Detection of Human Papillomavirus Genotypes in Breast Cancer Patients

Background: The role of viruses in the aetiology of breast cancer (BC) have been signicantly investigated in recent years. Human papillomavirus (HPV) has been detected in invasive breast carcinomas, and most studies have only focused on the detection of viral DNA therefore the aim of this study was to determine the genotypes of each papillomaviruses. Methods: We collected and analyzed 70 blocks of Formalin Fixed Paran Embedded (FFPE) including 59 BC samples, and 11 benign breast lesions as control from Iranian patients using nested PCR. Real-time PCR supported nested PCR ndings. Genotyping of HPV positive samples was performed, the samples were also subjected to a multiplex PCR to detect herpes simplex-1 (HSV-1), herpes simplex-2 (HSV-2), varicella zoster virus (VZV) and cytomegalovirus (CMV) in BC. Results: Papillomavirus DNA was present in 7 of 59 BC samples (11.8%); while none was detected in control samples. The most prevalent type was HPV18, followed by HPV 6, but we didn’t nd HPV16 in these cases. All HPV positive patients had high tumor grades (II/ III), and the age range of them were from 33 to 73 years (average was about 51). Most of HPV positive patients had low levels of education. Histologic diagnosis of all positive cases were ductal carcinoma. On the other hand 5 of 59 BC specimens (8.47%), were positive for HSV-1. Conclusions: Our results suggest that high risk HPVs may play a role in breast carcinogenesis. Also we detected HPV6 and HSV1 in our BC samples, taking into account a carcinogenesis role for low risk HPVs, in addition to high risk HPVs. Further large scale studies are warranted to assess the signicance of our ndings.

Over 200 HPV types have been identi ed and most of them fall within three genera: alpha (α), beta (ß) and gamma ( ). Alpha (α) mainly found on mucosal surfaces, beta (ß) and gamma ( ) are generally found on skin and hair follicles (23,24). HPVs are classi ed into low and high-risk types based on their oncogenic potential (25). Low risk HPVs although not usually associated with cancer development, can cause benign genital warts. The HPV types that have the ability to cause cancer are considered high risk. Low risk HPVs are including: 6, 11, 34, 40, 42, 43, 44, 53, 54, 66, 70, and 74. High risk HPV types are 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 69 (26, 27). For the rst time in 1992 Di Lonardo A and colleagues suggested that HPV could play a crucial role in the genesis of breast carcinoma by detecting HPV16 DNA sequence in 29.4% of breast and lymph node samples (28). Following the rst study, several studies have shown the presence of HPV in malignant breast tumors (29)(30)(31). Conversely, con icting studies were declared the negative relationship between HPV and breast cancer (32)(33)(34). So the association between HPV and breast carcinogenesis is not yet clear completely. Due to the high rate of breast cancer in Iranian women and the role of high-risk genotypes of papillomaviruses in carcinogenesis, this study was conducted to determine the presence of two high-risk genotypes HPV16 and HPV18 in malignant breast tissue.

Tissue samples
This cross sectional, case control study was performed between 2008 and 2019 in the Modarres Hospital (Tehran, Iran). A total of 70 blocks of Formalin Fixed Para n Embedded (FFPE) including 59 samples diagnosed as breast carcinomas, and 11 benign breast lesions as control were collected from the pathology department archives of Modarres hospital. Also, several parameters; such as type of BC, grading of breast cancer, age and level of education were taken in written format as exclusion criteria and summarized in Table 1. The specimens were carried to the school of Medicine in Shahid Beheshti University for the following studies.

DNA Extraction for PCR
The genomic DNA was extracted from FFPE breast tissues. Then we performed standard polymerase chain reaction (PCR) to detect HPV DNA. For extraction at rst, FFPE tissues were cut in 10 μm thickness by microtome. Depara nization was performed by adding 1 ml xylene and spinning. Afterwards we centrifuged the samples for 5 min and supernatants were removed. This step was repeated once, then 1 ml of 96% ethanol was added. Microtubes were put in 50ºC heating block until ethanol was entirely dried up. Then the digestion was performed by digestion buffer and proteinase K solution. At the end they incubated overnight, and next day they were placed in 95ºC heater. Afterwards, phenol, phenol-chloroform and only chloroform solution were added respectively followed by centrifugation and subsequent removal of the supernatant. Ultimately, ethanol was added, then they were kept in the incubator overnight.
In the last day of extraction samples were centrifuged at 12000 g for 30 min. Then threw away supernatant and left the ethanol to evaporate in 37ºC heater. Finally distilled water was added and then the DNA extracts were quanti ed with a NanoDrop spectrophotometer.

GAPDH PCR ampli cation
The quality of DNA extracted from para n-embedded tissue samples was assessed by using a forward (ATGTTCGTCATGGGTGTGAA) and reverse (GGTGCTAAGCAGTTGGTGGT) primer pair targeting a sequence within the GAPDH gene. PCR ampli cation protocol consisted of 30 cycles of denaturation at 95 °C for 30 s; hybridization at 55°C for 30 s, and elongation at 72°C for 30s. A nal elongation step was performed at 72 °C for 10 min.

HPVdetection by nested PCR
All samples were screened for the HPV L1 conserved region. The nested PCR assay was performed using two sets of primers (MY09/11 and GP5+/6+) for two consecutive ampli cation reactions. The rst reaction was performed in 25 μl using 12. Negative controls containing water instead of DNA were used. We used HPV18 HeLa cell line as positive control.

Sequencing and Phylogenetic Analysis
The positive PCR samples were sequenced for HPV genotyping. The DNA sequence was determined with the Big-Dye terminator cycle sequencing kit and an ABI 377A sequencer (Applied Biosystems Inc.).
The HPV sequences were edited with the BioEdit program version 7.2.3, and then phylogenetic and molecular analyses were organized using MEGA software program version 6.0.6 [27]. The neighbor joining and the Kimura 2-parameters methods were used for phylogenetic reconstructions that were implemented in the MEGA 6.0.6 program. Statistical signi cance for the phylogenetic tree was assessed by bootstrap method (1000 replicates).

Nucleotide Sequence Accession Number
The nucleotide sequences of HPV isolates that found out in this study have been settled in GenBank data base [accession numbers QED55703-QED55709]. The GenBank accession numbers for HPV6 types are: QED55703 and QED55704. And for HPV18 types are: QED55705, QED55706, QED55707, QED55708 and QED55709.

Real-time PCR
The PCR ampli cation was performed in a 20 μl volume containing 4 μl 5x HOT FIREPol® Probe qPCR Mix Plus (no ROX) (Solis BioDyne, Estonia), 0, 5 μl of each forward and reverse primers (Pishgam company, Tehran, Iran), 0.5 μl of probes (Sinaclon Co., Tehran, Iran) which listed in Table2, and 2 μl of each sample or control, and rest of the total volume was obtained by adding distilled water. Ampli cation and detection was performed by real-time PCR machine (Rotor-Gene-Q 6000 thermocycler (Corbett, Australia)). Thermal cycles used for quanti cation of HPV18 E6 gene and HPV type 16 E7 gene were 95°C for 12 minutes, 95°C for 15 seconds, and 60°C for 60 seconds for 40 cycles. mPCR (multiplex PCR) conditions were as follows: Initial denaturation (95°C for 5min), 35 cycles of denaturation (95°C for 30sec) annealing (60°C for 30 seconds) and extension (72°C for 30 seconds); Final extension was given at 72°C for 10 minutes. Samples were preserved at 4°C and then the PCR product was detected by gel electrophoresis.

Statistical analysis
Data was analyzed by SPSS statistical software program version 16.0. The correlations were subjected to χ2 (Pearson chi-square) and Fisher's exact test. Odds ratios and logistic regression were also calculated.
Statistical signi cance was set as a P-value less than 0.05.

Results
Study population and include criteria HPV gene sequences were identi ed. The age ranges of patients were from 29 to 81 years (mean of 50 years). In our study we had 4 patients (6.7%) with a low grade malignancy (tumor grade I), and all remaining had high grade tumor (II and III) (93.3%). Regarding histologic types, 5% of tumors were lobular types and remaining (95 %) were ductal carcinomas (Table 3). We reviewed the level of education of patients. High level of education (having university's degree) found in 24% of patients, low level (having diploma or less) in 56%, and no education in 20%.
Assessing DNA quality from FFPE samples In our study the age range of HPV positive patients were from 33 to 73 years (average is about 51). All HPV positive patients had high tumor grades (II/ III). Most of patients had low levels of education (57.1%), 14.2% had no education and remaining had universities degrees (28.5%). Histologic diagnosis of all positive cases was ductal carcinoma. (Table 3) Detection of HPV18 DNA in samples using real-time PCR Analysis of real-time PCR supported the results of nested PCR and con rmed the presence of HPV18 in 5 breast tissues. And no positivity for HPV16 were detected.
Sequencing HPV positive samples and constructing phylogenetic tree All positive PCR products were sequenced. Five out of 7 HPV positive samples were HPV18 genotype (~72%), and the rest two (~28%) were HPV6 genotype Phylogenetic tree was shown in Fig 3. According to our results, the most prevalent type was HPV18, followed by HPV 6. Contrary to our expectations, we didn't nd HPV16 in cases.
All of HPV positive patients with different genotypes, had invasive ductal carcinomas and high tumor grades. The age range of 30 to 40 years had three positive samples; 2 of them were HPV18 and one of them was HPV6. The age range of 50 to 60 years had one HPV6 and one HPV18 positive samples. The age range of 60 to 70 years and 70 to 80 years, each with one HPV18 positive samples. Most of the positive samples (80%) had high tumor grades (II/ III) ( Table 3). Three of the patients had high school diploma, one bachelor degree and the last one had no education. Therefore, we veri ed that two women (3. 3%) had HPV18 and HSV-1 coinfection.

Discussion
Incidence and mortality rates of breast cancer are enhancing considerably worldwide (40). This indicates the value of identifying risk factors that are responsible and preventable for breast cancer. Viral agents play a crucial role as a risk factor in cancer development as they are associated with nearly 20% of cancers. High risk HPVs, which can cause different cancers, based on development of molecular techniques and subsequent investigations, are also detected in breast cancer (41). Various theories indicate HR-HPVs can induce cancer by modifying cellular pathways and inducing tumorigenesis by E6 and E7 genes and suppressing the immune system (42,43). There are numerous publications studied the association of HPV and cervical cancer (44)(45)(46). But the correlation of HPV and breast cancer remains controversial. Lonardo A and his colleagues considered the correlation of HPV16 in breast cancer for the rst time in 1992, after they found this virus DNA in 29.4% of BC samples (28). Later, various studies were conducted around the world. In 2011, according to a meta-analysis study, the prevalence of HPV in breast cancer in Europe was the lowest (12.91%) but highest in Oceania (42.11%) followed by Asia (32.42%) (47). Based on their statistics and our ndings, Iranian women with breast cancer are more vulnerable to HPV infection than European women but less vulnerable compared to Australian and North American women. One-fourth of Iranian women with BC are infected with HPV coincidentally.  (29,55,56). On the other hand, several studies performed in Denmark, Australia, India, etc. couldn't nd any association between the presence of HPV and breast cancer (32,33,57). There are studies declared that HPV6 (low risk HPV) was the most common type of virus among their specimens (30, 58) since we also found HPV6 in our samples so we recommend to consider low risk HPVs in breast cancers studies. Our result was similar to N Khodabandehlou et al., (59) and JS Lawson et al., (60), which reported HPV18 as the most prevalent type in positive samples. Noticeably, we did not nd HPV type 16, which was demonstrated to be prevalent in previous studies (61,62).
Differences in results may be due to variation in patient age range, geographical distribution of viruses, techniques and the quality of samples used in each study. According to a few published papers, the mean age of breast cancer in Iranian women is between 47.95 to 54.6 years. (63)(64)(65)(66). In our study, the age range of patients was from 29 to 81 years (mean of 50 years) while according to a study in 2008 in Syrian women, the age range was 26 to 66 (with median age of 52) with higher prevalence of HPV (61.06%) (67). Also, in our study, the age range of HPV positive patients were from 33 to 73 years with a viral prevalence of 11.8%, but in a study in Venezuela the age range was 51 to 60 years with a viral prevalence of 41.67% (55). Therefore, the age range can be considered as an important issue in the rate of viral detection.
BC is often labeled as a "welfare disease" Because the incidence rate was reported higher in women in Western societies with higher social class and higher education. Of course, this can be related to timely diagnosis and regular checkups, as well as better training in this group (68).
But in 2017, J Bahk and colleagues conducted age-period-cohort (APC) analysis in Korea. They reported the BC mortality was higher in women with a higher education than in women with no education or a primary education during 1983-1992, and the reverse was true in 1993-2012 (69).
In this study, low level of education found in 56%, and no education in 20% of patients. And in HPV positive patients, 57.1% had low levels of education and 14.2% had no education. So due to our results, most of BC patients and HPV positive patients had low levels of education. Therefore, it can be noted that education, training and socio-economic level of individuals can play a signi cant role in the prevention and timely treatment of breast cancer.
In this study, regarding histologic types, 95% of the total tumor specimens were ductal carcinoma and 5% were lobular carcinoma. And all positive HPV samples were ductal carcinoma. Similarly, in 2019, N Khodabandehlou et al. showed ductal carcinoma as the most frequent type in their HPV positive patients (59).
In addition, in UK base study, NA Salman reported higher prevalence of HPVs in patients with invasive ductal carcinoma (56). From these studies, it can be concluded that mammary ducts provide an entry point for HPV infection (70).
Few studies surveyed the relation of tumor grades and HPV positive breast cancers. As the study of C Kroupis and et al. which recorded high grade tumor in 70.6% of HPV-positive breast cancer cases (61). In our study, all of positive samples had a high grade tumor.
Ni Li and colleagues stated a remarkable point about the type of samples collected for studies. They showed higher prevalence of HPV DNA in FFPE samples than fresh samples (47) According to their ndings the age of specimen also may affect on prevalence of virus detection. Our results indicated that samples collected earlier (from 2018 to 2019) showed higher HPV positivity compared to older collected samples (from 2008 to 2018).
Besides, due to the important role of herpesviruses as a co-factor in the oncogenesis of BC (39,71,72), we decided to detect this virus in our specimens. In 2013, BM Khashman et al. detected HSV-1 Ag with 31.8% prevalence using direct immuno uorescence in FFPE breast tissues (36). We suggest utilizing a different technique, since it may effect results. In the literature there are several studies investigating the relationship between HPV and herpesviruses (73,74). So we used multiplex PCR with different primers for HSV-1, HSV-2, CMV and VZV. We detected only HSV-1 with 8.47% prevalence. Using individual immuno uorescence method for each virus might be more speci c and provide higher accuracy.
These data determine further evidence of a possible role for Human papillomavirus and Herpesvirus in breast carcinogenesis, however additional studies in larger patient population and by techniques with higher speci cities can ascertain our results.

Conclusion
Overall, in the current study we showed the presence of HPV18 and HPV6 genomes in FFPE tissue Samples in women with breast cancers in Tehran (11. 8%). Most of positive samples were at grade II and III. Also HSV-1 was detected in 8.47% of BC samples. And coinfection of HPV and HSV-1 was declared in 3.3% of them. More epidemiological and molecular studies with large samples are needed to validate the association of breast cancer and HPV genotypes, and role of coinfection of HPV and herpesviruses in carcinogenesis of breast cancer.

Declarations
Ethics approval and consent to participate

Availability of data and material
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.     Phylogenetic trees of HPV16 and HPV18 Based on L1 Nucleotide was Constructed Using neighbor joining method and the Kimura 2-Parameter model by MEGA 6 package.