Complement Factor H in cSCC: Evidence of a Link Between Sun Exposure and Immunosuppression in Skin Cancer Progression

Abstract

Cutaneous squamous cell carcinoma (cSCC) is a common form of skin cancer with an estimated 750,000 cases diagnosed annually in the United States. Most cases are successfully treated with a simple excision procedure, but ~5% of cases metastasize and have a 5-year survival rate of 25-45%. Thus, identification of biomarkers correlated to cSCC progression may be useful in the early identification of high-risk cSCC and in the development of new therapeutic strategies. This work investigates the role of complement factor H (CFH) in the development of cSCC. CFH is a regulatory component of the complement cascade which affects cell mediated immune responses and increases in complement proteins are associated with poor outcomes in multiple cancer types. We provide evidence that sun exposure may increase levels of CFH, suggesting an immunomodulatory role for CFH early in the development of cSCC. We then document increased levels of CFH in cSCC samples, compared to adjacent normal tissue (ANT) routinely excised in a dermatology clinic which, in paired samples, received the same level of sun exposure. We also provide evidence that levels of CFH are even greater in more advanced cases of cSCC. To provide a potential link between CFH and immune modulation, we assessed immune system function by measuring interferon gamma (IFN-γ) and FOXP3 in patient samples. IFN-γ levels were unchanged in cSCC relative to ANT which is consistent with an ineffective cell-mediated immune response. FOXP3 was used to assess prevalence of regulatory T cells within the tissues, indicating either a derailed or inhibitory immune response. Our data suggest that FOXP3 levels are higher in cSCC than in ANT. Our current working model is that increased CFH downstream of sun exposure is an early event in the development of cSCC as it interferes with proper immune surveillance and decreases the effectiveness of the immune response, and creates a more immunosuppressive environment, thus promoting cSCC progression.

Keywords: Complement Factor H; FOXP3; complement cascade; cutaneous squamous cell carcinoma (cSCC); immunoevasion; immunomodulation; interferon gamma (IFNγ); sun exposure.

Figure 1

CFH Transcript Number is Increased in Sun-Exposed Skin. GTEx data of 278 paired sun exposed (lower leg) and non-sun exposed (suprapubic) patient samples were analysed. CFH levels are increased in exposed skin (p<0.0001, two-tailed Wilcoxon matched-pairs signed rank test) but not CFI (A, C). Levels of CFH mRNA (TPM) in exposed skin correlates with unexposed skin (p<0.0001, Spearman’s rank-order correlation) (B). When sorted by non-exposed CFH levels, the levels and fold increase in paired exposed samples decreases with the exception being at the highest values of CFH in non-exposed skin (D). (****p<0.0001; ns, not significant).

Figure 2

CFH and FH-L Expression in cSCC is Increased. Immunofluorescent microscopy of cells cultured from cSCC tissue show CFH staining (green) in cytosolic vesicles (A). Bands at the expected molecular weight for CFH and Factor H-like (FH-L) are detected in both adjacent normal tissue and cSCC samples (B). The ratio of CFH and FH-L band intensities, normalized to histone H3 intensity, was higher in cSCC tissue compared to ANT [(C), p=0.031 & (D), p=0.034, respectively]. In paired samples for non-invasive cSCC, CFH levels normalized to GAPDH are not significant. When CFH levels in ANT are compared to invasive cSCC, the difference is highly significant [(E), p<0.0001]. (*p<0.05; ***p=0.0001; ****p<0.0001; ns, not significant).

Figure 3

IFN-γ Expression in cSCC is Unchanged. Bands at the expected molecular weight for glycosylated IFN-γ monomer and IFN-γ dimer are detected at in both adjacent normal tissue and cSCC samples (A). The ratio of intensity for both IFN-γ bands in paired samples was normalized to histone H3 intensity and is not significant (ns) (B). The ratio of intensity for both IFN-γ bands in unpaired non-invasive and invasive cSCC samples normalized to GAPDH compared to control was also not significant (ns) (C).

Figure 4

Levels of FOXP3 are Increased in cSCC. Immunoblots detected a band at the expected molecular weight for FOXP3 in both adjacent normal tissue and cSCC samples (A). The ratio of FOXP3 band intensity normalized to GAPDH intensity was higher in cSCC tissue compared to paired ANT [(B), p<0.001] and non-invasive cSCC compared to unpaired ANT but there is no difference between non-invasive and invasive cSCC (C). (*p<0.05; ***p=0.0001; ns, not significant).

Figure 5

CFH in Routine Mohs and Advanced cSCC Samples by IHC. cSCC removed from routine clinic patients by Mohs surgery [fixed after cryosectioning; (A)] and an array of advanced cSCC [formalin fixed; (B)] were labeled with mouse anti-CFH (OX-24) and an AP-conjugated secondary antibody with permanent red stain. Mayer’s Hematoxylin was used as a counterstain (nuclei). The degree of CFH staining in the ANT and cSCC samples was semi-quantitatively determined by using a 0-3+ scale, with 0 indicating no staining, 1+ indicating <10% staining, 2+ indicating 10-50% staining, and 3+ indicating >50% staining. ANT samples [(A, B), top panels] were scored as either 0 or 1+, the Mohs cSCC samples [(A), bottom panel] were scored as 2+, and the advanced cSCC samples [(B), bottom panel] were scored as 3+. The boxes within the 40x and 100x images delineate the tissue location shown in the 100x and 400x images, respectively.

Figure 6

FOXP3 in Routine Mohs and Advanced cSCC Samples by IHC. cSCC removed from routine clinic patients by Mohs surgery [fixed after cryosectioning; (A)] and an array of advanced cSCC [formalin fixed; (B)] were labeled with a rabbit monoclonal anti-FOXP3 (Cell Marque) and stained with an HRP-conjugated secondary antibody and DAB stain (for Array samples) and with an AP-conjugated secondary antibody and permanent red stain (for ALBR samples). Mayer’s Hematoxylin was used as a counterstain (nuclei). The degree of FOXP3 staining in the ANT and cSCC samples was semi-quantitatively determined by using a 0-3+ scale, with 0 indicating no staining, 1+ indicating <10% staining, and 2+ indicating 10-50% staining, and 3+ indicating >50% staining. ANT samples [(A, B), top panels] were scored as 0, the Mohs cSCC samples [(A), bottom panel; arrows in the 400x image denote positive nuclear localization] were scored as 1+, and the advanced cSCC samples [(B), bottom panel; arrows in the 400x image denote positive nuclear localization] were scored as 2+. The boxes within the 40x and 100x images delineate the tissue location shown in the 100x and 400x images, respectively.