Introduction
Colorectal cancer (CRC) is the third most common cancer and the second leading cause of cancer-related death worldwide (Sung et al., 2021). Surgical resection of the primary CRC tumor is the mainstay of treatment (Argilés et al., 2020; Cunningham et al., 2010; Seo et al., 2013) unfortunately, up to 50% of these patients – despite chemotherapy and targeted adjuvant therapies – often develop life-threatening liver metastatic disease in the following years (Argilés et al., 2020; Sargent et al., 2009). While the overall benefit of surgery is well established (Seo et al., 2013), it has been also proposed that this procedure may foster liver metastases by increasing the dissemination of CRC cells into the portal circulation (Chow and Chok, 2019; Denève et al., 2013), enhancing the adhesion of CRC cells to the liver endothelium (Chambers et al., 2002; Gül et al., 2011) or promoting transient immunosuppression awakening dormant intrahepatic micrometastases (Ananth et al., 2016).
Accordingly, there is growing recognition that the use of perioperative immunotherapies in CRC patients undergoing surgical resection may represent a unique treatment window to prevent metastatic colonization and control minimal residual disease (Badia-Ramentol et al., 2021; Bakos et al., 2018; Horowitz et al., 2015). In this context, interferon-alpha (IFNα), a pleiotropic cytokine with multiple antitumor effects such as the direct inhibition of cancer cell growth and angiogenesis (Indraccolo, 2010), the sustained upregulation of major histocompatibility complexes (Gessani et al., 2014) and the induction of innate and adaptive antitumor immune responses (Aichele et al., 2006; Curtsinger et al., 2007; Fuertes et al., 2013), has been used as adjuvant immunotherapy in various solid cancers such as renal cell carcinoma (Flanigan et al., 2001), melanoma (Lens and Dawes, 2002) and colorectal cancer (Köhne et al., 1997; Link et al., 2005). Unfortunately, systemic administration of IFNα has shown limited clinical efficacy, likely due to its short plasma half-life (~1 hr) (Bocci, 1994) and the use of high and pulsed doses, which often resulted in systemic side effects (Weber et al., 2015). To overcome these limitations, several strategies to prolong IFNα half-life and target the tumor microenvironment have been tested (Fioravanti et al., 2011; Herndon et al., 2012; Jeon et al., 2013; Li et al., 2017; Liang et al., 2018; Yang et al., 2014), including a preclinical gene/cell therapy approach that can deliver constant amounts of IFNα into the liver to significantly curb CRC metastatic growth (Catarinella et al., 2016).
Herein, we adopted a continuous intraperitoneal (ip) IFNα delivery strategy to show that steady and tolerable IFNα doses reduce liver CRC metastatic spreading and improves survival in several CRC mouse models. Our results showed that the antimetastatic effects of IFNα rely neither on the direct inhibition of tumor cell proliferation nor on the indirect stimulation of hepatocytes, hepatic stellate cells, liver DCs, Kupffer cells (KCs) and liver capsular macrophages (LCMs). Rather, the results identify HECs, including LSECs, as key mediators of IFNα-dependent anti-tumor activities that involve the impairment of CRC trans-sinusoidal migration and the development of long-term anti-tumor CD8+ T cell immunity.
Results
Selection of the optimal IFNα dosing regimen
To avoid well-known toxicities, especially myelotoxicity, caused by high IFNα doses (Weber et al., 2015) and to define a delivery strategy providing prolonged and non-fluctuating IFNα levels in blood and tissues, normal inbred mice were implanted intraperitoneally with mini-osmotic pumps (MOP) constantly releasing different rates (i.e. 50 ng/day, 150 ng/day, or 1050 ng/day) of recombinant mouse IFNα1 (termed IFNα from now on) over time. Serum IFNα levels peaked at day 2 after MOP implantation and relative IFNα amounts (from ~100 pg/ml to ~1200 pg/ml) reflected the different MOP loading doses (Figure 1A). Serum IFNα levels decreased, albeit not uniformly, at days 5 and 7 post implantation (Figure 1A), mirroring the pharmacokinetic-pharmacodynamic (PK-PD) behavior of other long-lasting formulations of IFNα (Jeon et al., 2013). A reduction in circulating white blood cells (WBCs) but not in platelet (PLT) counts or hematocrit (HCT) was detected only with the highest dose (Figure 1B and Figure 1—figure supplement 1A,B). Looking at liver toxicity, we observed no increases in serum alanine aminotransferase (sALT) at all tested doses and time points (Figure 1—figure supplement 1C) and no abnormal changes in liver morphology at autopsy (Figure 1—figure supplement 1D). Looking at the intrahepatic induction of the interferon-stimulated gene (ISG)
Figure 1.
Selection of the optimal IFNα dosing regimen.
(A) Quantification of plasma IFNα concentration from mice continuously treated with different IFNα doses at the indicated time points. Mean values ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values refer to the IFNα 50 ng/day group, since NaCl-treated animals had IFNα plasma levels below the assay detection limit. p≤0.05; **p≤0.01; ***p≤0.001. (B) White blood cell (WBC) counts of mice treated with different IFNα doses at indicated time points. Horizontal dashed lines delimit normal WBC range. Mean value ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values are referred to the NaCl group. *p≤0.05; **p≤0.01. (C) Quantitative real-time PCR analysis of
Figure 1—figure supplement 1.
Selection of the optimal IFNα dosing regimen.
(A) Platelet (PLT) count from mice treated with NaCl (n=9) or different IFNα doses at the indicated time points (n=3 each). Horizontal dashed lines delimit normal PLT range. Mean value ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values are referred to NaCl group. **p≤0.01. (B) Hematocrit (HCT) from mice treated with NaCl or different IFNα doses at indicated time points. Horizontal dashed lines delimit normal HCT range. Mean value ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values are referred to NaCl group. **p≤0.01. (C) Serum Alanine Aminotransferases (sALT) quantification in plasma of mice treated with NaCl or different IFNα doses at indicated time points. Horizontal dashed line delimits normal sALT range. Mean values ± SEM are shown. (D) Representative hepatic H&E micrographs from NaCl and IFNα-treated mice with the indicated IFNα doses 7 days after MOP implantation. Scale bar = 100 µm. (E) Representative microscopic images of phospho-STAT1 (pSTAT1) IHC of the liver of mice treated continuously with NaCl or 150 ng/day IFN for 7 days and euthanized 30 min after ip injection with NaCl or 1 µg IFNα. Scale bar = 50 µm. Note that pSTAT1 is expressed only in small clusters of cells in close proximity to the portal circulation (i.e. HECs, stellate cells, or leukocytes) in mice continuously treated with IFNα at day 7 after treatment initiation. Of note, injection of pSTAT1-synchronising doses of IFNα resulted in complete activation of all hepatic cells in both groups of mice, suggesting that IFNα-dependent counterregulatory mechanisms are not induced by continuous IFNα treatment.
Continuous IFNα administration reduces liver CRC metastatic burden and improves survival
We next tested the ability of continuous IFNα administration (150 ng/day for 28 days) to reduce CRC metastatic growth in the liver. Groups of H-2bxd F1 hybrids of C57BL/6 J x BALB/c (CB6) mice were implanted with either control MOP-NaCl (termed NaCl) or MOP-IFNα (termed IFNα) (Figure 2A). Seven days later, a time frame compatible with the perioperative period in humans (Horowitz et al., 2015), CB6 mice were intrasplenically challenged with either the immunogenic microsatellite instable (MSI) MC38 CRC cell line (Efremova et al., 2018; Rosenberg et al., 1986) or the poorly immunogenic microsatellite stable (MSS) CT26 CRC cell line (Brattain et al., 1980; Efremova et al., 2018). Rapid removal of the spleen after CRC cell injection was implemented to avoid intrasplenic tumor growth (Catarinella et al., 2016). Each CRC cell line was injected at doses known to induce similar survival rates in age- and sex-matched CB6 recipients that, carrying hybrid H-2bxd alleles, are immunologically permissive to both MC38 and CT26 cells (Catarinella et al., 2016). After treatment initiation, well-tolerated serum IFNα levels of ~300 pg/ml at day 2 and ~100 pg/ml thereafter were observed (Figure 2B and Figure 2—figure supplement 1A,B), which subsequently declined to undetectable levels. The intrahepatic expression of
Figure 2.
Continuous IFNα administration reduces liver CRC metastatic burden and improves survival.
(A) Schematic representation of the experimental procedure. Intrasplenic injection of 7x104 MC38 or 5x103 CT26 cells was performed 7 days after continuous NaCl or IFNα therapy. (B) Quantification of plasma IFNα concentration at different time points after continuous IFNα administration (n=6). The time of intrasplenic CRC cell injection has also been depicted. Mean values ± SEM are shown. (C–D) Representative T1 contrast-enhanced magnetic resonance images (MRI) of the liver of mice treated with NaCl (red frame) and IFNα (blue frame) at 21, 28, and 50 days after MC38 (C) or CT26 (D) cells injection. Red dashed lines highlight CRC liver metastases, characterized as hypointense regions in T1-weighted sequences. n.a.=not assessed, is referred to mice euthanized before the specified time point; scale bar = 5 mm. (E–F) Percentage of mice treated with NaCl (MC38 n=3 + 3; CT26 n=5 + 5 for each of two independent experiments) or IFNα (MC38 n=5 + 6; CT26 n=5 + 6 for each of two independent experiments) bearing at least one CRC liver metastasis estimated by MRI analysis at indicated time points after MC38 or CT26 injection. The oblique black line pattern within columns depicts the percentage of mice euthanized before the indicated time point. Mean values are shown; p-values were calculated by Fisher’s exact test. (G–H) Kaplan-Meier survival curves of Sham (n=3), NaCl- (MC38 n=6; CT26 n=10) or IFNα-treated (MC38 n=11; CT26 n=11) mice after MC38 or CT26 cells injection. The blue pattern indicates the time frame of IFNα ip release; p-values were calculated by log-rank/Mantel-Cox test.
Figure 2—figure supplement 1.
Continuous intraperitoneal IFNα administration reduces MC38 and CT26 metastatic tumor burden without causing significant hematologic toxicity.
(A–B) WBC counts of mice treated with NaCl or IFNα and injected with MC38 (A) or CT26 (B) at the indicated time points. Horizontal dashed lines delimit normal WBC range. (Arrows indicate the time of MC38 or CT26 cells injection.) Mean values ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values are referred to NaCl group. *p≤0.05; **p≤0.01; ***p≤0.001. (C–D), Quantitative real-time PCR analyses of the relative expression levels of the
Continuous IFNα administration prevents spontaneous hepatic colonization of orthotopically implanted CT26LM3 cells
To confirm the above-mentioned results in a different metastatic setting, we developed an orthotopic CRC model of liver metastases by implanting invasive CRC cells into the mouse cecal wall. As previously reported (Zhang et al., 2013), invasive CRC cells were generated by serial intracecal injections of the parental CT26 cells into CB6 mice (Figure 3—figure supplement 1A). The percentage of metastatic livers in intracecally implanted mice significantly increased as CT26 cells were passaged, with an almost 100% of animals bearing multiple liver metastases after 3 rounds of in vivo selection (Figure 3—figure supplement 1B-D). Three-time passaged cells (termed CT26LM3) were then orthotopically implanted in the cecal wall of CB6 mice and 7 days later the animals were treated with either NaCl or IFNα (Figure 3A).
Figure 3.
Continuous IFNα administration prevents spontaneous hepatic colonization of orthotopically implanted CT26LM3 cells.
(A) Schematic representation of the experimental procedure. Seven days after intracecal injection of 2x105 CT26LM3 cells, mice were randomly assigned to receive either continuous NaCl (n=7) or IFNα (n=6) administration and analyzed by MRI 14 days later. (B) Quantification of plasma IFNα concentration at the indicated time point after cecal wall injection of CT26LM3 cells in mice described in a. The arrow indicates the time of NaCl or IFNα therapy initiation. Mean values ± SEM are shown. (C) WBC counts from mice described in (A) continuously treated with NaCl or IFNα at indicated time points. Horizontal dashed lines delimit the normal WBC range. The time of MOP implantation has also been depicted. Mean value ± SEM are shown; p-values were calculated by one-way ANOVA Tukey’s multiple comparison test. Significant p-values are referred to NaCl group. ***p≤0.001. (D) Representative images (top panels) of the hepatic lesions and intracecal tumors observed in NaCl- (red frame) and IFNα-treated (blue frame) mice, 21 days after CT26LM3 cells intracecal wall injection, and the corresponding hepatic contrast-enhanced MRI, T1-weighted (middle panels) and T2-weighted (bottom panels) sequences. Red dashed lines identify macroscopic liver metastatic lesions. Scale bars = 5 mm. (E) Quantification of the weight of primary CRC tumors 21 days after CT26LM3 cells intracecal wall injection of mice described in D. (F) Quantification of the number of hepatic lesions and total tumor volume of liver metastases by MRI analysis of mice described in D. Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test. (G) Representative H&E, Ki67, CD34, F4/80, and CD3 immunohistochemical micrographs of liver metastatic lesions found in NaCl- (red frame) and IFNα-treated (blue frame) mice, 21 days after intracecal injection of CT26LM3 cells. Scale bar = 100 μm. (H) Quantification of the percentage of mice bearing liver metastases, as well as the percentage of tumor area and the number of cells expressing Ki67, CD34, F4/80, and CD3 per mm2 determined by IHC. Immunohistochemical measurements were conducted on at least 1000 mm2 of total liver area for both experimental conditions. Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test.
Figure 3—figure supplement 1.
In vivo selection of CT26LM3 cells with increased spontaneous liver metastatic potential.
(A) Schematic representation of the experimental procedure. (1) 2x106 CT26 cells were grown subcutaneously (sc) c into NSG immunodeficient mice. (2) Excised tumors were disaggregated and cultured in vitro to generate CT26sc cells. (3) 2x106 CT26sc cells were injected into the cecal wall of immunocompetent CB6 mice. (4) After 28 days, the number of liver metastatic foci that arised from primary cecal tumors were counted and (5) metastatic liver tumor cells were isolated and cultured in vitro to generate CT26LM1 cells. (6) 2x105 CT26LM1 cells were injected into the cecal wall of a new immunocompetent mice to obtain CT26LM2 cells. The procedure was repeated with 2x105 CT26LM2 cells to obtain CT26LM3 cells. (B) Percentage of mice bearing at least one macroscopic liver metastasis 28 days after intracecal wall injection of CT26LM1 (n=5), CT26LM2 (n=7) or CT26LM3 (n=5) cells. Mean values are shown; p-value was calculated by Fisher’s exact test. (C) Number of hepatic lesions in the same mice described in (B). Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test. (D) Representative image of the hepatic lesions observed in mice 28 days after intracecal injection of CT26LM3. Black dashed line identifies macroscopic cecal primary tumor whereas white dashed lines delineate liver metastatic lesions.
Figure 3—figure supplement 2.
Immunophenotypic analysis of primary tumors and liver metastases in the orthotopic CT26LM3 model.
(A) Representative H&E, Ki67, CD34, F4/80, and CD3 immunohistochemical micrographs of primary CRC cecal tumors developed in NaCl- (red frame) and IFNα-treated (blue frame) mice, 21 days after intracecal injection of CT26LM3 cells. Scale bar = 100 μm. (B) Quantification of the number of cells expressing Ki67, CD34, F4/80, and CD3 per mm2 determined by IHC. Immunohistochemical measurements were conducted on at least 1000 mm2 of total liver area of sham mice and within all metastatic lesions found in both experimental conditions. Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test. (C) Representative H&E, Ki67, CD34, F4/80, and CD3 immunohistochemical micrographs showing liver tumor characteristics of metastases found in NaCl- (red frame) and IFNα-treated (blue frame) mice, 21 days after intracecal injection of CT26LM3 cells. Scale bar = 100 μm. (D) Quantification of the number of cells expressing Ki67, CD34, F4/80, and CD3 per mm2 determined by IHC. Immunohistochemical measurements were conducted on at least 1000 mm2 of total liver area of sham mice and within all metastatic lesions found in both experimental conditions. Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test.
Consistent with our previous results (Figure 2B), serum IFNα levels peaked at day 2 after MOP implantation (Figure 3B), without causing myelotoxicity (Figure 3C), and MRI analyses performed 14 days later revealed that continuous IFNα therapy did not alter the growth of primary intracecal tumors (Figure 3D and E), while IFNα treatment significantly reduced both number and size of hepatic lesions (Figure 3D and F) with 60% of mice spared from metastatic lesions (Figure 3H). The primary intracecal tumors (Figure 3—figure supplement 2A) and liver metastases (Figure 3G) detected after orthotopic implantation of CT26LM3 cells were also characterized by immunohistochemistry (IHC). This analysis showed that primary intracecal tumors and liver metastatic lesions of NaCl-treated control mice were highly proliferative (as denoted by Ki67 positivity), exhibited marked signs of angiogenesis (as denoted by CD34 staining) and, accordingly with previous reports (Catarinella et al., 2016; Tauriello et al., 2018), were devoted of F4/80+ resident macrophages and CD3+ T cells (Figure 3G and H). Similar results were also observed in IFNα-treated primary intracecal tumors (Figure 3—figure supplement 2A, B). The absence of liver metastases in the majority of IFNα-treated mice is reflected by a reduced Ki67 or CD34 staining and an apparently normal distribution of F4/80+ macrophages and CD3+ T cells (Figure 3G and H). The few small hepatic lesions detected in 40% of mice continuously treated with IFNα (Figure 3H and Figure 3—figure supplement 2C, D) did not show differences in Ki67 positivity, CD34 staining or amount of F4/80+ resident macrophages and CD3+ T cells in relation to NaCl-treated mice (Figure 3—figure supplement 2C, D), consistent with the notion that CRC tumors may deregulate the
Altogether, these results indicate that continuous IFNα therapy does not significantly alter the growth of primary established CRC tumors but reduces the liver metastatic potential of invasive CRC cells emerging from the cecum.
HECs mediate the anti-metastatic activity of IFNα
As the
MRI analysis at day 21 after CRC challenge revealed that, in comparison with liver metastases observed in NaCl-treated controls, the lesions produced by MC38- or MC38
Figure 4.
HECs mediate the antimetastatic activity of IFNα.
(A) Representative hepatic contrast-enhanced MRI of wild-type mice (C57BL/6) at day 21 after injection of 5x104 MC38 cells or 5x104 MC38
Figure 4—figure supplement 1.
Characterization of conditional
(A) T7E1 mismatch detection assay was used to screen three representative single-cell MC38 clones (MC38B2 and MC38C8 transfected with crRNA1 or MC38H11 transfected with crRNA2). Additionally, untransfected MC38, the bulk crRNA1 MC38 line (defined as MC38Bulk) and a no DNA control were also treated or not with T7E1 and analyzed as further controls. A 100 bp Marker lane was used as reference. Note that only MC38 transfected with crRNA1 showed cleaved mismatched products and were further characterized. (B) Sanger sequencing was performed to validate the location and nature of indel events of the MC38C8 cell clone. The location of crRNA1 is depicted in red, insertions are highlighted in blue and dashed lines indicate deleted regions. Sequence alignment of MC38 and the two
Figure 4—figure supplement 2.
LSECs mediate the antimetastatic activity of IFNα.
(A) WBC counts at indicated time point from C57BL/6, injected with MC38 or MC38
Figure 4—figure supplement 3.
Characterization of liver metastases developed in
(A) Representative H&E, CD34, F4/80 and CD3 immunohistochemical micrographs of liver metastatic lesions found in NaCl- (red frame, n=4) and IFNα-treated (blue frame, n=7)
Next, we crossed
Continuous IFNα administration limits trans-sinusoidal migration of CRC cells by strengthening the liver vascular barrier
We next took advantage of fluorescence-based techniques to investigate the initial steps of liver colonization. First, we assessed the intrahepatic localization of GFP-expressing MC38 cells (MC38GFP) (Talamini et al., 2021) that were intramesenterically challenged 5 min earlier. Most MC38GFP cells in
Figure 5.
Continuous IFNα administration limits trans-sinusoidal migration of CRC cells by strengthening the hepatic vascular barrier.
(A) Total number of MC38GFP cells per area (top) and total number of intravascular MC38GFP cells per tissue area (bottom). The total hepatic area was approximately 5 mm2 for each experimental group. Intravascular localization was measured on 20 x images, 5 randomly selected images per mouse (n=3 per group). Mean values ± SEM are shown; p-values were calculated by Mann-Whitney test. (B) Confocal reconstruction of the liver vasculature from
Figure 5—figure supplement 1.
Continuous IFNα administration induces hepatic endothelial capillarization strengthening the liver vascular barrier.
(A) Representative composite immunofluorescence images showing a pan-endothelial cell marker (CD31, blue) and a liver sinusoidal marker (LYVE-1; red) 5 min after MC38GFP cells (green) injection in
Figure 5—figure supplement 2.
Continuous IFNα administration induces hepatic endothelial capillarization that is reversed after discontinuation of IFNα therapy.
(A) Representative composite immunofluorescent micrographs of mouse liver cryosections, from
Figure 5—video 1.
3D reconstruction of a representative extravasated MC38GFP cell found within the liver parenchyma of a NaCl-treated
Confocal reconstruction of the liver vasculature of a
Figure 5—video 2.
3D reconstruction of an intravascular MC38GFP cell found within a blood vessel of an IFNα-treated
Confocal reconstruction of the liver vasculature of a
Figure 5—video 3.
3D reconstruction of an extravasated MC38GFP cell found within the liver parenchyma of a NaCl-treated
Confocal reconstruction of the liver vasculature of a
Figure 5—video 4.
3D reconstruction of extravasated MC38GFP cells found within the liver parenchyma of an IFNα-treated
Confocal reconstruction of the liver vasculature of a
To unravel phenotypic modifications associated with such antitumor function of HECs, including LSECs, the liver microvasculature of NaCl- or IFNα-treated
Next, we evaluated the status of the vascular glycocalyx (GCX), a fibrous network of glycoproteins and proteoglycans that lines the LSECs and projects intraluminally (Reitsma et al., 2007). Notably, enhanced GCX deposits can act as a repulsive barrier that prevents tumor cell interactions with endothelial cells, adhesion molecules or chemokines have been previously identified as negative correlates of transendothelial migration (Glinskii et al., 2005; Mitchell and King, 2014; Offeddu et al., 2021; Wilkinson et al., 2020). Continuous IFNα treatment modified this network as well, increasing its thickness (Figure 5E and F top) and the expression of one of its major components, the heparan sulfate (HS) (Reitsma et al., 2007; Figure 5E and F bottom). Of note,
HECs acquire an antimetastatic transcriptional profile upon continuous IFNα sensing
To confirm the above-mentioned data and to shed new light on the transcriptional changes that HECs adopt to limit CRC trans-sinusoidal migration, we performed RNA-seq analyses on CD31+ endothelial cells isolated from the liver of
Figure 6.
HECs acquire an antimetastatic transcriptional profile upon continuous IFNα sensing.
(A) Volcano plots of differential gene expression (DGE) results obtained from the comparisons between HECs-derived from
Figure 6—figure supplement 1.
HECs acquire an antimetastatic transcriptional profile upon continuous IFNα sensing.
(A) Flow cytometry gating strategy for the sorting and isolation of CD31+ HECs isolated from the livers of
Keeping the HECs transcriptional profile of NaCl-treated
Continuous IFNα sensing improves immunostimulatory properties of HECs to provide long-term tumor protection
First, HECs/LSECs isolated from the liver of
Figure 7.
Continuous IFNα sensing improves immunostimulatory properties of HECs to provide long-term tumor protection.
(A–B) Quantification of the percentage of OT-I CD8+ T cells expressing CD44+IFNγ+ generated after the co-cultured with HECs, including LSECs, or sDCs isolated from
Figure 7—figure supplement 1.
Continuous IFNα sensing improves immunostimulatory properties of HECs to provide long-term tumor protection.
(A) FACS analysis showing the mean fluorescence intensity (MFI) of MHC-I (left) and the percentage of CD86 (middle) and IL-6RA cells (right) within the hepatic CD31+ non-parenchymal cells (NPC) population of
Figure 7—figure supplement 2.
Isolation and flow cytometry characterization of HECs and sDCs.
(A) Schematic diagram of the protocol used to isolate HECs, including LSECs, from murine liver tissue (left) and sDCs from mouse spleen (right). (B) Assessment of LSEC purity by flow cytometry using CD31 as a pan-endothelial marker, CD146 as a specific marker for LSECs and F4/80 to identify Kupffer cells contaminants. Analysis was performed in
To assess whether IFNα-stimulated HECs and LSECs promoted memory responses endowed with antitumor potential,
Discussion
In this study, we used different mouse models of CRC liver metastasis to show that the continuous perioperative administration of relatively low IFNα doses provides significant antitumor potential in vivo without provoking overt toxicity. Moreover, under the pharmacological conditions we defined (route, dosage, treatment duration, and chemical nature of the recombinant protein), we did not observe counter-regulatory mechanisms affecting IFNα efficacy (Katlinski et al., 2017), or significant systemic side effects, as our strategy avoids the short tissue-oscillatory IFNα bursts that are often achieved after high and pulsed administrations, often associated with efficacy-limiting toxicities (Weber et al., 2015). These results are consistent with previous preclinical work indicating that the intrahepatic delivery of IFNα through a gene/cell therapy approach curbs CRC liver metastases by acting primarily on unidentified non-hematopoietic stromal cell populations (Catarinella et al., 2016).
Given the pleotropic nature of IFNα, we demonstrated that the antimetastatic activity of IFNα is neither based on the direct inhibition of primary intracecal tumor growth, favoring the hypothesis that IFNα therapy does not modify the number of cells that spread from primary tumors and seed into the liver – nor on the direct inhibition of metastatic cell growth within the liver. These data are consistent with the high IFNα concentrations required to activate the ‘tunable’ direct antiproliferative functions of this cytokine, likely exceeding the levels achieved in our system (Catarinella et al., 2016; Schreiber, 2017). In addition, IFNα therapy does not require indirect stimulation of hepatocytes, HSCs, DCs, KCs, or LCMs to exert its antimetastatic functions. Rather, the results pinpointed HECs/LSECs as key local and early sensors of IFNα that ultimately limit CRC cell invasion into the liver.
Mechanistically, we showed that IFNα-stimulated LSECs inhibit the trans-sinusoidal migration of circulating CRC cells normally occurring within 24 hr of their initial intrahepatic landing. This effect is associated with phenotypic changes that IFNα-stimulated LSECs acquire or induce in the liver microenvironment. Among these changes, we observed a reduction in the overall LSEC porosity (i.e. sinusoidal fenestrae were reduced in number and size), an enhancement in the subendothelial deposition of basal membrane components (including Collagen IV and Laminin) and an upregulation of LYVE-1, a marker of hepatic capillarization (Pandey et al., 2020; Wohlfeil et al., 2019). Along these lines, it is noteworthy that in the ‘healthy’ liver, functioning as a common site for CRC metastases, LSECs contain numerous fenestrae of up to 200 nm in diameter and normally lack the typical basal membrane that characterizes the microvasculature of most other tissues and organs (Jacobs et al., 2010). It is also interesting to note that IFNα-stimulated LSECs promote microvascular alterations like those typifying pathological conditions (e.g. initial hepatic capillarization and liver fibrosis Pandey et al., 2020; Wohlfeil et al., 2019) associated with impaired immune cell extravasation and reduced immune surveillance (Guidotti et al., 2015) and reduction of hepatic metastases from solid tumors including CRCs (Wohlfeil et al., 2019). This fits with the evidence that CRC patients suffering from chronic viral liver fibrotic diseases characterized by hepatic endogenous type I interferon production display lower incidence of hepatic metastases (Augustin et al., 2013; Baiocchini et al., 2019; Li Destri et al., 2013). The existence that fibrotic liver diseases not associated with reduced metastatic risk (Kondo et al., 2016) suggests that changes in the vascular hepatic niche other than matrix deposition play additional roles in this process. Indeed, IFNα stimulated LSEC-governed changes hampering CRC extravasation including the modification of the sinusoidal GCX that, by increasing its thickness and modifying its chemical composition, recapitulated conditions known to negatively regulate the trans-endothelial migration of tumor cells in other settings (Glinskii et al., 2005; Mitchell and King, 2014; Offeddu et al., 2021; Wilkinson et al., 2020). The continuous administration of therapeutic low-doses of IFNα thus stimulate HECs/LSECs to shape a vascular antimetastatic barrier preventing the interaction between tumor cells and endothelial cells that are known to promote the extravasation of the former cells (Glinskii et al., 2005; Mitchell and King, 2014; Wilkinson et al., 2020). Accordingly, the enhanced expression of ‘pro-migratory’ adhesion molecules and integrins that we observed in the liver of animals bearing IFNα-responsive LSECs appear to be efficiently counteracted by the creation of such vascular barrier.
Of note, the functional consequences of LSEC capillarization (especially the induction of hepatic fibrosis) during states of chronic liver injury highly depend on the relative magnitude and duration of the underlying liver disease (DeLeve, 2015). Additionally, both LSEC capillarization and hepatic fibrosis are reversed when chronic liver injury resolves (DeLeve, 2015). In keeping with these concepts and the absence of sALT elevation or morphological evidence of liver disease during continuous IFNα therapy, it is not surprising that we observed a complete recovery of fenestrae abundance and LSEC porosity 40 days after therapy discontinuation. This supports the notion that a continuous but relatively short IFNα therapy promotes changes in the structure and function of LSECs that are mild and reversible and should not result in persistent hepatic fibrinogenesis. Such notion is also supported by the absence of hepatic toxicity (Weber et al., 2015) and the significant reduction in established fibrosis in patients with chronic viral liver diseases treated with recombinant IFNα for up to 48 weeks (Li et al., 2019; Poynard et al., 2002). The notion that IFNα treatment failed to shape the vascular antimetastatic barrier in mice carrying the
In addition, to hindering the initial trans-sinusoidal migration of CRC cells in vivo, IFNα-stimulated LSECs efficiently cross-presented nominal tumor antigens to naive CD8+ T cells in vitro, enabling degrees of T cell priming and effector differentiation that were comparable to those induced by professional APCs. In keeping with this, we demonstrated that the in vivo IFNα stimulation of LSECs resulted in the upregulation of proteins and transcripts associated with antigen processing and presentation or co-stimulation (e.g. MHC-I, CD86, IL-6RA,
These data are consistent with the notion that IFNα-stimulated LSECs, due to their anatomical proximity and efficient endocytosis capacity that is among the highest of all cell types in the body (Sorensen and Smedsrod, 2020) – rapidly remove CRC-derived antigens from the intravascular space and productively and rapidly contribute to the development of effective antitumor immunity, since this process does not require the time-consuming step of migration to lymphatic tissue (Böttcher et al., 2014). This concept is also supported by the upregulation by IFNα-stimulated LSECs of
Altogether, we have identified a novel MoA by which IFNα functions as antitumor drug against CRC liver metastases. Whether the adoption of similar LSEC-stimulating IFNα treatments may also curb the hepatic growth of metastatic cells originating from other solid tumors, or if continuous IFNα treatment promote the generation of vascular barriers in other metastasis-prone organs remains to be determined (Crist and Ghajar, 2021). Based on the findings of this report, we propose the following model: CRC cells emerging from the primary tumor reach the hepatic sinusoids via the portal circulation and arrest – mostly because of size constrains – at the portal side of the sinusoidal circulation. CRC cells then trans-sinusoidally migrate into the liver parenchyma and develop micrometastases that will eventually grow overtime, promoting the generation of an immunosuppressive microenvironment. Continuous therapy with well-tolerated doses of recombinant IFNα, stimulates HECs/LSECs to limit CRC trans-sinusoidal migration and parenchymal invasion by building up a vascular barrier typified by the reduction of LSECs porosity, the increased thickness of GCX and the appearance of a basal membrane. Continuous IFNα therapy also promotes long-term antitumor immunity in cured mice and protection from secondary tumor challenge, by stimulating LSECs to efficiently cross-prime tumor antigens to naïve CD8+ T cells (Figure 7F).
In terms of future clinical applications, our strategy could be used as perioperative neoadjuvant immunotherapy in CRC patients undergoing resection of their primary tumor who are at high risk for developing metachronous liver metastases (Engstrand et al., 2019; van Gestel et al., 2014). Indeed, several technologies have already been developed for the sustained release of drugs, such as osmotic pumps, electronic devices, hyaluronic acid-based hydrogels (Park et al., 2018; Stewart et al., 2018; Yun and Huang, 2016), FDA-approved polymer miscellas – such as pegylated (PEG)-IFNα (Foser et al., 2003; Glue et al., 2000) – and IFNα cell/gene therapy approaches (Catarinella et al., 2016), which could quickly translate our results into clinical practice. Of note, the use of clinically approved doses of pegylated-IFNα has shown improved serum stability and clinical efficacy and reduced side effects, with serum IFNα concentrations similar to those achieved in our system (Foser et al., 2003; Glue et al., 2000).
All in all, the results of this study support the use of continuous low doses of IFNα as an antimetastatic drug during the perioperative period, due to its ability to transform a metastases-prone liver into a metastases-resistant organ.
Materials and methods
Animal studies
Eight- to 10-week-old C57BL/6 J and BALB/c mice were purchased from Charles River Laboratory, Calco, Italy. CB6 mice were obtained by crossing
The conditional deletion of
Study approval
All animal experiments were approved by the Animal Care and Use Committee of the San Raffaele Scientific Institute (
CRC cell lines
CT26 (H-2d, BALB/c-derived) cell line was purchased from ATCC. MC38 (H-2b, C57BL/6-derived), have been previously described (Catarinella et al., 2016). MC38 cells were transduced with a PGK-GFP lentiviral vector, cloned and sorted by FACS to establish MC38GFP fluorescently tagged cell lines (Talamini et al., 2021). All cells were routinely tested for mycoplasma contamination using the N-GARDE Mycoplasma PCR reagent set (EuroClone). CT26 and CT26LM3 cells were cultured under standard condition at 37 °C in a humid atmosphere with 5% CO2 in RPMI GlutaMAX medium (Gibco) supplemented with 10% FBS (Lonza) and 1% penicillin/streptomycin (P/S) (Gibco). MC38, MC38GFP, MC38
Mouse models of liver metastases
Eight-to ten-week-old sex- and age-matched mice were injected with 5x103 CT26 or 5x104 MC38 CRC cell lines either through intrasplenic or superior mesenteric vein injections as previously described (Catarinella et al., 2016; van der Bij et al., 2010). For early time point experiments, 7x105 MC38GFP cells were injected in the superior mesenteric vein of anesthetized mice as described (van der Bij et al., 2010). For intrasplenic or superior mesenteric vein injections, deep anesthesia was induced by isoflurane inhalation (5% induction and 2% for maintenance in 2 l/min oxygen). The indicated number of CRC cells was injected into spleen or the superior mesenteric vein using a 29 G needle and to prevent excessive bleeding vein puncture was compressed with a sterile and absorbable hemostatic gauze (TABOTAMP). The peritoneum and skin were sutured with silk 4.0 and 7 mm wound clips as described (Catarinella et al., 2016). This experimental setting may mimic the vascular spreading of CRC cells during primary tumor resection and, thus, preventive IFNα infusion may be considered as a neoadjuvant treatment.
Mouse model of orthotopic colorectal cancer liver metastases
The generation of highly metastatic CT26 CRC cells was obtained by three consecutive rounds of in vivo selection as previously reported (Zhang et al., 2013). Briefly, 2x106 CT26 cells were first injected subcutaneously into the right flank of immunodeficient NSG mice. After 28 days, tumors were excised, dissected, sliced into small fragments, and digested for 30 min at 37 °C in DMEM containing collagenase type IV (200 units/ml; Sigma-Aldrich) and Dnase I (100 units/ml; Sigma-Aldrich). The resulting cell suspension defined as CT26sc, were maintained at 4 °C, filtered through a 70 μm nylon cell strainer (BD Biosciences, Bedford, MA), washed in PBS, and grown in RPMI 10% FBS. Sub-confluent CT26sc cells were harvested, resuspended in PBS:Matrigel (1:1) (Corning, MERK) and then injected into the cecal wall of immune competent anesthetized (isoflurane, 5% induction and 2% for maintenance in 2 l/min oxygen), CB6 recipient mice as described (Zhang et al., 2013). Briefly, a midline incision was made to exteriorize the cecum. Using a 33 G micro-injector (Hamilton, USA), 10 µl of a 50% Matrigel solution (BD Bioscience, USA) containing 2x105 CT26sc cells were injected into the cecum wall. To avoid intraperitoneal spreading of CT26sc, the injection site was sealed with tissue adhesive (3 M Vetbond, USA) and washed with 70% alcohol. The cecum was replaced in the peritoneal cavity, and the abdominal wall and skin incision was sutured with silk 4.0 and 7 mm wound clips as described (Catarinella et al., 2016). Twenty-eight days after injection, mice were euthanized and single cell suspensions of liver metastatic lesions, defined as CT26LM1, were obtained as described above. This cycle was repeated twice to obtain the highly metastatic CT26LM3 cells.
Recombinant mouse IFNα therapy
Continuous intraperitoneal IFNα delivery (IFNα1 carrier-free, Biolegend, San Diego, CA, USA) was achieved by intraperitoneal implantation of mini-osmotic pumps (MOP, ALZET, Cupertino, CA, USA) able to deliver either 50, 150, or 1050 ng IFNα a day for 14 or 28 days. NaCl-containing MOP were used as controls. Within each specific experiment, mice of each genotype were randomly assigned to receive either NaCl- or IFNα-containing MOP. MOP filling, priming and implantation within the peritoneum was performed following manufacturer’s instructions. To avoid MRI artifacts due to the presence of metallic components within MOP, the day before MRI acquisition, MOP were removed from the peritoneum. To directly investigate responsiveness of liver cells to IFNα, signaling downstream of
Tumor rechallenge of IFNα cured mice
IFNα-cured mice that were designated as MC38-tumor free for at least 50 days after challenge, were subcutaneously rechallenged with 5x103 MC38 cells resuspended in 200 µl of PBS:Matrigel (1:1). Age-matched naïve syngeneic mice were used as control. Tumor volumes were measured twice a week and euthanized for ethical reasons when tumor size reached ~500 mm3.
Magnetic resonance imaging (MRI)
All MRI studies were carried out at the Experimental Imaging Center of SRSI on a preclinical 7-Tesla MR scanner (Bruker, BioSpec 70/30 USR, Paravision 6.0.1, Germany) equipped with 450/675 mT/m gradients (slew rate: 3400/4500 T/m/s; rise time 140 µs), coupled with a dedicated 4 channels volumetric mouse body coil. All images were acquired in vivo, under inhalational anesthesia (Isoflurane, 3% for induction and 2% for maintenance in 1 L/min oxygen) with mice laid prone on the imaging table. A dedicated temperature control system was used to prevent hypothermia; respiratory rate and body temperature were continuously monitored (SA Instruments, Inc, Stony Brook, NY, USA) during the whole MRI scan. An intravenous injection of gadoxetic acid (Gd-EOB-DTPA; Primovist, Bayer Schering Pharma) at a dose of 0.05 μmol/g of body weight was administered via the tail vein before placing the mice on the scanner table. As previously described (Sitia et al., 2012), the MRI studies relied on an axial fat-saturated T2-weighted sequence (TurboRARE-T2: TR = 3394ms, TE = 33ms, voxel-size=0.125 × 0.09 x0.8mm, averages = 3) acquired immediately after Gd-EOB-DTPA injection and an axial fat-saturated T1-weighted scan (RARE-T1: TR = 581ms, TE = 8.6ms, voxel-size=0.125 × 0.07 x0.8mm, averages = 4) acquired thereafter, during the hepatobiliary phase (HBP) of contrast excretion (starting from 10 min after Gd-EOB-DTPA injection). Two board certified radiologists skilled in clinical and preclinical abdominal MR imaging, blinded to any other information, reviewed all MRI studies using an open-source image visualization and quantification software (Mipav, 5.3.4 and later versions, Biomedical Imaging Research Services Section, ISL, CIT, National Institute of Health, USA). Liver metastases were identified as focal lesions showing slight hyper-intensity on T2-weighted images and concurrent hypo-intensity on contrast-enhanced HBP T1-weighted images. Liver metastases segmentation was performed by manual drawing of regions-of-interest (ROIs) on each slice, yielding volumes-of-interest (VOIs; lesion area x slice thickness) for the entire sequence. The total CRC metastatic mass was obtained by summing up the volumes of all single VOIs that were semi-automatically provided by the software.
MC38 gene editing
To knockout
Peripheral blood analyses
At the indicated time points after IFNα administration, whole anti-coagulated blood of MOP-NaCl and MOP-IFNα-treated mice was collected from the retro-orbital plexus of anesthetized animals (isoflurane, 5% for induction and 2% for maintenance in 2 l/min oxygen) using Na-heparin coated capillaries (Hirschmann Laborgeräte GmbH, Germany) and vials (Microvette, Sarstedt, Germany). Hematologic parameters were evaluated using an automated cell counter (ProCyte Dx, IDEXX Laboratories, USA). The extent of hepatocellular injury was monitored by measuring serum ALT (sALT) activity at several time points after IFNα treatment, as previously described (Sitia et al., 2012).
Measurement of plasma IFNα by ELISA
Circulating levels of IFNα were quantified in plasma collected from NaCl controls or IFNα-treated mice at indicated time points using VeriKine-HS Mouse IFNα all Subtype ELISA Kit (PBL) according to the manufacturer’s instructions. The IFNα titer in the samples was determine by plotting the optical density (OD) subtracted of blank OD to eliminate background, using a 4-parameter logistic fit for the standard curve by using Prism v8 (GraphPad). Detection range is comprised between 2.38 and 152 pg/ml of IFNα.
RNA extraction and quantitative real-time PCR gene expression analyses
Total RNA was isolated from liver homogenates of MOP-NaCl control and MOP-IFNα-treated mice by using the ReliaPrep RNA Tissue Miniprep System (Promega) and DNAse TURBO (Thermo Fisher Scientific) following manufacturer’s recommendation. The extracted RNA was subsequently retro-transcribed to cDNA as previously described (Sitia et al., 2011). Quantitative real-time PCR analysis was performed utilizing the ViiA7 Fast Real-Time PCR System (Applied Biosystems). The ISG,
Immunohistochemistry
At time of autopsy for each mouse, livers were perfused with PBS, harvested and different pieces were sampled, fixed in zinc-formalin, processed and embedded in paraffin for histological and immunohistochemical analysis, as previously described (Sitia et al., 2012). Immunohistochemical staining using a Bond RX Automated Immunohistochemistry (Leica Microsystems GmbH, Wetzlar, Germany) was performed on 3-μm-thick sections. First, tissues were deparaffinized and pre-treated with the Epitope Retrieval Solution [ER1 Citrate Buffer for Ki-67 (dilution 1:200, clone SP6, Thermo Fisher Scientific) and F4/80 (dilution 1:200, clone A3-1, Bio-Rad); ER2 EDTA for CD3 (dilution 1:100, clone SP7, Abcam),CD34 (dilution 1:300, clone MEC14.7, Biolegend) and pSTAT1 (dilution 1:800, clone 58D6, Cell Signaling)] at 100 °C for 30 min. After washing steps, peroxidase blocking was carried out for 10 min using the Bond Polymer Refine Detection Kit DS9800 (Leica Microsystems GmbH). Then, tissues were washed and incubated for 1 hr RT with the primary antibody diluted in IHC Antibody Diluent (Novocastra, Leica RE7133). Subsequently, tissues were incubated with polymer-HRP or Rat-on-Mouse HRP (Biocare Medical, RT517H), developed with DAB-Chromogen for 10 min and counterstained with Hematoxilin for 5 min. For image acquisition and analysis, eSlide Manager (Aperio Leica Biosystems) was used. All images were acquired using the Aperio AT2 system (Leica Biosystems). Quantifications were performed by automated image analysis software through dedicated macros of the ImageScope program, customized following manufacturer’s instructions (Leica Biosystems). The images shown were identified as representative area of interest within the total area of the specimen analyzed and exported as ImageScope snapshots.
Immunofluorescence and confocal microscopy
Livers were perfused with PBS, harvested and fixed over-night in 4% paraformaldehyde (PFA), equilibrated in 30% sucrose in PBS over-night at 4 °C prior to embedding in OCT (Bio-Optica) for quick freezing at –80 °C. Thirty-μm-thick cryosections were adhered to Superfrost Plus slides (Thermo Scientific). For immunofluorescence staining, sections were blocked and permeabilized with PBS containing 5% FBS and 0.1% Triton X-100 (Sigma-Aldrich) for 30 min at room temperature and subsequently incubated with 10% of donkey serum (DS; Sigma-Aldrich) in PBS for 30–60 min at room temperature. Staining with primary and secondary antibodies, were performed with staining buffer (PBS with 1.5% DS, 0.2% Triton X-100 and 1% BSA), using the following antibodies and dilutions: anti-PDGFRβ/CD140b (dilution 1:200, clone APB5, eBioscience)+anti rat AF647 (dilution 1:200, Jackson IR); anti-CD11c AF647, (dilution 1:100, clone N418, Biolegend); anti-GFP (dilution 1:100, rabbit polyclonal, A11122 Thermo Fisher Scientific)+anti-rabbit AF 488 (dilution 1:200, Thermo Fisher Scientific); anti-CD31/PECAM-1 (dilution 1:300, goat polyclonal, AF3628 R&D Systems)+anti goat AF546 (dilution 1:200, Thermo Fisher Scientific); anti-Heparan Sulfate (dilution 1:50, clone F58-10E4, Amsbio)+anti IgM conjugated to APC (dilution 100, clone II/41, Thermo Fisher Scientific); anti-LYVE-1 (dilution 1:300, rabbit polyclonal, Novus Biologicals)+anti-rabbit AF647 (dilution 1:200, Thermo Fisher Scientific); anti-Collagen type IV (dilution 1:100, rabbit polyclonal, Abcam)+anti-rabbit AF488 (dilution 1:200, Thermo Fisher Scientific); anti-Laminin (dilution 1:300, rabbit polyclonal, Sigma-Aldrich)+anti-rabbit AF488 (dilution 1:200, Thermo Fisher Scientific); anti-CD54/ICAM1 (dilution 1:100, clone YN1/1.7.4, Biolegend)+anti rat AF647 (dilution 1:200, Jackson IR); anti-CD62E/E-selectin (dilution 1:100, clone 10E9.6, BD Bioscience)+anti rat AF647 (dilution 1:200, Jackson IR). Confocal images were acquired using a Leica SP8 confocal systems (Leica Microsystems) that are available at the SRSI Advanced Light and Electron Microscopy BioImaging Center (ALEMBIC). Fifteen–20 μm z-stacks were projected in 2D and processed using Fiji image processing software (Schindelin et al., 2012). Localization of MC38GFP tumor cells within liver vessels, 20–30 square xy sections (1024x1,024 pixel) confocal xyz stacks, from NaCl- and IFNα-treated mice, were acquired with 0.5 µm z-spacing on a Leica TCS SP8. The Imaris Surpass View and Surface Creation Wizard were used to create 3D renderings of MC38GFP cells and CD31+ liver vessels as previously reported (Guidotti et al., 2015). A tumor cell was considered intravascular when at least 95% of its surface-reconstructed body was inside the vessel lumen in all the acquired sections projected in the horizontal (xy), transversal (yz) and longitudinal (xz) planes. Entire liver sections were acquired using MAVIG RS-G4 confocal microscope (MAVIG GmbH Research, Germany) to quantify the number of MC38GFP tumor cells in relation to the total liver area. The quantification of the percentage of liver area covered by endothelial markers and extracellular matrix components, such as Heparan Sulfate, Laminin, Collagen type IV, ICAM1 and E-Selectin, was evaluated using ImageJ software, applying the same threshold to the different experimental groups for each channel and measuring the percentage of area limited to threshold. Colocalization analysis of GFP in the different Cre recombinant mouse models was performed using an unsupervised ImageJ plugin algorithm termed Colocalization, which was developed by Pierre Bourdoncle (Institut Jacques Monod, Service Imagerie, Paris; 2003–2004).
Scanning and transmission electron microscopy
Electron microscopy (EM) fixative composition was 2,5% glutaraldehyde, 2% paraformaldehyde, 2 mM CaCl2 and 2% sucrose in 0.1 M Na cacodylate buffer (pH 7,4). For the analysis of endothelial GCX, the EM fixative was supplemented with 2% Lanthanum nitrate (MERK) as previously reported (Inagawa et al., 2018). Warm PBS and EM fixative at 35–37°C was used to ensure tissue integrity. When mice were under deep anesthesia, with a single ip injection of 50–60 mg/kg Tribromoethanol (Avertin), a Y incision was made in the abdomen to expose the liver and the portal vein. The portal vein was cannulated with an appropriately sized IV cannula of 22 G and the liver was perfused with 15 ml of warm PBS at a constant rate of 3 rpm using a peristaltic pump (Peri-Star Pro, 2Biological Instruments) as previously reported (Guidotti et al., 2015). In situ fixation was achieved by perfusing EM fixative for approximately 5 min at 3 rpm. Fixed liver was harvested and cut into 5 mm blocks using a scalpel. Liver blocks were ulteriorly immersed in EM fixative for 24–72 hr at 4 °C and finally EM fixative was replaced with 0.1 M sodium cacodylate buffer and stored at 4 °C until processed for TEM or SEM analysis. Liver blocks were post-fixed in 1% osmium tetroxide (OsO4), 1,5% potassium ferricyanide(K4[Fe(CN)6]) in 0.1 M Na Cacodylate buffer for 1 hr on ice. Afterwards, for SEM, 150-µm-thick sections were obtained from perfused livers using a vibratome (Leica VT1000S). Sections were further post-fixed in 1% OsO4 in 0.1 M Na Cacodylate, dehydrated through a series of increasing concentration of ethanol and immersed in absolute hexamethyldisilazane (HMDS) that was left to evaporate overnight. Dried sections were sputter-coated with Chromium using a Quorum Q150T ES sputter coater. Sections were then mounted on SEM stubs using conductive adhesive tape and observed in a field-emission scanning electron microscope Gemini 500 (ZEISS, Oberkochen, Germany). The LSEC fenestra measurements were performed from SEM microphotographs taken under a magnification of 20,000 X, using three independent samples from each experimental condition [NaCl-
For TEM, tissue pieces were rinsed in Na Cacodylate buffer, washed with distilled water (dH2O) and en bloc stained with 0.5% uranyl acetate in dH2O overnight at 4 °C in the dark. Finally, samples were rinsed in dH2O, dehydrated with increasing concentrations of ethanol, embedded in Epon resin and cured in an oven at 60 °C for 48 hr. Ultrathin sections (70–90 nm) were obtained using an ultramicrotome (UC7, Leica microsystem, Vienna, Austria), collected, stained with uranyl acetate and Sato’s lead solutions, and observed in a Transmission Electron Microscope Talos L120C (FEI, Thermo Fisher Scientific) operating at 120kV. Images were acquired with a Ceta CCD camera (FEI, Thermo Fisher Scientific). TEM microphotographs were taken under a magnification of 3.400 X, using three independent samples from each experimental condition and a total area of approximately 2.575 µm2 was analyzed for each mouse. LSECs thickness and the width of the space of Disse were measured using ImageJ software. For collagen deposition analysis, at least 10 randomly selected sinusoids from each mouse were analyzed as previously reported (Gissen and Arias, 2015; Warren et al., 2007).
Isolation of liver non-parenchymal cells (NPCs)
Liver NPCs, including leukocytes, were isolated from NaCl control or IFNα-treated mice 7 days after MOP implantation, as previously described (Bénéchet et al., 2019). Briefly, after euthanasia, the liver was perfused through the vena cava with 5–8 ml of PBS to remove most blood cells. Livers were weighted and 50% of the tissue was sliced in small pieces and incubated 30 min at 37 °C in 10 ml of digestion medium (RPMI GlutaMAX medium [Gibco] containing 200 U/ml of collagenase type IV [Sigma-Aldrich] and 100 U/ml of DNAse I [Sigma-Aldrich]). Remaining undigested fragments were syringed with an 18 G needle and filtered through a 70 µm cell strainer to obtain a single cell suspension. Cells were centrifuged 3 min at 50 g at room temperature and the pellet containing hepatocytes was discarded. The resultant cell suspension of NPCs was incubated for 30 s with ACK lysis buffer (Lonza) to deplete red blood cells, washed with cold RPMI. NPCs were counted and processed for flow cytometry analysis or sorting.
Isolation of splenocytes and naïve CD8+ T cells
Spleens were obtained from
Flow cytometry and cell sorting
Cells were resuspended in PBS and LIVE/DEAD Fixable Near-IR or Green dead cell dyes (Thermo Fisher Scientific) and incubated 15 min at RT in the dark for cell viability assessment. Subsequently, cells were blocked with FACS buffer (PBS supplemented with 2% FBS) containing InVivo Mab anti-mouse CD16/CD32 (BioXCell) and stained for surface markers using the following antibodies: anti-CD45R/B220 (clone RA3-6B2, Biolegend), anti-CD11b (clone M1/70, Biolegend), anti-mouse CD11c (clone N418, Biolegend), anti-CD3 (clone 145–2 C11, Thermo Fisher Scientific), anti-CD45 (clone 30-F11, Bioegend), anti-CD8a (clone 53–6.7, Biolegend), anti-F4/80 (clone BM8, Biolegend), anti-CD126/IL-6RA (clone D7715A7, Biolegend), anti-CD18/ITGB2 (clone M18/2, Biolegend), anti-CD49d/ITGA4 (clone 9C10(MFR4.B), Biolegend), anti-H-2Kb/H-2Db (clone 28-8-6, Biolegend), anti-CD86 (clone GL-1, Biolegend), anti-CD146 (clone ME-9F1, Biolegend), anti-CD44 (clone IM7, Biolegend), anti-CD62L (clone MEL-14, Biolegend), anti-IFNγ (clone XMG1.2, BD Biosciences), anti-CD31 (clone MEC13.3, BD Biosciences)for 20 min at 4 °C. For intracellular IFNγ staining, cells were then fixed, permeabilized and stained following Foxp3/Transcription Factor Staining buffer set (Thermo Fisher Scientific) manufacturer’s guidelines. When preparing samples for FACS sorting, NPCs were directly blocked and stained with CD45-APC and CD31-BV421. Viability was evaluated by 7-AAD (Biolegend) staining that was added to samples 5 min before sorting. Cell sorting was performed on a BD FACSAria Fusion (BD Biosciences) equipped with four lasers: Blue (488 nm), Yellow/Green (561 nm), Red (640 nm) and Violet (405 nm). 85 µm nozzle was used and sheath fluid pressure was at 45 psi. A highly pure sorting modality (four-way purity sorting) was chosen. The drop delay was determined using BD FACS AccuDrop beads. Unstained and a single-stained controls have been used to set up compensation. Rainbow beads (SPHEROTM Rainbow Calibration Particles) were used to standardize the experiment and were run before each acquisition. Samples were sorted at 4 °C to slow down metabolic activities. Sorted cells were collected in 1.5 ml Eppendorf tubes containing 200 µl of DMEM 10% FBS medium and immediately processed for RNA extraction using ReliaPrep RNA Cell Miniprep System (Promega) and DNAse TURBO (Thermo Fisher Scientific) following manufacturer’s recommendation.
Isolation of HECs, including LSECs and splenic DCs for in vitro studies
Mouse liver perfusion was performed as described in the section Electron Microscopy. After PBS perfusion, liver digestion was achieved in situ by perfusing warm digestion medium at 4 rpm for 10 min. The cava vein was squeezed tight several times to build up some pressure within the liver in order to fill all liver lobes with digestion medium. The resultant digested liver was excised, placed on a petri dish containing digestion medium and the Glisson’s capsule was removed. Disaggregated tissue was filtered using a 70-µm cell strainer and centrifuged at 50 g for 3 min to discard hepatocytes. The supernatant containing NPCs was counted and Kupffer cells (KC) were removed using anti-F4/80 Ultrapure microbeads (Miltenyi Biotec) following manufacturer’s guidelines. The flow-through of unlabeled NPCs was placed on top of a 25–50% Percoll gradient and centrifuged at 850 g for 20 min at RT without brake and the LSECs located at the 25/50% interface were collected. LSECs were counted and 105 cells were seeded in a 48-well plate and cultured in collagenated plates with EGM-2 medium (Lonza) for 3 days. For the isolation of splenic DC, spleens were slowly injected with 1 ml of digestion medium until they changed from dark maroon to reddish-orange color. Then, the spleen was minced and pipetted vigorously several times in digestion medium. The cell suspension was filtered using a 70 µm cell strainer, and larger undigested fragments were ulteriorly incubated with digestion medium at 37 °C for 30 min. After tissue digestion, splenocytes were centrifuged at 500 g for 5 min and washed three times with plain RPMI supplemented with 5 mM EDTA to disrupt DC-T cell complexes. Red blood cells were lysed with ACK lysis buffer and DCs cells were isolated using anti-CD11c microbeads UltraPure (Miltenyi) according to manufacturer’s instructions. CD11c+ DCs were counted and seeded at 105 cells in a 48-well plate using RPMI GlutaMAX supplemented with 10% FBS, 1% P/S, 1 x Na Pyruvate (Gibco), 1 x MEM Non-essential Amino Acid Solution (Gibco), 20 µM β-mercaptoethanol and 40 ng/ml GM-CSF (BioXcell). Every 48 hr 200 µl of fresh medium were added to cultured cells and DCs were grown for 7 days to stimulate DCs differentiation.
Antigen cross-priming
Prior to naïve CD8+ T cell co-culture, HECs, including LSECs, and sDCs were stimulated for 18 hr with 1 µg/ml of SIINFEKL peptide (OVA 257–264, Proimmune) or with 1 mg/ml of low endotoxin soluble ovalbumin protein (sOVA; Sigma-Aldrich) in combination with NaCl or 5 ng/ml of IFNα. Subsequently, after extensive washes, cells were co-culture with 5x105 naive CD8+T cells isolated from OT-I mice in complete RPMI (containing 10% FBS and 50 µM β-mercaptoethanol) in combination with NaCl or 5 ng/ml of IFNα. After 3 days, CD8+ T cells were stimulated for 4 hr with 1 µg/ml of SIINFEKL peptide, 5 µg/ml of Brefeldin A (Sigma-Aldrich) and 2.5% EL-4 supernatant in complete RPMI. Finally, the production of IFNγ and the expression of activation markers, such as CD44, were measured by FACS.
RNA-seq and bioinformatic data analysis
RNA integrity, isolated from sorted liver CD31+ cells, was evaluated using the Agilent 4100 TapeStation (Agilent Technologies) and samples with RNA integrity number (RIN) above 7.0 were used for subsequent RNA-Seq-based profiling. Libraries were prepared using the Illumina Stranded mRNA ligation kit, according to the manufacturer’s instructions. The RNA-Seq library was generated following the standard Illumina RNA-Seq protocol and sequenced on an Illumina NovaSeq 6000 machine (Illumina, San Diego, CA) obtaining an average of 40 millions of single-end reads per sample. The raw reads produced from sequencing were trimmed using Trimmomatic, version 0.32, to remove adapters and to exclude low-quality reads from the analysis. The remaining reads were then aligned to the murine genome GRCm38 using STAR, version 2.5.3 a. Reads were eventually assigned to the corresponding genomic features using featureCounts, according to the Gencode basic annotations (Gencode version M22). Quality of sequencing and alignment was assessed using FastQC, RseQC, and MultiQC tools. Expressed genes were defined as those genes showing at least 1 Count Per Million reads (CPM) on at least a selected number of samples, depending on the size of the compared groups (Chen et al., 2016). Low-expressed genes that did not match these criteria were excluded from the corresponding dataset. Gene expression read counts were exported and analyzed in R environment (v. 4.0.3) to identify differentially expressed genes (DEGs), using the DESeq2 Bioconductor library (v. 1.30.1, Love et al., 2014). P-values were adjusted using a threshold for false discovery rate (FDR)<0.05 (Benjamini and Hochberg, 1995). Significant genes were identified as those genes showing FDR <0.05. Functional enrichment analysis was conducted using the enrichR R package (v. 3.0, Kuleshov et al., 2016), starting from the lists of differentially expressed genes as defined by FDR <0.05. Selected pathways were grouped and summarized according to their general biological functions and the hypergeometric test was performed to test the enrichment of these custom genesets, exploiting the hypeR R package (v. 1.8.0). Pre-ranked Gene Set Enrichment Analysis (GSEA Subramanian et al., 2005) was performed for each DGE comparison, on all the expressed genes ranked for Log2 fold change. The gene-sets included in the GSEA analyses were obtained from Canonical Pathways, Hallmark and the Gene Ontology collections as they are reported in the MsigDB database.
Statistical analysis
In all experiments, values are expressed as mean values ± SEM. Statistical significance was estimated by two-tailed non-parametric Mann-Whitney test (e.g. to evaluate differences generated as a consequence of tumor growth) or by one-way ANOVA with Tukey’s multiple comparison test when more than two groups were analyzed. Contingency tables were tested by two-tailed Fisher’s exact test and Chi-square test. Statistical significance of survival experiments was calculated by log-rank/Mantel-Cox test. All statistical analyses were performed with Prism 8 (GraphPad Software) and were reported in Figure legends. p-values <0.05 were considered statistically significant and reported on graphs. If not mentioned, differences were not statistically significant.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2022, Tran, Ferreira, Alvarez-Moya et al. This work is published under https://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Hepatic metastases are a poor prognostic factor of colorectal carcinoma (CRC) and new strategies to reduce the risk of liver CRC colonization are highly needed. Herein, we used mouse models of hepatic metastatization to demonstrate that the continuous infusion of therapeutic doses of interferon-alpha (IFNα) controls CRC invasion by acting on hepatic endothelial cells (HECs). Mechanistically, IFNα promoted the development of a vascular antimetastatic niche characterized by liver sinusoidal endothelial cells (LSECs) defenestration extracellular matrix and glycocalyx deposition, thus strengthening the liver vascular barrier impairing CRC trans-sinusoidal migration, without requiring a direct action on tumor cells, hepatic stellate cells, hepatocytes, or liver dendritic cells (DCs), Kupffer cells (KCs) and liver capsular macrophages (LCMs). Moreover, IFNα endowed LSECs with efficient cross-priming potential that, along with the early intravascular tumor burden reduction, supported the generation of antitumor CD8+ T cells and ultimately led to the establishment of a protective long-term memory T cell response. These findings provide a rationale for the use of continuous IFNα therapy in perioperative settings to reduce CRC metastatic spreading to the liver.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer