Dynamic control of lymphocyte trafficking by fever-range thermal stress

Abstract

Migration of blood-borne lymphocytes into tissues involves a tightly orchestrated sequence of adhesion events. Adhesion molecules and chemokine receptors on the surface of circulating lymphocytes initiate contact with specialized endothelial cells under hemodynamic shear prior to extravasation across the vascular barrier into tissues. Lymphocyte-endothelial adhesion occurs preferentially in high endothelial venules (HEV) of peripheral lymphoid organs. The continuous recirculation of naïve and central memory lymphocytes across lymph node and Peyer's patch HEV underlies immune surveillance and immune homeostasis. Lymphocyte-endothelial interactions are markedly enhanced in HEV-like vessels of extralymphoid organs during physiological responses associated with acute and chronic inflammation. Similar adhesive mechanisms must be invoked for efficient trafficking of immune effector cells to tumor sites in order for the immune system to have an impact on tumor progression. Here we discuss recent evidence for the role of fever-range thermal stress in promoting lymphocyte-endothelial adhesion and trafficking across HEV in peripheral lymphoid organs. Findings are also presented that support the hypothesis that lymphocyte-endothelial interactions are limited within tumor microenvironments. Further understanding of the molecular mechanisms that dynamically promote lymphocyte trafficking in HEV may provide the basis for novel approaches to improve recruitment of immune effector cells to tumor sites.

Fig. 1

Analysis of lymphocyte-endothelial interactions in nodal venules by intravital microscopy. A Superficial epigastric artery (SEA), superficial epigastric vein (SEV), and nodal venular structure were observed under low power (10X; left photomicrograph) in surface inguinal LN of C57BL/6 mice by epifluorescence intravital microscopy as described previously [11, 91]. Interactions between lymphocytes and nodal venules of different orders were visualized under high power (20×) in the same field following injection of fluorescent-labeled LN cells (~2.5×10⁷ cells/mouse; labeled with calcein [1 μg/ml, Molecular Probes, Eugene, OR, USA]) via the femoral artery (right photomicrograph). The majority of fluorescent-labeled, firm sticking cells accumulate in order III–V vessels. B Rolling or sticking lymphocytes in different order venules were quantified in two mice. Rolling fraction was defined as the percentage of cells transiently interacting with HEV in the total number of cells passing through the vessel during the observation period, as described by von Andrian and M’Rini [91]. The median velocity of 30 non-interacting cells and 20 rolling cells in order IV venules is shown. Sticking fraction was the percentage of rolling cells that adhered in HEV for ≥30 s. Sticking efficiency was the percentage of total cells that arrest on vessel walls for ≥30 s [91]. See also Video 1, available at http://www.roswellpark.org/sse/cii2005.

Fig. 2

Expression of PNAd and JAM-1 in PLN HEV. PNAd expression was analyzed on cuboidal HEV of peripheral LN cryosections (9 μm-thick) by immunohistochemical staining (left panel; note brown staining of individual HEC by rat anti-mouse PNAd primary mAb [BD Bioscience, San Diego, CA, USA] and biotin-conjugated goat anti-rat secondary Ab [BD Bioscience]). JAM-1 on HEV was detected by immunofluorescent staining (right panel; green fluorescent staining with goat anti-mouse JAM-1 primary antibody [R&D System, Minneapolis, MN, USA] and FITC-conjugated mouse anti-goat secondary Ab [Jackson ImmunoResearch, West Grove, PA, USA])

Fig. 3

Fever-range thermal stress activates lymphocyte adhesion to PLN HEV in vitro. Lymphocytes were isolated from spleen (SPC) or LN organs (LNC; pooled PLN and MLN) of BALB/c mice and then cultured in vitro at 37°C or 40°C for 6 h. Lymphocyte adherence to HEV in cryosections of BALB/c PLN was evaluated under mechanical shear as described in [14, 15, 41]. Photomicrographs show typical images of toluidine-stained lymph node cells (LNC) (black arrows) bound to HEV in PLN tissue cryosections. The number of adherent lymphocytes was quantified by light microscopy (Olympus, Spectra Services Inc., Ontario, NY, USA) in a total of 300–500 HEV in PLN cryosections. For consistency in double-blind evaluation, HEV were quantified only if they contained ≥1 adherent cell. The dotted line indicates the level of adhesion when lymphocytes were treated with functional blocking antibody to mouse L-selectin (Mel-14; American Type Culture Collection [ATCC, Rockville, MD, USA]). Data are the mean ± SE of triplicate samples in two experiments. Results are representative of ≥3 experiments. The differences between adhesion of untreated cells and hyperthermia-treated cells were significant, *P<0.0001, using an unpaired two-tailed Student’s t test

Fig. 4

Fever-range WBH stimulates lymphocyte homing to PLN in vivo. Calcein-labeled splenocytes were injected intravenously (5×10⁷ cells/mouse) into normothermal (NT) control BALB/c mice (core temperature, 36.8±0.2°C) or mice pretreated with fever-range WBH (core temperature, 39.5±0.5°C, 6 h) and allowed to resume normothermal temperatures, as described in [38, 41]. After 1 h, PLN and pancreatic organs were removed and cryosections were prepared. Calcein-labeled green-fluorescent cells were observed and quantified by fluorescence microscopy. A Micrographs are images from different organs; the arrows indicate the typical morphology of calcein-labeled cells that were included in the quantification. B Numbers of fluorescent cells were counted in 10 fields (0.335 mm²/field) of non-sequential tissue sections. Data are the mean ± SE (n=2 mice per group; data are representative of four independent experiments). The differences between splenocyte homing to PLN in NT control mice and WBH-treated mice were significant, *P<0.0001, using an unpaired two-tailed Student’s t test

Fig. 5

Analysis of blood flow in s.c. murine colon 26 tumors by intravital microscopy. Dorsal skinfold window chambers were implanted in BALB/c mice as described in [92, 93]. In brief, a 12-mm-diameter hole was dissected through one layer of dorsal skinfold to expose the fascial plane in the other layer of skinfold. Colon 26 cells (2×10⁴) were injected into the fascial plane at the time of surgery. In 9–14 days, tumors grew to 3–4 mm in diameter and were well vascularized inside the window chamber. The structure of tumor microvessels was observed under epifluorescence light microscopy (left panel). Blood flow in the same field was detected by injection of fluorescent-labeled FITC-dextran (10 mg/ml, 10 ml/kg body weight, Sigma-Aldrich, St. Louis, MO, USA) via the tail vein (right panel). See also Video 2, available at http://www.roswellpark.org/sse/cii2005.

Fig. 6

CD3 lymphocyte infiltration and expression of adhesion molecules is restricted to the peritumoral region of RIP-Tag5 pancreatic tumors. Dense leukocyte (L) infiltrates containing CD3⁺ T cells (indicated by brown staining obtained using rat anti-CD3 primary mAb [Serotec, Raleigh, NC, USA] and biotin-conjugated goat anti-rat secondary Ab [BD Bioscience, San Diego, CA, USA]) were detected in RIP-Tag5 mice (22–23 weeks) by immunohistochemical staining in the peritumoral region, outside of the edge of pancreatic islet tumors (Tu) demarked by the capsule (C). An enlargement of the designated region is shown in the inset in the upper left panel. CD3⁺ T cells were rarely observed inside pancreatic islet tumors or in exocrine (E) pancreatic tissues. SV40 large T antigen expression was detected in pancreatic islet tumor cells, but not in exocrine pancreatic tissues by immunofluorescence staining (mouse anti-SV40 large T antigen primary mAb [BD Bioscience] and FITC-labeled goat anti-mouse secondary Ab [BD Bioscience]). Immunofluorescence microscopy revealed that vessels expressing the pan-endothelial adhesion molecule, CD31, (indicated by green fluorescence staining obtained using rat anti-mouse CD31 primary mAb [BD Bioscience] and FITC-labeled goat anti-rat secondary Ab [BD Bioscience]) are evident throughout the intratumoral region, in the exocrine pancreatic tissue and in the capsular region, while expression of ICAM-1 (indicated by red fluorescent staining obtained using hamster anti-mouse ICAM-1 mAb [BD Bioscience] and PE-conjugated mouse anti-hamster secondary Ab [BD Bioscience]) was primarily limited to the peritumoral region associated with the tumor capsule

Fig. 7

Model for regulation of vascular adhesion and trafficking in response to fever-range thermal stress. Fever-range thermal stress acts independently on lymphocytes and cuboidal HEV to enhance trafficking in LN and PP HEV. No change in vascular adhesion or homing is observed in response to thermal stress across squamous, non-activated endothelium of extralymphoid organs. It remains to be determined if thermal enhancement of lymphocyte infiltration in tumor sites is mediated by changes in adhesion in tumor microvessels