Journal of Surgical Research 103, 203–207 (2002)
doi:10.1006/jsre.2002.6351, available online at http://www.idealibrary.com on
Expression of PH-20 in Normal and Neoplastic Breast Tissue
Derrick J. Beech, M.D., F.A.C.S., Atul K. Madan, M.D., and Nan Deng
Department of Surgery, University of Tennessee–Memphis, Memphis, Tennessee 38163
Submitted for publication June 26, 2001; published online February 13, 2002
Background. Tumor metastasis involves a sequence
of interrelated steps, of which penetration beyond the
basement membrane is an essential component. Hyal-
uronic acid (HA) is a major structural component of
the complex proteoglycans found in extracellular
matrices and basement membranes. Hyaluronidase
(PH-20) degrades HA, resulting in the disruption of
basement membrane integrity and possible tumor dis-
semination.
Materials and methods. Total RNA was extracted from
samples (n ⴝ 51) of normal breast tissue (n ⴝ 12), ductal
carcinoma in situ (DCIS) (n ⴝ 12), infiltrating ductal
breast adenocarcinoma (n ⴝ 13), and metastatic breast
cancer to lymph nodes (n ⴝ 14). RT-PCR was used to
determine the relative level of PH-20 in each specimen.
Results. PH-20 was detected in 41/51 (80.4%) of the
specimens evaluated. PH-20 was present in 12/12
(100%) normal breast tissues; 8/12 (66.7%) DCIS; 13/13
(100%) invasive breast cancers; and 8/14 (57.1%) metas-
tases. Of those specimens in which PH-20 was de-
tected, there were increased levels of PH-20 in meta-
static breast cancer to lymph nodes compared to DCIS
and invasive breast cancer. Stratification of specimen
by race revealed that African American women had
higher levels of PH-20 with invasive and metastatic
beast cancer.
Conclusions. Increased levels of PH-20 are noted in
invasive and metastatic breast cancer compared to
DCIS. Tumors from African American women with inva-
sive and metastatic breast cancer demonstrated higher
levels of PH-20 than Caucasians. Varying levels of PH-20
in mammary tissue may contribute to early invasion and
metastasis of breast cancer. © 2002 Elsevier Science (USA)
Key Words: hyaluronidase; breast cancer; neoplasm;
PH-20.
INTRODUCTION
Breast cancer is the second leading cause of cancer-
related deaths in American women. It is estimated that
203
more than 190,000 women will be diagnosed with
breast cancer in 2001 in the United States [1]. In 2001,
an estimated 40,600 women will die of breast cancer in
the United States [1]. Early detection and improved
understanding of the molecular basis for breast cancer
metastasis may assist in decreasing disease-related
mortality.
Survival is largely dependent on tumor size, axillary
lymph node status, and the presence of metastatic
disease [2]. Although several tumor-related variables
have been found to have prognostic significance (size,
grade, estrogen and progesterone receptor status,
S-phase) [3], there are no tumor-related pathologic
markers that accurately predict the risk of axillary
nodal metastasis or distant disease spread.
Tumor metastasis involves a series of complex cellu-
lar and biochemical events that ultimately result in
disease dissemination [4]. Hyaluronic acid (HA) is a
proteoglycan present in basement membrane and ex-
tracellular matrix. HA is important for maintaining
the integrity of tissue architecture and provides a bar-
rier for tumor invasion. The enzyme hyaluronidase
(PH-20) is necessary for the degradation of HA, allow-
ing penetration of cancer cells through the basement
membrane and entry into systemic circulation and
lymphatics [5]. Also, breakdown products from HA deg-
radation have been associated with an angiogenic re-
sponse and tumor dissemination [6].
Hyaluronidase is a group of enzymes that degrade HA
and can be divided into two general categories according
to optimal pH for enzymatic function. The neutral hyal-
uronidase, found in testicular tissue, is homologous to an
enzyme known as PH-20 and has been implicated in
cancer progression and metastasis. The acidic hyaluron-
idase, typically produced by the liver, is identical to the
enzyme hemopexin and has not been correlated with
cancer progression or metastasis [7].
Our previous investigations have demonstrated a
direct correlation between increased expression of
0022-4804/02 $35.00
© 2002 Elsevier Science (USA)
All rights reserved.
204 JOURNAL OF SURGICAL RESEARCH: VOL. 103, NO. 2, APRIL 2002
FIG. 1. Distribution of specimens. The graph demonstrates the
variety of histological types in our specimens.
PH-20 and invasiveness from prostate cancer [8], la-
ryngeal cancer [9], and breast adenocarcinoma [10].
While elevated levels of hyaluronidase has been previ-
ously demonstrated in small sample studies for breast
cancer with a direct correlation to invasiveness and
metastasis, there has been no large-scale pathologic
study to demonstrate the significance of elevated levels
of PH-20 as a prognostic marker for breast cancer
metastasis. Further delineation of the relation be-
tween PH-20 levels and breast cancer invasiveness or
metastasis will potentially provide a useful tumor-
related marker to predict the risk of axillary nodal
disease or distant disease spread. Furthermore, racial
differences in PH-20 expression might suggest a mo-
lecular marker associated with the current disparity in
breast cancer survival among different ethnic groups.
MATERIALS AND METHODS
Sections of paraffin-embedded and frozen breast tissue samples
(n ⫽ 51) were obtained from the Cooperative Tissue Network. These
specimens included normal breast tissue (n ⫽ 12), ductal carcinoma
in situ (DCIS) (n ⫽ 12), invasive breast adenocarcinoma (n ⫽ 13),
and metastatic adenocarcinoma to axillary lymph nodes (n ⫽ 14)
(Fig. 1). The paraffin sections were deparaffinized with xylene
(Sigma, St. Louis, MO) for incubation at 60°C for 10 min twice. The
sections were then washed for 10 min at room temperature without
absolute ethanol twice to remove the xylene. Specimens were dried
by heating at 65°C for 30 min on a hot block. Tissues were lysed by
incubation at 50°C for 24 h in 200 ␮l of lysis buffer: 20 mM Tris/HCl
(pH 8.0), 20 mM EDTA (pH 8.0), 2% SDS, and 500 ␮g/ml proteinase
K (Fisher, Houston, TX). RNA was extracted by adding the solution
to 0.1 ml of 2 M sodium acetate (pH 4.0) (Sigma), 1 ml of phenol
(water saturated) (Fisher), and 0.2 ml of chloroform/isoamyl alcohol
(49:1) (Fisher). The solution was shaken and cooled in ice for 15 min.
The test tube was centrifuged for 20 min at 10,000g at 4°C in and
Eppendorf centrifuge 541C (Germany). The aqueous phase, which
contained the RNA, was transferred into a 2.5-ml microcentrifuge
tube, mixed with 1 ml 100% isopropanol (Fisher), and placed at
⫺20°C for at least 1 h in order to precipitate the RNA. The solution
was centrifuged at 10,000g for 20 min at 4°C and the supernatant
was discarded. In a 1.5 microcentrifuge tube, the RNA pellet was
dissolved in 0.3 ml denaturing solution: 4 M guanidinium thiocya-
nate, 25 mM sodium citrate (pH 7.0), 0.1 mM 2-mercaptoethanol,
and 0.5% sarkosyl (N-lauroylsarcosine) (Sigma). The RNA was then
precipitated in 0.3 ml of 100% isopropanol (Fisher) for 30 min at
⫺20°C and centrifuged for 10 min at 10,000g. The supernatant was
discarded. The RNA pellet was resuspended in 300 ␮l of 75% ethanol,
vortexed, and incubated 10 to 15 min at room temperature to dis-
solve the amounts of guanidinium contaminating the pellet. The
solution was centrifuged at 10,000g for 5 min and the supernatant
was discarded. The RNA pellet was dried in a vacuum for 5 to 15 min.
The RNA pellet was placed in 100 to 200 ␮l of diethylpyrocarbonate
(DEPC)-treated water (Fisher). The RNA was quantified by diluting
10 ␮l to 1 ml DEPC-treated water and the absorption was read with
a spectrophotometer (Model DU 640 spectrophotometer; Beckman
Instruments, Inc., Sullerton, CA) at an A 260 /A 280 ratio.
For the reverse transcription polymerase chain (RT-PCR) reaction,
2.5 ␮l total RNA and 1 ␮l oligo (dT) primer (100 ng) (Midland
Certified Reagent Company, Midland, TX) were combined to bring
the volume to 12 ␮l and the solution was heated at 60°C for 15 min.
After the mixture was cooled, 5 ␮l 5⫻ RT (reverse transcriptase)
buffer (Fisher), 1.5 ␮l (16u) AMV rt (Fisher), and 5 ␮l 2 mM 4dNTP
mix (Fisher) were added. The mixture was incubated at 42°C for 1 h
and then 150 ␮l 10 mM Tris/HCL, 10 mM EDTA (pH 7.5) and 200 ␮l
phenol were added. The solution was vortexed and then centrifuged
for 5 min at 14,000g at room temperature. The upper aqueous phase
was saved and then 200 ␮l of choloroform/isoamyl alcohol (24:1) was
added. The solution was again vortexed and centrifuged for 5 min at
14,000g at room temperature. The upper aqueous phase was saved
and 20 ␮l of 3 M sodium acetate (pH 5.5) and 500 ␮l of 100% ethanol
were added. The solution was precipitated 15 min at ⫺70°C. The
solution was microcentrifuged at 14,000 cpm at 4°C for 15 min and
the supernatant was discarded. The pellet, which contained the
cDNA, was dried and resuspended in 40 ␮l of water.
For the PCR reaction, three primers, which corresponded to PH-20
mRNA, were used for each sample: (i) 5⬘CCA-TGT-TGC-TTG-ACT-
CTG-AAT-TTC-A 3⬘, (ii) 5⬘CCA-GAA-GAT-TTC-CTT-ACA-AGA-CC
3⬘, and (iii) 5⬘CCG-AAC-TCG-ATT-GCG-CAC-ATA-GAG-T 3⬘. All
primers were synthesized from Midland Certified Reagent Company.
The first PCR reaction used primers i and iii. Five microliters of the
diluted cDNA was combined with 1 ␮l 3⬘ primer (50 ng), 1 ␮l 5⬘
primer (50 ng), 5 ␮l 10⫻ Taq buffer, 0.5 1 Taq polymerase, and 4 ␮l
5 mM 4 dNTP mix (Fisher) to a total volume of 50 ␮l. The tube was
covered with 50 ␮l of mineral oil (Aldrich Chemical Co., Inc., Mil-
waukee, WI) and then heated to 94°C for 5 min. The PCR was carried
out for 35 cycles: 94°C for 1 min, 63°C for 1 min, and 72°C for 10 min
(Thermolye, Barnstead/Thermolyne Corp., Dubuque, IA). The prod-
ucts were run on a 1% agarose gel at 200 V for 30 min (Model H⬘5
Horizontal Gel Electrophoresis Apparatus; Life Technologies, Inc.,
Grand Island, NY) and visualized with 1% ethidium bromide (Sigma)
stain. The expected PCR product was a 759-bp sequence. PCR was
performed again using primers ii and iii. The expected PCR product
was a 504-bp sequence. A photograph of the PCR products was
scanned on a densitometer (Fig. 2). Densities of the PCR product
were compared to densities of ␤-actin as a housekeeping gene. Ratios
of the densities were compared by ANOVA and two-tailed unpaired
t test when appropriate with GraphPAD INSTAT version 1.12a.
RESULTS
Our investigation evaluated 12 normal breast spec-
imens, 12 patient samples of ductal carcinoma in situ,
13 invasive breast adenocarcinomas, and 14 specimens
of metastatic adenocarcinoma to axillary lymph nodes.
PH-20 levels (1.06) in nonneoplastic breast tissue were
increased compared to neoplastic breast tissue, which
demonstrated an overall value of 0.69 (P ⬍ 0.03).
Nonneoplastic breast tissue from patients with a
known diagnosis of breast cancer (NLBC; n ⫽ 7) re-
sulted in a PH-20 level of 1.12 compared to 0.86 (P ⬍
 BEECH, MADAN, AND DENG: PH-20 IN BREAST TISSUE
FIG. 2. Photograph of the RT-PCR results, demonstrating the
expected bands of PH-20 at 759 bp.
0.8) for patients without a cancer diagnosis (NLNC;
n ⫽ 5) (Fig. 3). When normal breast tissue (n ⫽ 12)
was compared to DCIS (n ⫽ 8), primary invasive ad-
enocarcinoma (n ⫽ 13), and lymph node metastasis
(n ⫽ 8), the levels of PH-20 were 0.65, 0.61, and 0.91,
respectively (Table 1). Hence, PH-20 levels were
greater in metastatic adenocarcinoma to the axillary
lymph nodes compared to DCIS and invasive adenocar-
cinoma (P ⬍ 0.06). Similar levels of PH-20 expressions
were evident in DCIS and invasive breast cancer spec-
imens (Fig. 4). When the PH-20 level was compared by
race (Fig. 5), there was a trend of increased hyaluron-
idase levels in invasive (P ⬍ 0.5) and metastatic (P ⬍
0.3) breast cancer for African American compared to
Caucasian women.
DISCUSSION
There has been a substantial increase in the inci-
dence of breast adenocarcinoma over the past several
years with a trend toward earlier stage presentation
[11]. There has not, however, been a significant de-
205
FIG. 3. PH-20 levels in normal breast tissue. Normal breast
tissue data include specimens from patients with breast cancer
(NLBC) and patients with no evidence of breast cancer (NLNC).
*P ⬍ 0.8 compared to NLNC.
crease in associated mortality rates resulting from
breast cancer [11].
Prognostic factors critical to the therapeutic deci-
sions regarding breast adenocarcinoma are the size
and status of the regional lymph node basin. Generally,
patients with tumors less than 2 cm in diameter have
an overall 5-year survival of approximately 90%, while
those with tumors having diameters of 2 to 4.9 cm have
approximately 80% overall survival at 5 years [2]. Pa-
tients with tumors that exceed 5 cm in diameter have
an associated survival of less than 60% at 5 years [2].
The size of the primary lesion is directly correlated
with the status of the lymph node basin. As the size of
the tumor increases, the number of affected lymph
nodes also increases. Although size has been found to
directly correlate with the rate of lymph node spread
for breast adenocarcinoma, there are no other primary
tumor-related prognostic factors to predict lymph node
metastasis or distant disease spread. As such, the focus
on a molecular marker associated with the primary
tumor capable of predicting the risk of metastatic dis-
ease is paramount.
Tumor metastasis involves a sequence of interre-
lated events which ultimately lead to tumor dissemi-
nation to distant sites. While there are several steps
involved in the metastatic cascade, two vital compo-
TABLE 1
PH-20 Levels by Histological Type
Specimen PH-20 level*
Normal 1.06
DCIS 0.65
Invasive 0.61
Metastasis 0.91
* P ⬍ 0.06.
206 JOURNAL OF SURGICAL RESEARCH: VOL. 103, NO. 2, APRIL 2002
FIG. 4. PH-20 levels. The graph displays PH-20 levels in normal
breast tissue, DCIS, invasive breast cancer, and metastatic breast
cancer. There are statistical significantly higher PH-20 levels in
normal breast tissue compared to neoplastic tissue. *P ⬍ 0.06.
nents involve the enzymatic degradation of the base-
ment membrane and the process of angiogenesis. Base-
ment membrane disruption along with associated
degradation of the extracellular matrix provides access
to vascular channels for the tumor cell. PH-20 plays a
role in this degradation process along with the associ-
ated angiogenesis related to the degradation products
of hyaluronic acid.
Hyaluronic acid is a nonsulfated glycosaminoglycan
which is composed of repeated disaccharide units,
D-glucuronic acid and N-acetyl D-glucosamine. It is a
key component of the extracellular matrix as well as
being found in various body fluids and connective tis-
sue. Hyaluronic acid appears to play a critical role in
maintaining matrix integrity as well as influencing
some migration and proliferation [12]. Hyaluronic acid,
when associated with the cell surface receptor CD-44,
is intimately involved in the process of cell migration.
West and associates demonstrated the degradation of
hyaluronic acid. Hyaluronic acid produces approxi-
mately 3–25 disaccharide fragments better known to
be angiogenic, thus producing a heightened level of
actual access for distant disease dissemination [6].
Furthermore, certain isoforms of the CD-44 receptor
expressed on tumor cells aid in the attachment of hy-
aluronic acid, stimulating migration and proliferation
of the tumors cells [13]. Again, this is also associated
with the angiogenic effects of the breakdown products
of hyaluronic acid. Thus, elevated levels of hyaluroni-
dase along with the expressions of certain forms of
C-44 receptor may play a role in the process of disease
dissemination, particularly in patients with high-risk
breast cancer [14].
Our investigation expanded the previous study eval-
uating six specimens containing breast neoplasia [10].
We explored the significance of PH-20 in the metastatic
cascade for 52 patients. Forty patients had ductal car-
cinoma in situ, invasive breast cancer, or metastatic
adenocarcinoma to the axillary lymph nodes. The other
12 patients with normal breast tissue were primarily
those undergoing breast reduction surgery for nonma-
lignant causes. PH-20 levels were found to be increased
in normal breast tissue compared to neoplastic breast
disease. When comparing subgroups of these patients,
interesting trends were found although the results
were not statistically significant. Of course, increasing
the sample size may help demonstrate statistically sig-
nificant results. The data suggest that metastatic ad-
enocarcinomas have higher levels of PH-20 compared
to other neoplastic breast disease. Interestingly, only
57% of specimens with metastatic adenocarcinoma
demonstrated PH-20, while 100% of normal and inva-
sive breast specimens demonstrated PH-20. Also, the
levels of PH-20 expression were decreased in DCIS,
invasive adenocarcinoma, and metastatic adenocarci-
noma compared to normal breast tissue. Since we be-
lieve that hyaluronidase may be necessary in tumor
metastasis, the disparity of these observations can be
explained by several hypotheses. One hypothesis is
that cells that have already metastasized may have
“lost” the need and ability to express PH-20. Also, there
may be regulation of PH-20 via a negative feedback
mechanism. Since RT-PCR only measures RNA pro-
duction, not the amount of protein, the cells in meta-
static adenocarcinoma may have decreased the produc-
tion of hyaluronidase RNA due to the high level of
hyaluronidase protein. Also, as shown in Fig. 4, PH-20
can vary with ethnicity in various stages of breast
cancer development. Ethnicity may account for the
differences noted in our observations. Further investi-
gations must focus on these possible hypotheses.
This evaluation was able to include a diverse groups
of patients with 57% Caucasian patients, 28% African
American patients, 3% Hispanic patients, and 6%
Asian American patients. The total RNA extraction
and RT-PCR were standardized and performed with-
out difficulty. Our evaluation demonstrated increased
FIG. 5. PH-20 levels by ethnicity and histologic types. The graph
demonstrates a decrease in PH-20 levels as metastatic potential
increases in Caucasian women, but an increase in PH-20 levels as
metastatic potential increases in African American women. *P ⬍
0.5; **P ⬍ 0.3.
 BEECH, MADAN, AND DENG: PH-20 IN BREAST TISSUE
expression of PH-20 in both malignant and benign
breast tissue. An interesting finding was the associa-
tion of PH-20 level of expression with race. African
American women demonstrated a higher level of PH-20
for invasive adenocarcinoma of the breast and meta-
static axillary lymph node disease. It also seems that in
Caucasian women PH-20 levels decreased as the met-
astatic potential increased, while African American
women have increased levels as the metastatic poten-
tial increases. This is a highly relevant finding consid-
ering that there is a paucity of data linking molecular
and genetic factors to survival and breast cancer out-
come.
These data suggest that PH-20 serves as a molecular
marker for outcome and perhaps suggest the founda-
tion for current disparity and survival based on race.
Further investigation is warranted to explore these
preliminary findings and target the potential for
hyaluronidase-positive breast cancer with regard to
outcomes and metastasis. Long-term follow up is re-
quired with primary patient data to assess the true
significance of these preliminary discoveries. Future
investigations may lead to the development of a new
cancer marker and possibly a new target for breast
cancer therapy.
REFERENCES
1. Greenlee, R. T., Hill-Harmon, M. B., Murray, T, and Thun, M.
Cancer statistics, 2001. CA Cancer. J. Clin. 51: 15, 2001.
2. Carter, C., Allen, C., and Henson, D. Relation of tumor size,
lymph node status, and survival in 24,740 breast cancer cases.
Cancer 63: 181, 1989.
207
3. Mustafa, I., Cole, B., Wanebo, H., Bland, K. I., and Chang, H. R.
The impact of histopathology and nodal metastases in minimal
breast cancer. Arch. Surg. 132: 384, 1997.
4. Zetter, B. The cellular basis of site-specific metastases. N. Engl.
J. Med. 322: 605, 1990.
5. Fraser, J., and Laurent, T. Turnover and metabolism of hyalu-
ronan. In J. Whelon (Ed.), The Biology of Hyaluronan. New
York: Wiley, Ciba Foundation Symposium, 1989. Pp. 41–59.
6. West, D., Hampson, I., Arnold, F., and Kumar, S. Angiogenesis
induced by degradation products in hyaluronic acid. Science
228: 1324, 1985.
7. Lui, D., Pearlman, E., Diaconu, E., Guo, K., Mori, H., Haqqi, T.,
Markowitz, S., Willson, J., and Sy, M. S. Expression of hyal-
uronidase by tumor cells induces angiogenesis. Proc. Natl.
Acad. Sci. USA 93: 7832, 1996.
8. Madan, A. K., Pang, Y., Wilkiemeyer, M. B., Yu, D., and Beech,
D. J. Increased hyaluronidase expression in more aggressive
prostate adenocarcinoma. Oncol. Rep. 6: 1431, 1999.
9. Godin, D. A., Fitzpatrick, P. C., Scandurro, A. B., Belafsky,
P. C., Woodworth, B. A., Amedee, R. G., Beech, D. J., and
Beckman, B. S. PH20: A novel tumor marker for laryngeal
cancer. Arh. Otolaryngol. Head Neck Surg. 126(3): 402, 2000.
10. Madan, A. K., Beech, D. J., Yu, K., Dhurandhar, N., Cullinane,
C., and Pang, Y. Association of hyaluronidase and breast ade-
nocarcinoma invasiveness. Oncol. Rep. 6: 607, 1999.
11. Bal, D. G. Cancer statistics 2001: Quo vadis or whither goest
thou? CA Cancer J. Clin. 51: 11, 2001.
12. Stetler-Stevenson, W.,Aznavoorian, S., and Liotta, L. Tumor
cell interactions with the extracellular matrix during invasion
of metastasis. Ann. Rev. Cell Biol. 9: 541, 1993.
13. Thomas, L., Etoh, T. Stamenkovic, S., Mihm, M., and Byers, H.
Migration of human melanoma cells on hyaluronate is related
to CD-44 expression. J. Invest. Dermatol. 100: 115, 1993.
14. Sy, M., Guo, Y., and Stomenkovic, I. Distinct affects of two
CD44 isoforms on tumor growth in vivo. J. Exp. Med. 174: 859,
1991.