Quantitative assessment of STAT3 and HPV16 E6 transcripts using Flow-FISH approach for early detection of progressive cervical lesions

Abstract

BACKGROUND

Dysregulated signal transducer and activator of transcription 3 (STAT3) signaling is a key feature of human papillomavirus (HPV)-driven cervical carcinogenesis. Our earlier work demonstrated elevated STAT3 expression plays a regulatory role in oncogenic transcription of HPV16 E6/E7. A combined analysis of STAT3 and HPV E6/E7 mRNA by fluorescence in situ hybridization (FISH) showed diagnostic clinical relevance in screening for pre-cancerous cervical lesions. However, use of FISH does not provide clear objective diagnostic threshold to establish accurate signal positivity.

AIM

To assess the feasibility of FISH for quantitative detection of STAT3 and HPV E6/E7 transcripts using a flow cytometry-based approach.

METHODS

HPV-negative (C33a), HPV16-positive (SiHa) and HPV18-positive (HeLa) cervical cancer cell lines were analyzed for STAT3 and HPV E6/E7 transcript expression using FISH approach. Signal specificity and distribution were assessed by fluorescence microscopy using target-specific and scrambled probes. The same probes were further evaluated using flow cytometry-based FISH to determine their suitability for quantitative transcript detection.

RESULTS

Fluorescence microscopy revealed strong and discrete STAT3-associated signals in SiHa and HeLa cells, whereas C33a cells exhibited comparatively diffuse fluorescence. Scrambled control probes produced minimal background staining, supporting the specificity of STAT3 probe. HPV16 E6 probe also produced detectable signals but comparable fluorescence intensity was observed across all cell lines irrespective of HPV status and was similar to scrambled probe controls. Moreover, when assessed by flow cytometry-based FISH, the same probes displayed limited performance. STAT3-positive populations were low and accounted for approximately 10% of cells, while no clearly distinguishable HPV16 E6-positive population could be identified in any of the tested cell lines. These findings indicate that optimized microscopy-based FISH assay may not be directly compatible with flow-based transcript detection platforms.

CONCLUSION

Therefore, re-optimization of assay is required for quantitative transcript detection for HPV-associated cervical cancer screening using flow cytometry-based FISH. The manuscript addresses the bottlenecks and potential strategies to mitigate the issues.

Key Words: Cervical cancer; Human papillomavirus; Signal transducer and activator of transcription 3; Fluorescence in-situ hybridization; Flow-cytometry

Core Tip: This study evaluates the feasibility of translating a microscopy-optimized fluorescence in situ hybridization (FISH) assay for signal transducer and activator of transcription 3 and human papillomavirus (HPV) 16 E6 transcripts into a flow cytometry-based FISH (Flow-FISH). While microscopy-based FISH reliably detected transcripts in cervical cancer cell lines, pre-published probes demonstrated limited sensitivity and specificity in Flow-FISH, particularly for HPV16 E6. These findings highlight critical technical bottlenecks in adapting slide-based FISH assays to flow-cytometry and underscore the need for platform-specific probe design and assay re-optimization before Flow-FISH can be applied as a scalable and objective screening tool for HPV-associated cervical cancer.

INTRODUCTION

Cervical cancer (CaCx) is a major threat to women’s health globally. It is the fourth most common malignancy in women worldwide with an estimated annual incidence and mortality of 604127 and 341831, respectively[1]. More than 98% of CaCx cases is caused by persistent infection of high risk (HR)-human papillomavirus (HPVs). Among these HR-HPVs, HPV16 is the most prevalent[2] and is strongly associated with cervical carcinogenesis through sustained expression of the viral oncogenes E6 and E7[3]. These oncogenes drive malignant transformation by inactivating key tumor suppressor genes p53[4] and pRb[5] as well as reprogramming host transcriptional networks[6]. Consequently, detection of HPV oncogene expression at the transcript level has emerged as a clinically relevant strategy for identifying biologically active infections associated with disease progression[7,8]. Nevertheless, despite its applicability, current diagnostic approaches for identifying transcriptionally active HPV remain limited, as they show high positive predictive value but do not reliably predict whether a lesion will subsequently regress or progress.

This necessitates inclusion of other markers with HPV mRNA based test to increase its negative predictive value. There are studies of combined evaluation of HPV mRNA/protein with Ki67, p16, p27 in research use but they are not available clinically[9-12]. Our previous study revealed that there is a strong correlation between constitutively active host transcription factor signal transducer and activator of transcription 3 (STAT3) and HPVs E6/E7 oncoproteins. The persistent activation of STAT3 is closely related to the occurrence of HPVs in CaCx. Increased STAT3 activity in CaCx cells is associated with elevated expression of HPV16 E6 and E7[13,14], HPV-induced cervical carcinogenesis[15,16]. These leads established a strong biomarker value of STAT3 in HPV oncogene expression and disease progression during cervical carcinogenesis.

A combined analysis of STAT3 and HPV E6/E7 mRNA using fluorescence in situ hybridization (FISH) assay has been used in screening for pre-cancerous cervical lesions[17]. This microscopy-based FISH approach has demonstrated feasibility for detecting HPV oncogene transcripts in clinical samples of Cacx. However, such methods are inherently limited by qualitative or semi-quantitative interpretation, lack of objective signal thresholds, low throughput, and observer-dependent variability[18]. These limitations restrict their broader clinical applicability, particularly in large-scale screening settings where standardized and quantitative readouts are essential.

Flow cytometry-based FISH (Flow-FISH) represents a potential advancement of FISH by combining molecular specificity with high-throughput, single-cell quantitative analysis. This approach offers objective signal measurement making it valuable for translational and diagnostic applications. In this context, the present study aimed to evaluate the feasibility of translating a previously established FISH-based detection strategy for quantitative detection of STAT3 and HPV16 E6 transcripts on a flow cytometry-based platform.

MATERIALS AND METHODS

Cell culture

C33a cells were cultured in minimum essential medium, whereas SiHa and HeLa cells were grown in Dulbecco’s modified eagle’s medium. All media were supplemented with 10% fetal bovine serum and 1 × antibiotic-antimycotic solution. Cultures were maintained at 37 °C in a humidified incubator with 5% CO₂.

Total RNA isolation and cDNA preparation

Total cellular RNA from cells was isolated by using TRI reagent (Invitrogen) using previously published protocol[19]. The concentration of the isolated RNA was quantified using NanoDrop™ One (Thermo Fisher Scientific) using nuclease free water as blank and quality was assured as per the readings of A260/A280 and A260/A280 ratios. Using the Verso cDNA Kit (Thermo Scientific), a minimum of 2.0 μg of sample was used for cDNA synthesis in a 20 μL reaction.

Real-time reverse transcription-polymerase chain reaction

Real-time reverse transcription-polymerase chain reaction (RT-PCR) amplification of HPV16/18 E6 and E7, STAT3, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (internal control) transcripts was carried out using a CFX-96™ thermal cycler (96-well block) in a 10 μL reaction system. Primer sequences, annealing temperatures, and reaction conditions are provided in Supplementary Table 1. No template control was used as the negative control. For positive controls, the International Standard control for HPV16 was used for SiHa cells, and the International Standard control for HPV18 was used for HeLa cells. For GAPDH and STAT3, pre-confirmed positive samples were used.

Synthesis of probes

The pre-published probe sequences were used for preparation of STAT3, and HPV16 E6 and E7 probes[17] and were synthesized and procured from Merck-Sigma (Supplementary Table 2). Sequences were cross checked and verified by comparing with STAT3 transcript variant 1 (NM_139276.3) and HPV16 E6E7 transcript variant (NC_001526.4) using N-BLAST online tool (http://www.ncbi.nlm.nih.gov/bLAST/). STAT3 probe was labeled with Alexa flour 488, HPV16 E6 and E7 were tagged with biotin at the 5’ end. The single locus scrambled probes (SCR) were designed using the Sequence Scramble Bioinformatics tool available on the online platform Genescript (https://www.genscript.com/tools/create-scrambled-sequence). Initially, the species was selected, and the input sequence of the probe for a specific gene was provided, followed by the submission of the sequence. After generating 1000000 combinations during the scrambling process, a scrambled sequence was produced. Sequences were compared using N-BLAST online tool (http://www.ncbi.nlm.nih.gov/bLAST/) for any lack of any non-specific binding. STAT3 scrambled probe (STAT3-SCR) were labeled with Alexa flour 488 and HPV16 E6 and E7 scrambled probes (HPV16 E6-SCR) were tagged with biotin at the 5’ end and were visualized using Streptavidin-fluorescein isothiocyanate (BD Biosciences, United States).

FISH

Adherent cells were grown on 18 mm cover glass in 6-well cell culture plate, washed with PBS with 0.2 mg/mL RNase-free bovine serum albumin in nuclease-free water. Then the cells were fixed in 250 μL BD PermFix buffer at room temperature for 40 minutes and washed twice with 500 μL BD PermWash for 5 minutes each. Permeabilized cells were equilibrated using wash buffer and incubated at room temperature for 5 minutes. The cells were then incubated with 100 μL drop of Hybridization Buffer containing probe at concentration of 0.1 μmol/L in the dark at 37 °C for overnight incubation. The cells were then washed again by wash buffer for 30 minutes in the dark at 37 °C followed by another wash for 2-5 minutes in the dark at RT. For biotin labelled probes, an additional incubation step with fluorescein isothiocyanate-labelled streptavidin in ratio 1:2000 was performed. The washed cells were mounted on glass slides using a small drop of approximately 15 μL of Flouroshield Mounting Medium.

Image acquisition and analysis

Image acquisition was performed by selecting a minimum of 5 and a maximum of 14 cells per field in each cell line. The individual cells were manually outlined as regions of interest, and total fluorescence intensity (TFI = area × mean intensity) was calculated for each cell. Average TFI values were used for graphical representation. Although some representative fields may appear visually brighter, variability in cell size and intracellular signal distribution can result in apparent differences that are not proportionally reflected in averaged data. All images were acquired using identical exposure settings, and fields were selected randomly to avoid bias.

Flow-cytometry

Cells were trypsinized and made single cell suspension. Cells were hybridized with probes in suspension as described earlier for adherent cells. Immediately following hybridization and washing, cells were subjected to flow cytometry analysis in wash buffer. Flow cytometry was performed on FACSAria III (BD Biosciences) flow cytometer (530/30 nm emission filter) for AlexaFL488 detection. Prior to screening for the appropriate fluorochrome, cells were gated based on size and granularity. Data analysis was performed with either FACSDiva or with FlowJo (Becton Dickinson, San Jose, CA, United States) software. Cells were gated on singlets and percent populations were determined with unstained or stained negative controls as required.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 10.3.1. Data distribution was assessed for normality prior to hypothesis testing. For datasets with adequate sample size, normality was evaluated using the Shapiro Wilk, D’Agostino Pearson, and Anderson Darling tests. For datasets with small sample sizes, normality was assessed using the Shapiro-Wilk test. Datasets that satisfied the assumption of normality were analyzed using parametric tests, including the unpaired Student’s t-test or one-way analysis of variance, followed by Sidak’s multiple comparisons test where appropriate. When the assumption of normality was not met, non-parametric analyses were applied using the Kruskal-Wallis test followed by Dunn’s multiple comparisons test. A P value < 0.05 was considered statistically significant.

RESULTS

Detection of STAT3 and HPV16 E6 transcripts by RT-PCR

The presence of STAT3, HPV16 E6E7, and HPV18 E6E7 transcripts were confirmed by RT-PCR in CaCx cell lines C33a, SiHa and HeLa using cDNA prepared from isolated RNA of these cell lines (Figure 1). All CaCx cell lines showed amplification of STAT3 transcripts. SiHa and HeLa showed detectable and type-specific oncogene E6E7 transcripts, corresponding to HPV16 and HPV18, respectively. Multiple bands of different E6 splice variants were detected in HeLa cell cDNA. RNA integrity was confirmed by amplifying GAPDH.

Figure 1 Analysis of human papillomavirus and signal transducer and activator of transcription 3 transcripts in cervical cancer cell line using real-time reverse transcription-polymerase chain reaction. ¹Spliced variant of E6. A: Representative gel images showing amplicons of reverse transcriptase-glyceraldehyde-3-phosphate dehydrogenase; B: RT-human papillomavirus (HPV) 16 E6; C: RT-HPV16 E7; D: RT-signal transducer and activator of transcription 3; E: RT-HPV18 E6. In addition to the full-length HPV18 E6 transcript, a 245 bp amplicon corresponding to the spliced variant HPV18 E6 was observed; F: RT-HPV18 E7. RT-PCR: Real-time reverse transcription-polymerase chain reaction; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; HPV: Human papillomavirus; STAT3: Signal transducer and activator of transcription 3.

Detection and quantification of STAT3 and HPV16 E6 transcripts in CaCx cell lines using fluorescence microscopy

Next, the presence of STAT3 and HPV16 E6 transcripts in CaCx cell lines was detected using fluorescence microscopy. The assay was performed using pre-published probes reported to detect these transcripts. STAT3 probes showed signals in all three cell lines, however the intensity varied. C33a showed the lowest fluorescence intensity/cell whereas HeLa showed the highest fluorescence intensity indicating higher STAT3 positivity (Figure 2A). In contrast, HPV16 probes showed positivity in HPV16 infected SiHa cells, however diffused signals were also detected in HPV-negative C33a cells and HPV18-positive HeLa cells (Figure 2B). Upon quantitative evaluation of TFI, there was no difference found in the extent of signals in all the three cell lines for HPV16 E6.

Figure 2 Evaluation and quantitation of signal transducer and activator of transcription 3 and human papillomavirus 16 E6 transcripts in CaCx cell lines using fluorescence microscopy. Data are presented as mean ± SD. ^aP < 0.05 vs C33a. A: Representative fluorescence photomicrographs showing fluorescence intensity of signal transducer and activator of transcription 3 (STAT3); B: Representative fluorescence photomicrographs showing fluorescence intensity of human papillomavirus (HPV) 16 E6. CaCx cells were grown on cover glass, fixed with BD Permfix and stained with STAT3 probes labelled with Alexa Fluor 488 and biotin-linked HPV16 probes visualized with streptavidin-fluorescein isothiocyanate. Cells were counterstained with 4’,6-diamidino-2-phenylindole (blue) and examined at magnification 1000 ×. Bar graphs represent the mean fluorescence intensity of STAT3 and HPV16 E6 stained cells. HPV: Human papillomavirus; STAT3: Signal transducer and activator of transcription 3.

Estimation of specificity of probes

To further assess the specificity of the signals observed from probe-stained cells, the samples were stained in parallel with custom-designed (GeneScript) non-specific scrambled probes labelled with corresponding fluorescence of the probes. Cells stained with SCR-STAT3 showed signal, however the intensity was low compared to STAT3 probes (Figure 3A, upper panel). Same experiment was repeated with higher probe concentration, and the result remained same (Figure 3A, lower panel). In contrast, there was no difference found in the signals for HPV16 E6 probes even after increase in the probe concentration (Figure 3B). Moreover, the scrambled probes produced the signal intensity similar to the HPV16 E6 probe producing non-specific signals.

Figure 3 Estimation of specificity of signal transducer and activator of transcription 3 and human papillomavirus 16 E6 at different probe concentrations. ^aP < 0.05 vs C33a scrambled, ^bP < 0.0001 vs SiHa scrambled. A: Representative fluorescence photomicrographs showing expression of signal transducer and activator of transcription 3 (STAT3) transcripts in C33a and SiHa at different concentrations of the probe. The upper panel represent the signal intensity at 0.1 μmol/L probe concentration and lower panel represent the signal intensity at 0.2 μmol/L probe concentration. Non-specific scrambled probe was used as a negative control to estimate the label of non-specific binding of the probe (scale bar: 20 μm). Bar graph showing fluorescence intensity of STAT3 probes and STAT3 scrambled probes; B: Representative fluorescence photomicrographs showing expression of human papillomavirus (HPV) 16 E6E7 transcripts in SiHa (scale bar: 20 μm). HPV16E6 scrambled probe was used as a negative control to estimate the specificity of the probe. Bar graph showing fluorescence intensity of HPV16 E6 probes and HPV16 E6 scrambled probes in SiHa. HPV: Human papillomavirus; STAT3: Signal transducer and activator of transcription 3; SCR: Single locus scrambled probes.

Detection and quantification of STAT3 and HPV16 E6 transcripts in CaCx cell lines using flow cytometry

The STAT3 and HPV16 E6-mRNA positive cells were quantified by probes using respective probe-FISH followed by flowcytometric evaluation. Probe positivity was set against unstained control population where 95% unstained cells on left side of the fluorescence channel were gated out as negative and 5% was attributed to auto-fluorescence in all the analysis. With respect to the 5% cut off, nearly 10% of probe-stained cells showed detectable STAT3 signal that also included sample auto-fluorescence (represented as dotted line) (Figure 4A). Overall, STAT3-probe labelled cells showed poor resolution from the unstained control. Similarly, HPV16 E6 probes also showed very low non-specific detection of HPV16 E6-mRNA positive cells (Figure 4B).

Figure 4 Evaluation and quantitation of signal transducer and activator of transcription 3 and human papillomavirus 16 E6 transcripts using flow-cytometry. The bar graph shows the percent population of signal transducer and activator of transcription 3 and human papillomavirus 16 E6 positive cells in cervical cancer cell lines. Horizontal dotted bar represents the cutoff of 5% with respect to unstained control. A: Flow-cytometric evaluation of cervical cancer cells hybridized to signal transducer and activator of transcription 3 probes; B: Flow-cytometric evaluation of cervical cancer cells hybridized to human papillomavirus 16 E6 probes. HPV: Human papillomavirus; STAT3: Signal transducer and activator of transcription 3.

DISCUSSION

The present study aimed to assess the feasibility of translating a microscopy-based FISH assay in a flow cytometry-based platform for quantitative detection of STAT3 and HPV16 E6 transcripts. The fluorescence microscopy analysis, using pre-published probes, demonstrated strong and specific STAT3 signals in HeLa and SiHa cells whereas, the HPV16 E6 specific did not show differential detection, exhibiting comparable fluorescence across all cell lines irrespective of HPV status. Scrambled probes used as specificity controls produced minimal background for STAT3, confirming probe specificity, whereas the HPV16 E6 probe displayed non-specific binding with signal intensities similar to the scrambled control. Further evaluation by flow-cytometry revealed limited assay sensitivity, with low STAT3 positivity (about 10%) and no clearly distinguishable HPV16 E6-positive populations across the tested cell lines thereby compromising the efficiency of assay for quantitative flow-cytometry based detection.

Diagnostic and prognostic markers are of utmost importance in clinical medicine for accurate and early detection, diagnosis and prognosis of CaCx. The E6 and E7 gene of HR-HPV have been implicated in the early development of CaCx, as it plays a role in host cell cycle regulation and viral genome integration, contributing to carcinogenesis[20]. The expression of HR-HPV E6 and E7 mRNA indicates that the virus has started the process of integration and carcinogenesis. Another marker, STAT3, a member of the signal transducers and activators of transcription family is implicated as a key factor in CaCx progression, with studies showing a strong correlation between HPV16 E6 and E7 expression and STAT3 activation. STAT3 plays a regulatory role in oncogenic transcription of HPV16 E6 and E7[14]. Despite its clinical significance, no diagnostic or prognostic assays targeting STAT3 have been developed.

Currently, considerable efforts are being directed toward the detection of these transcripts using diverse molecular assays. United States published application No. US20120214152A1 describes an RNA scope-based assay for targeting HPV E6 and E7 mRNA in head and neck as well as CaCx, employing RNA in situ hybridization. Another United States published application No. US20180230556A1, outlines a rapid assay for HPV E6 and E7 mRNA detection using a combination of ISH and flow-cytometry to identify pre-cancerous and malignant cervical cells in liquid-based cervical cytology samples. Although this approach incorporates signal amplification strategies for single-target detection, it does not evaluate lesion progression and requires specialized instrumentation, thereby limiting its accessibility, particularly in low-resource clinical settings. Similarly, Enzo Life Sciences previously developed a flow cytometry-compatible FLOWSCRIPT E6/E7 assay for single-cell detection of HPV E6/E7 transcripts, demonstrating the feasibility of flow-based identification of transcriptionally active HPV. However, the assay relied on proprietary reagents and specialized instrumentation, limiting its broader applicability. Moreover, the platform has since been discontinued, highlighting the unmet need for alternative, adaptable flow-based assays capable of detecting HPV oncogenic transcripts alongside other regulatory markers.

In another study, Fan and Shen[17] demonstrated the feasibility of detecting HPV E6/E7 transcripts using FISH-based microscopy in cervical lesions with potential diagnostic relevance. While microscopy-based FISH offers high spatial resolution and allows visualization of intracellular transcript localization, it is inherently limited by qualitative or semi-quantitative interpretation. Moreover, it does not provide a clear objective diagnostic threshold to establish accurate signal positivity. The signal positivity often depends on subjective assessment, manual thresholding, and field selection, which can introduce observer biasness and limit reproducibility, particularly in a clinical screening context[18]. Also, the operation of FISH is considered complicated requiring higher level of technical personnel and experimental apparatus making it a costly technique. In addition, slide-based FISH preparation can be technically demanding, as uneven cell attachment, sensitivity to fixation conditions, and careful handling during multiple washing and processing steps often affect signal quality and reproducibility, making consistent slide preparation challenging[21]. These limitations underscore the challenge of translating microscopy-dependent FISH assays into objective and scalable diagnostic tools.

Flow-cytometry on the other hand offers several advantages over microscopy-based methods, including objective quantification and the ability to define reproducible diagnostic threshold to establish accurate signal positivity. Flow-FISH combines the molecular specificity of FISH with the quantitative power of flow cytometry, making it more relevant for clinical and translational applications. However, our findings from present study indicate that probe performance, signal amplification, fixation, permeabilization, and hybridization conditions optimized for microscopy are not directly transferable to flow-based platforms.

The published probes for STAT3 and HPV reported earlier performed very weakly and their signals were not very specific and limited applicability in flowcytometry. The reason for observed differences between our study and of published study[17] are not clear. However, difference in hybridization conditions cannot be ruled out. These technical challenges highlight the need for platform-specific probe design and assay optimization when adapting FISH assays for flow cytometric applications.

Probe specificity can be improved by targeting multiple non-overlapping regions of the same transcript and by avoiding repetitive or homologous sequences that contribute to off-target hybridization[22]. In silico validation should be performed against the human transcriptome and HPV genomic variants to minimize cross-reactivity of these probes. Stringent hybridization and washing conditions should be optimized for flow-cytometry to increase signal intensity[23]. Moreover, co-detection of oncogenic transcripts can also be applied to increase signal specificity; however, probes should be labeled with spectrally distinct fluorophores to avoid spectral overlap and minimize the need for compensation. Collectively, these approaches can be used to increase the signal intensity and specificity as well as setting up the threshold for accurate detection of signal positivity for determination of progressive cervical lesions.

The present study demonstrates that feasibility of extending microscopy-based FISH assay to flow cytometry-based platforms is not straightforward. The limited sensitivity of pre-published probes observed in Flow-FISH, emphasizes the necessity for comprehensive re-optimization of probe chemistry and assay conditions. Such optimization is essential to realize the full potential of Flow-FISH as a quantitative, scalable, and clinically applicable tool for HPV-associated CaCx screening.

CONCLUSION

The present study demonstrates that while FISH assay enables reliable detection of STAT3 and HPV16 E6 transcripts in HPV-associated CaCx cell lines, its direct translation to a flow cytometry-based platform is limited by reduced sensitivity and inconsistent signal detection, particularly for HPV16 E6 transcripts. These findings highlight that FISH assays optimized for microscopy are not compatible with quantitative flow-based detection without substantial methodological modification. Future efforts should focus on re-engineering probe design and hybridization conditions specifically tailored for Flow-FISH platforms. Such advancements would be critical for translating combined STAT3 and HPV oncogene transcript analysis into a scalable and clinically applicable screening approach for CaCx.