ , ¹⁹ These effects are potentiated by genotoxic insults, revealed by hyperproliferation of normal intestinal epithelium induced by the elimination of GCC in mice carrying inactivating mutations in Apc (Apc^Min/+) or exposed to the chemical carcinogen azoxymethane (AOM). Corrupting the regulation of proliferation and accelerating the cell cycle by eliminating GCC signaling promotes tumor growth in Apc^Min/+ and AOM models, as reflected by the contribution of adenoma size to the increased tumor burden and the associated crypt hyperplasia in normal adjacent mucosa. ⁷

Cell cycle acceleration in tumorigenesis is often associated with chromosomal instability, reflecting DNA damage generated by premature entry into S phase, ⁴¹ or compromised DNA damage detection and repair due to failed cell cycle arrest at the S/G2 interface in response to checkpoint signaling. ⁴³ Unexpectedly, even in the absence of hyperproliferation, elimination of GCC signaling disrupts the genetic circuits maintaining chromosomal integrity. ⁷ In the colons of Apc^Min/+ mice, elimination of GCC increased tumor initiation, as reflected by both increased tumor incidence and tumor multiplicity, without altering the proliferation of colonocytes. ⁷ Tumor initiation produced by the elimination of GCC signaling was associated with an increased DNA damage in the colonocytes of wild‐type mice in the absence of genotoxic insults. DNA damage was potentiated in the colonocytes of Apc^Min/+ mice in the absence of GCC, quantified by phosphorylated phosphorylated histone 2AX (γ‐H2AX), a marker of double‐strand DNA breaks (DSBs). ⁷ , ⁴⁴ In that context, a loss of heterozygosity at the Apc locus ⁴⁵ , ⁴⁶ occurred in 80% of tumors from Apc^Min/+Gcc^{−/ −} mice, compared to 10% in Apc^Min/+Gcc^+/+ mice, reflecting genomic instability associated with DSBs induced in the absence of GCC signaling. ⁷ Moreover, elimination of GCC signaling doubled tumor multiplicity in mice exposed to AOM and increased the rate of single‐base mutations in β‐catenin, the key mutation inducing transformation in mice exposed to this carcinogen. A discreet partitioning of the regulation of proliferation and chromosomal stability implies that GCC signaling contributes to genomic integrity through independent, but mutually reinforcing, mechanisms collaborating across cell cycle and DNA damage and repair machineries. ⁷ The precise contribution of GCC signaling to the steady‐state maintenance of the genome, including damage detection and assessment, mutational repair, and the associated coordination of replicative decision‐making remains to be defined. ⁴² , ⁴³

Taken together, these data suggest that GCC signaling suppresses intestinal neoplastic transformation by coordinating the restriction of cell proliferation and maintenance of chromosomal stability. In that context, disruption of GCC signaling reflecting the early loss of guanylin and uroguanylin expression might be a key initiating event in intestinal tumorigenesis, releasing cell cycle restriction and compromising genetic integrity, producing downstream accumulation of genetic mutations in tumor suppressors or oncogenes promoting tumor progression.

GCC Is a Specific Hormonal Target for Colon Cancer Prevention and Treatment

GCC expression is restricted to intestinal epithelial cells. Indeed, GCC expression has been detected in more than 1,000 samples of normal intestine from human, but not in >1,000 extragastrointestinal tissues, including the kidney, pancreas, lung, or biliary epithelium. ⁴⁷ , ⁴⁸ , ⁴⁹ GCC expressed in the intestine is functionally active, quantified by competitive radioligand binding and cGMP production. ⁴⁷ , ⁴⁸ Compartmentalization in the apical membranes, ⁵⁰ on the lumenal side of the epithelial barrier, suggests that GCC is functionally restricted to the mucosae, confirmed by radioligand imaging and biodistribution studies. ⁵¹ , ⁵² , ⁵³ These observations underscore the specificity of GCC as a target universally expressed only by intestinal epithelial cells, but not by cells outside the intestine, limiting collateral of‐target effects.

Unlike its endogenous ligands, whose expression is lost early in tumorigenesis, GCC continues to be expressed uniformly by the intestinal cells throughout neoplastic transformation. ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ GCC expression has been detected in >1,000 primary and metastatic colorectal tumors examined to date. ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ The novel role of GCC in suppressing intestinal tumorigenesis, together with its unique restricted pattern of expression and the near‐absolute specificity of its endogenous and exogenous ligands, make GCC an unambiguous structural and mechanism‐based target for the prevention and treatment of colon cancer. In the context of the standard of care in which hormone defficiencies are treated by replacement, the observation that intestinal paracrine hormones antagonize tumorigenesis through GCC underscores the potential for GCC ligand supplementation for a targeted prevention and therapy in colorectal cancer.

In that context, oral supplementation with uroguanylin suppresses intestinal tumor initiation and growth in Apc^min/+ mice. ⁵ Further, in human colorectal cancer cells, the exogenous ligand, ST, induces accumulation of intracellular cGMP, ⁶ , ¹³ , ¹⁴ producing an inhibition of cell growth ⁶ , ¹³ , ¹⁴ , ¹⁹ by suppressing DNA synthesis, ⁶ , ¹³ , ¹⁴ quantified by cell enumeration, protein content, and ³H‐thymidine incorporation into DNA. Moreover, GCC signaling induces transient cell cycle arrest at the G1/S interface, ¹⁹ quantified by flow cytometry, associated with a decreased expression of key cell cycle mediators, including cyclin D1 and pRb, but an increased expression of the cell cycle inhibitor p27 regulating the transition through G1 to S in human intestinal cells. ¹⁹

Cell cycle regulation opposing the proliferation induced by GCC signaling is associated with the maintenance of chromosomal stability, reflecting a regulation in the production of DNA damaging agents and sensitization of the repair machinery. Activation of GCC signaling by ST in human colon carcinoma cells opposes the production of reactive oxygen species (ROS), ³⁶ the principal endogenous source of DNA oxidative damage that potentiates proliferation and genomic instability, contributing to tumor initiation and progression. ⁵⁴ Beyond production of DNA damage, GCC signaling in human colon cancer cells sensitizes reparative responses to DNA damage, including an increased phosphorylation of checkpoint homolog (Chk2), a key component of the machinery coordinating DNA damage detection and repair with cell cycle restriction. ⁴³ Upon DNA damage, the phosphorylation of Chk2 activates checkpoint control and arrest of the cell cycle to enable engagement of replicative decision‐making circuits specifying dichotomous cell fates along reparative or apoptotic pathways. ⁴⁰ , ⁴³

Moreover, GCC signaling reverts the tumorigenic metabolic phenotype in mouse and human colon cancer cells following treatment with ST. ³⁶ Ligand‐dependent activation of GCC, or its downstream efector cGMP, in human colon cancer cells transcriptionally induces the expression of the suite of critical transcription factors required for mitochondrial biogenesis, including proliferation activator receptor‐γ coactivator 1 (PGC1α), mitochondrial transcription factor A (mtTFA), and nuclear respiratory factor a (NRF1). ³⁶ Further, GCC increases the content of mitochondria and their associated genome and proteome, ³⁶ associated with an enhanced mitochondrial oxygen consumption, dehydrogenase activity, and ATP production. ³⁶ Moreover, GCC signaling reciprocally inhibits glycolysis and fatty acid synthesis, metabolic characteristics of rapidly proliferating tumor cells, by reducing key rate‐limiting enzymes including the glucose transporter, hexokinase, phosphofructokinase, and acid citrate lyase, tightly associated with a decrease in lactate production and glucose uptake. ³⁶

Future Perspectives

GCC signaling regulates critical homeostatic mechanisms maintaining the dynamic intestinal crypt‐surface axis that oppose tumorigenesis resulting from genotoxic stress by coordinating proliferative restriction, lineage specification, metabolic programming, and genetic integrity. ⁷ , ¹⁸ , ¹⁹ Disruption of GCC signaling reflecting near‐universal loss of paracrine hormone expression early in neoplastic transformation corrupts these fundamental systems‐level processes, resulting in a mutually reinforcing maladaptive evolution of cellular functions and the development of the tumorigenic phenotype. These novel insights into the molecular pathogenesis suggest an additional dimension to the current model in which tumorigenesis reflects the accumulation of mutations sequentially compromising homeostatic processes, co‐opting corrupted circuits to produce a selection advantage permitting the emergence of invasive cancer cells. Beyond this view, the present observations suggest a previously unanticipated pathophysiological paradigm in which colorectal cancer, in part, initiates as a disease of hormone insufficiency in which the disruption of a single key homeostatic signaling mechanism produces a simultaneous systems‐level disruption of fundamental homeostatic processes and an exponential leap in maladaptive evolution along the transformation continuum. Beyond pathogenesis, this emerging model suggests a correlative therapeutic hypothesis in which colorectal cancer can be prevented and treated employing the well‐established paradigm originating in endocrinology in which diseases of hormone insufficiency are resolved by hormone replacement therapy. In that context, the near‐universal loss of guanylin and uroguanylin early along the transformation continuum, ⁸ , ⁹ , ¹⁰ , ¹¹ the reciprocal compensatory overexpression of GCC in colorectal cancer cells, ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ and the compartmentalization of GCC in brush border membranes sampling the lumenal environment ⁵⁰ support the utility of oral GCC ligand replacement therapy to prevent and treat colorectal cancer. Oral replacement therapy should restore and preserve intestinal paracrine hormone balance and defend homeostatic processes including proliferative restriction, lineage specification, metabolic programming, and genomic stability that critically oppose intestinal neoplastic transformation.