Genetic deletion of the desmosomal component desmoplakin promotes tumor microinvasion in a mouse model of pancreatic neuroendocrine carcinogenesis

Abstract

We used the RIP1-Tag2 (RT2) mouse model of islet cell carcinogenesis to profile the transcriptome of pancreatic neuroendocrine tumors (PNET) that were either non-invasive or highly invasive, seeking to identify pro- and anti-invasive molecules. Expression of multiple components of desmosomes, structures that help maintain cellular adhesion, was significantly reduced in invasive carcinomas. Genetic deletion of one of these desmosomal components, desmoplakin, resulted in increased local tumor invasion without affecting tumor growth parameters in RT2 PNETs. Expression of cadherin 1, a component of the adherens junction adhesion complex, was maintained in these tumors despite the genetic deletion of desmoplakin. Our results demonstrate that loss of desmoplakin expression and resultant disruption of desmosomal adhesion can promote increased local tumor invasion independent of adherens junction status.

Figure 1. Desmoplakin and cadherin 1 expression in wild-type islets and RT2 PNETs.

Expression of desmoplakin (Dsp) and cadherin 1 (Cdh1) is lost in the IC2 but not the IT grade of PNET in RT2 mice. (A–C) H&E staining of a normal islet from a wild-type B6 mouse and of an IT and an IC2 tumor from an end-stage RT2 mouse. Dashed lines demarcate tumor margins. (D–F) Immunofluorescence staining with DAPI to visualize cellularity. (G–I) Immunofluorescence staining for Cdh1. (J–L) Immunofluorescence staining for Dsp. (M–O) Merge of Cdh1 and Dsp immunofluorescence staining (G–L). (P–R) Higher magnification of the boxed regions in M–O. Scale bars represent 200 µm (A–O) and 100 µm (P–R).

Figure 2. Genetic deletion of desmoplakin does not affect tumor growth parameters in RT2 PNETs.

Conditional genetic deletion of Dsp in angiogenic islet dysplasias and incipient solid tumors does not affect tumor formation or tumor growth parameters in RT2 mice. (A–B) Tumor burden and tumor number in RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. Cre activity was induced at 10 weeks, and mice were sacrificed at 14 weeks. Data shown are individual values plus mean. Groups are not statistically different for these metrics. (C–E) Ki67 staining on tumors from RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. (F–H) TUNEL staining on tumors from RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. (I) Quantification of C–E. Data shown are mean plus standard error. Groups are not statistically different. (J) Quantification of F–H. Data shown are mean plus standard error. Groups are not statistically different. Scale bars represent 100 µm (C–H).

Figure 3. Genetic deletion of desmoplakin leads to increased local tumor invasion in RT2 mice.

Conditional genetic deletion of Dsp in angiogenic islet dysplasias and incipient solid tumors increases the rate of progression to focally invasive IC1 tumors in RT2 mice. (A–C) H&E staining of a non-invasive IT tumor lesion, a focally invasive IC1 tumor lesion, and a broadly invasive IC2 tumor lesion from RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. (D–F) Higher magnification of the boxed regions in A–C. T indicates tumor region and Ex indicates exocrine pancreas. Dashed lines demarcate tumor margins. Arrowheads indicate regions of tumor invasion. (G) Quantification of tumor invasiveness represented as the percentage of IT lesions or total IC lesions (IC1+IC2) in RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice at 14 weeks of age. A minimum of 36 tumors per group was graded. * p<0.01 by Fisher's exact test. (H) Same as G except IC lesions are separated into the IC1 and IC2 subclasses. * p<0.01 by the Chi-square test. (I) Tumors from RT2+; Pdx1-Cre^ER+; Dsp^WT/WT, RT2+; Pdx1-Cre^ER+; Dsp^Flox/WT, and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice were genotyped for the presence of the Cre recombinase (∼530 bp), β2 microglobulin (∼290 bp), and the floxed (∼360 bp) or wild-type Dsp allele (∼230 bp). These same tumors were assessed for the recombination status of Dsp: wild-type allele (∼960 bp), non-recombined floxed allele (∼1200 bp), recombined floxed allele (∼650 bp). Scale bars represent 400 µm (A–C) and 200 µm (D–F).

Figure 4. Genetic deletion of desmoplakin does not affect cadherin 1 expression in RT2 PNETs.

Cdh1 expression is maintained in the IC1 grade of tumors in both RT2+; Pdx1-Cre^ER+; Dsp^WT/WT and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. (A–D) Immunofluorescence staining with DAPI to reveal cellularity in IT and IC1 tumors in RT2+; Pdx1-Cre^ER+; Dsp^WT/WT and RT2+; Pdx1-Cre^ER+; Dsp^Flox/Flox mice. (E–H) Immunofluorescence staining for the oncoprotein T antigen (Tag). (I–L) Immunofluorescence staining for Cdh1. (M–P) Merge of Tag and Cdh1 immunofluorescence staining (E–L). (Q–T) Higher magnification of the boxed regions in M–P. Arrowheads indicate regions of tumor invasion. Scale bars represent 200 µm (A–P) and 100 µm (Q–T).

Figure 5. Progression to an invasive growth state is governed by multiple factors in the RT2 model of PNET.

Multiple factors impact the progression to an invasive phenotype as illustrated by the RIP1-Tag2 (RT2) mouse model of pancreatic neuroendocrine tumorigenesis (PNET). This study demonstrates that the genetic deletion of desmoplakin (Dsp) and concomitant loss/attenuation of desmosomal adhesion can promote local tumor invasion, specifically to a focally invasive state typified by the IC1 tumor class. Other factors have also been demonstrated to affect tumor invasion in this model. Activation of heparanase or of cathepsin proteases – supplied by infiltrating immune cells or suppression of cadherin 1 (Cdh1, also known as E-cadherin) can each contribute to invasion. Upregulation of the type-1 insulin-like growth factor receptor (Igf1r) preferentially promotes progression to the IC2 stage, in part via a branched pathway from earlier neoplastic stages such as angiogenic islet dysplasias (AI), bypassing the canonical AI→IT→IC1→IC2 progression . Future research may well identify additional factors that impact tumor invasion.