Full text of DOI:10.1359/jbmr.2001.16.3.478

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 3, 2001
© 2001 American Society for Bone and Mineral Research
Human Osteoclast Cathepsin K Is Processed Intracellularly
Prior to Attachment and Bone Resorption
ROBERT A. DODDS,1 IAN E. JAMES,1 DAVID RIEMAN,1 REINA AHERN,1 SHING MEI HWANG,1
JANICE R. CONNOR,1 SCOTT D. THOMPSON,2 DANIEL F. VEBER,2 FRED H. DRAKE,1
STEPHEN HOLMES,3 MICHAEL W. LARK,1 and MAXINE GOWEN1
ABSTRACT
Cathepsin K is a member of the papain superfamily of cysteine proteases and has been proposed to play a
pivotal role in osteoclast-mediated bone resorption. We have developed a sensitive cytochemical assay to
localize and quantify osteoclast cathepsin K activity in sections of osteoclastoma and human bone. In tissue
sections, osteoclasts that are distant from bone express high levels of cathepsin K messenger RNA (mRNA) and
protein. However, the majority of the cathepsin K in these cells is in an inactive zymogen form, as assessed
using both the cytochemical assay and specific immunostaining. In contrast, osteoclasts that are closer to bone
contain high levels of immunoreactive mature cathepsin K that codistributes with enzyme activity in a
polarized fashion toward the bone surface. Polarization of active enzyme was clearly evident in osteoclasts in
the vicinity of bone. The osteoclasts apposed to the bone surface were almost exclusively expressing the mature
form of cathepsin K. These cells showed intense enzyme activity, which was polarized at the ruffled border.
These results suggest that the in vivo activation of cathepsin K occurs intracellularly, before secretion into the
resorption lacunae and the onset of bone resorption. The processing of procathepsin K to mature cathepsin K
occurs as the osteoclast approaches bone, suggesting that local factors may regulate this process. (J Bone
Miner Res 2001;16:478–486)
Key words: cathepsin K processing, osteoclast, cytochemistry, ruffled border
Newly synthesized lysosomal enzymes, packaged into trans-
port vesicles, are transported to this membrane, where they are
secreted into the underlying compartment. Two reports have
shown that osteoclasts in vitro remove the organic and inor-
ganic degradation products by endocytosis at the ruffled border
membrane, with subsequent transcytosis (vesicular) through
the cell to the apposing free membrane, and liberation into the
extracellular space.(6,7) This process is inhibited by a non-
selective cysteine protease inhibitor trans-epoxysuccinyl-l
leucylamido-(4-guanidino) butane (E64).(7)
INTRODUCTION
STEOCLASTS ARE specialized multinucleated cells that
resorb bone. The resorbing osteoclast creates an extracel-
O
lular compartment, which it maintains at an acid pH via a
vacuolar Hϩ-adenosine triphosphatase (ATPase) and into
which it secretes hydrolytic enzymes.(1–4) The acidity of the
compartment results in demineralization of the matrix and
provides an optimal pH for certain proteolytic enzymes,(1,3,5)
which ultimately degrade the demineralized matrix. An inten-
sively convoluted membrane, the ruffled border,(2) lines the
Lysosomal cysteine protease(s) plays a key role in bone
resorption lacuna that forms beneath a recruited osteoclast. matrix resorption.(8–12) An osteoclast selective cysteine pro-
1Department of Bone and Cartilage Biology, SmithKline Beecham Pharmaceuticals, King of Prussia, Pennsylvania, USA.
2Department of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals, King of Prussia, Pennsylvania, USA.
3Department of Molecular and Cellular Immunology, SmithKline Beecham Pharmaceuticals, Harlow, United Kingdom.
478

CATHEPSIN K IS PROCESSED INTRACELLULARLY PRIOR TO BONE RESORPTION
479
tease cathepsin K has been identified(13–18) that has high
sequence homology to cathepsins S and L. In situ hybrid-
ization, Northern blot, and immunolocalization studies in-
dicated that human cathepsin K was expressed abundantly
in osteoclasts but that cathepsins B, L, and S were expressed
at relatively low or negligible levels.(13,14,16,19)
Cathepsin K has a low pH optimum and degrades many
bone matrix proteins including type I collagen, osteopontin,
FIG. 1. Structure of SB 240314.
and osteonectin.(20–22) Selective human cathepsin K inhib-
itors have been described that potently inhibit osteoclast
resorption both in vitro and in vivo.(23–25) Patients with
pycnodysostosis, a disease characterized by abnormal bone
remodeling,(26–30) have mutations in the cathepsin K
gene.(31,32) Furthermore, mice with a null mutation in the
cathepsin K gene develop osteopetrosis of the long bones
and vertebrae.(33,34) Histological evaluation of resorption
sites from the bones of Ϫ/Ϫ mice revealed fully differen-
tiated osteoclasts apposed to small regions of demineralized
bone, suggesting that the cathepsin K–deficient osteoclasts
are capable of demineralizing the extracellular matrix but
are unable to degrade adequately the demineralized ma-
trix.(34)
Purified cathepsin enzyme studies, using synthetic fluoro-
genic substrates,(20,35) have shown that benzyloxycarbonyl-
Phe-Arg-(4-methyl)coumaryl-amide (Cbz-Phe-Arg-MCA)
FIG. 2. Measurement of cathepsin activity with time of reaction.
Assays were conducted at pH 5.5 using Z-Leu-Arg-4M␤NA (15 mM)
as substrate. Activity is expressed as MIE ϫ 100 Ϯ SEM. Osteoclasts
within human osteoclastoma section (open diamonds) and osteoclasts
within fetal human bone sections (closed diamonds) were evaluated.
No activity was recorded in fetal bone osteoclasts with the cathepsin B
substrates Z-Arg-Arg-4M␤NA (dotted line) and Z-Ala-Arg-Arg-
4M␤NA (hatched line). The cathepsin B assays were carried out at 12
mM substrate concentration at pH 6.0.
and Cbz-Leu-Arg-MCA are the favored substrates for both
cathepsin K and cathepsin L. In contrast, cathepsin K does
not cleave the cathepsin B selective substrates Cbz-Arg-
Arg-MCA and Cbz-Ala-Arg-Arg-MCA.(20,35) Cytochemical
analysis of the localization and activity of cysteine proteases
in tissue sections are possible using similar synthetic sub-
strates that contain the leaving group 4-methoxy-2-
naphthylamide (4M␤NA).(36) During the assay, performed
on cryostat sections, the cleaved naphthalamine is subse-
quently postcoupled with Fast Blue BB to form a quantifi-
able red precipitate within the cell. The modified quantita-
tive cytochemical assay described here uses the cathepsin K
substrates, Cbz-Phe-Arg-4M␤NA, and Cbz-Leu-Arg-
4M␤NA, and comparison was made using the cathepsin B
substrates Cbz-Arg-Arg-4M␤NA and Cbz-Ala-Arg-Arg-
4M␤NA. Cryostat sections of human fetal bone (tibia, 19-
week gestation) and osteoclastoma were used to identify
and characterize cathepsin K activity. Fetal bone contains
abundant resorbing osteoclasts at the site of endochondral
bone formation; however, nonresorbing osteoclasts are rare.
In contrast, osteoclastoma tissue contains abundant non-
resorbing osteoclasts distant from populations adjacent to or
attached to membranous bone, representing an ideal tissue
to investigate lysosomal cathepsin processing.
at the time of surgery and frozen as described previously.(37)
Adult kidney samples were obtained from the Anatomical
Gift Foundation (Laurel, MD, USA). Multiple serial sec-
tions from each tissue block (5 ␮m) were cut on a Hacker
cryostat (Hacker Instruments, Inc., Fairfield, NJ, USA),
equipped with a finely polished tungsten-tipped steel knife,
and flash-dried onto glass slides. Serial sections were pro-
cessed for enzyme cytochemistry (cysteine protease), im-
munolocalization, and in situ hybridization as described in
the following section. For in situ hybridization studies,
TESPA-coated (3-aminopropyltriethoxy silane; Sigma
Chemical Co., St. Louis, MO, USA) glass slides were used.
The present study was designed to investigate whether
cathepsin K in situ is secreted into the osteoclast resorption
lacunae in its inactive (proform) or mature form.
Cysteine protease activity
Cryostat sections of whole tissue were assayed for pro-
tease activity by a suitably modified azo coupling procedure
derived from the cathepsin B assay described by Van Noor-
den et al.(36) For the detection of cathepsin K, sections were
incubated at 37°C for 10–15 minutes in the following
reaction medium (total volume, 12 ml): 5 ml encapsin HPB
MATERIALS AND METHODS
Tissue processing
Human fetal bone and kidney (Anatomical Gift Founda- (Hammond, Cerestar, IN, USA) added to 6 ml of 40%
tion, Laurel, MD, USA) and osteoclastoma tissue (Jefferson polypeptide (dissolved in 100 mM phosphate buffer, pH
Hospital, PA, USA) were obtained (with informed consent) 5.3; Sigma), the cathepsin K substrate (dissolved in 1 ml

480
DODDS ET AL.
FIG. 3. (A) The effect of increasing concen-
trations of Z-Phe-Arg-4M␤NA on the cathepsin
activity of osteoclasts within sections of human
osteoclastoma (closed diamonds) and proximal
convoluted tubule cells within sections of human
kidney (closed squares). Assays were conducted
at pH 5.5 for 15 minutes. (B and C) The effect of
increasing concentrations of Z-Leu-Arg-4M␤NA
on the activity of (B) osteoclastoma and (C) fetal
osteoclast cathepsin activity; assays were con-
ducted at pH 5.5 for 10 minutes. Activity is
expressed as MIE ϫ 100 Ϯ SEM.
DMSO), Cbz (Z)-Leu-Arg-4M␤NA (15 mM) or Z-Phe-
Arg-4M␤NA (12 mM; Bachem, King of Prussia, PA, USA),
and 2.5 mM EDTA. Subsequently, the sections were post-
coupled at room temperature for 10 minutes with 0.25
mg/ml Fast Blue BB (Sigma) dissolved in phosphate-
buffered saline (PBS). The sections were then rinsed in
PBS. The sections were finally rinsed with 100 mM copper
sulfate for 10 minutes at room temperature. The final reac-
tion product is a deep red precipitate. Two cathepsin B
substrates also were profiled: Z-Arg-Arg 4M␤NA and
Z-Ala-Arg-Arg-4M␤NA (Bachem).
Specificity was assessed by using the nonselective cys-
teine protease inhibitor trans-epoxysuccinyl-l leucylamido-
(4-guanidino) butane (E64; Sigma), the aspartate protease
inhibitor pepstatin A (Sigma), the serine protease inhibitor
phenylmethylsulfonyl fluoride (PMSF; Sigma), and a selec-
tive irreversible inhibitor of cathepsin K SB 240314 (Fig.
1).(24) For SB 240314, the kobs/[I], MϪ1 sϪ1 for inhibition of
cathepsin K, B, and L are 3.1 ϫ 106, 1.3 ϫ 103, and 5.8 ϫ
104, respectively; the Ki for cathepsin S was 11 nM (i.e.,
Ͼ1000-fold selective for cathepsin K vs. cathepsin B, L,
and S).
The reaction product was measured on a per cell basis
using a Vickers M85 scanning and integrating microcyto-
photometer,(38) with a mask that encompasses a single cell
(ensuring that the optical density of a unit area of reaction
product was measured in each cell), a ϫ40 objective at 550
nm, and with a flying spot of a 0.5-␮m diameter in the plane
of the section. The individual flying spot measurements of
“relative extinction” was integrated by the microcytopho-
tometer and summed to give the relative integrated extinc-
tion or absorption over the area scanned. By suitable cali-
bration (with a filter of known absolute absorbance), the
absolute mean integrated extinction (MIE) was determined
for each cell. Thirty cells were measured in triplicate sets of
human osteoclastoma (osteoclasts), 19-week gestation fetal
femoral bone (osteoclasts), and kidney (proximal convo-
luted tubule cells). Randomized measurements are recorded
blind at distinct sites for each section. Results are presented
as MIE ϫ 100 Ϯ SEM or expressed as percent of control.
FIG. 4. Measurement of optimum pH. The assay was conducted on
sections of fetal tibia using 15 mM Z-Leu-Arg-4M␤NA for 10 minutes.
Activity is expressed as MIE ϫ 100 Ϯ SEM.
cathepsin K) is described elsewhere.(39,40) 10C7 shows no
cross-reactivity with procathepsin K or cathepsins L, B, D,
or S. A panel of soft tissues (cryostat sections) known to
have abundant levels of cathepsin B, L, or S (e.g., liver,
skin, muscle, and lung macrophages) also were stained with
10C7 and CE5 and no expression was observed. For immu-
nolocalization, an alkaline phosphatase-based immunoenzy-
matic technique was used according to the manufacturers
protocol (Dako, Carpenteria, CA, USA). A new fuchsin-
based substrate system was used to localize antibody bind-
ing and a red precipitate indicates positive reactivity. No ex-
pression was observed in isotype-matched negative controls.
Immunolocalization
In situ hybridization
The generation and characterization of the murine mono-
A complementary DNA (cDNA) clone (pBluescript SK)
clonal antibodies CE5 (procathepsin K) and 10C7 (mature containing the coding region of human cathepsin K was

CATHEPSIN K IS PROCESSED INTRACELLULARLY PRIOR TO BONE RESORPTION
481
FIG. 5. Inhibition of osteoclast cathepsin ac-
tivity by SB 240314. (A) Assays were conducted
with Z-Phe-Arg-4M␤NA at pH 5.5 for 15 min-
utes on sections of osteoclastoma (diamonds)
and adult kidney (triangles). SB 240314 had no
effect on cathepsin activity in kidney proximal
convoluted tubule cells (cathepsin L). Results
presented as percent of control Ϯ SEM. (B)
Assay conducted with Z-Leu-Arg-4M␤NA as
substrate on sections of fetal tibia. Results are
presented as percent of control Ϯ SEM.
hybridization. In situ hybridization was performed as pre-
viously described.(37)
RESULTS
To validate true enzyme activity cytochemically, the as-
say should show substrate dependancy, linearity with time
of reaction, and the appropriate pH optimum of the enzyme.
Optimization (to attain maximal activity) and confirmation
of cathepsin activity was performed on cryostat sections of
human osteoclastoma tissue (containing zones of intramem-
branous bone remodeling), fetal bone (19-week gestation),
and adult (70 years old, diseased) and fetal kidney (19-week
gestation); kidney proximal convoluted tubule cells contain
no cathepsin K but do contain abundant cathepsin B and L
expression.(13,37) The fetal kidney blocks were in excellent
condition compared with the adult kidney (rejected for
transplant purposes) and were thus used for the selective
versus nonselective inhibitor studies.
FIG. 6. Effects on nonselective protease inhibitors on cathepsin
activity in sections of fetal bone and fetal kidney proximal convoluted
cells (PCT). Pepstatin (10 ␮M, aspartyl protease inhibitor) and PMSF
(10 ␮M, serine protease inhibitor) had no effect on cathepsin activity in
fetal bone osteoclasts. In contrast SB 240314 (1 ␮M) and E64 (1 ␮M)
potently inhibited osteoclast cathepsin activity. In contrast to E64, SB
240314 had no effect on fetal kidney cell cathepsin activity. Assays
were conducted with Z-Leu-Arg-4M␤NA at pH 5.5 for 10 minutes.
Results are presented as percent of control Ϯ SEM.
Substrate specificity, linearity with time, and pH profile
For each experiment, representative data are presented
from one of four experiments. The selective cathepsin K
cytochemical substrates Z-Phe-Arg-4M␤NA and Z-Leu-
Arg-4M␤NA were excellent substrates for osteoclast ca-
obtained from an osteoclast library. cDNA templates were thepsin activity. The activity was linear with time of reac-
linearized and then transcribed from the T3 or the T7 tion up to 20 minutes (Fig. 2), and the pattern of activity was
promoter to generate sense and antisense probes, respec- comparable with either substrate. No activity was observed
tively. Riboprobes were prepared using the Promega (Mad- in osteoclastoma osteoclasts (not shown) or fetal human
ison, WI, USA) in vitro transcription kit with [35S]thiocy- osteoclasts with the cathepsin B substrates Z-Arg-Arg-
tosine triphosphate (CTP; Amersham, Arlington Heights, 4M␤NA or Z-Ala-Arg-Arg-4M␤NA (Fig. 2), consistent
IL, USA). After transcription, cDNA templates were di- with the negligible expression of cathepsin B messenger
gested with RQ1 RNAse-free DNAse I (Promega) and RNA (mRNA) and protein.(13) Chondrocytes and macro-
unincorporated nucleotides were removed by centrifugation phages showed cathepsin activity in sections of fetal bone
through Quick Spin Sephadex G-50 columns (Boehringer- using these cathepsin B substrates (data not shown). Oste-
Mannheim, Indianapolis, IN, USA). RNA transcripts with a oclast cathepsin activity was dose dependent with both
specific activity in excess of 108 cpm/mg were used for substrates (Fig. 3). Proximal convoluted tubule cells in

482
DODDS ET AL.
FIG. 7. Expression of pro- and
mature cathepsin K, cathepsin K
mRNA, and cathepsin activity in
serial sections of osteoclastoma.
(A) Section hybridized with the
cathepsin
K
antisense probe
shows osteoclasts (arrows) with
intense and selective levels of ca-
thepsin K mRNA. Counterstained
with methylene blue. (B) Serial
section reacted with the procathep-
sin K monoclonal antibody (CE5).
This population of osteoclasts
shows intense levels of procathep-
sin K (arrows). (C) Serial section
reacted with the mature cathepsin
K monoclonal antibody (10C7).
This population of osteoclasts
shows no mature cathepsin K (ar-
rows). (D) Serial section reacted
for cathepsin activity using the
substrate Z-Phe-Arg-4M␤NA. (E)
Serial section reacted for cathepsin
activity using the substrate Z-Leu-
Arg-4M␤NA. (F) Isotype matched
negative control for panel B (bar ϭ
80 ␮m).
sections of human kidney also showed cathepsin activity taining or not containing discrete sites of bone formation
(cathepsin L) with Z-Phe-Arg-4M␤NA (Fig. 3) and Z-Leu- and remodeling) were probed with specific antibodies to
Arg-4M␤NA (not shown).
either procathepsin K, mature cathepsin K, reacted for ca-
For both substrates (Z-Leu-Arg-4M␤NA shown as repre- thepsin activity, or hybridized with a cathepsin K riboprobe
sentative), the osteoclast cathepsin activity was maximal in to localize sites of cathepsin K mRNA expression. Multiple
the acidic pH range from 5.0 to 5.5 (Fig. 4), consistent with sections were examined from the different osteoclastoma
the published pH optimum of isolated cathepsin K.
blocks. The representative sites highlighted in Figs. 7 and 8
were selected from the same section for each probe/activity
described previously. Osteoclasts that were distant from
bone (Fig. 7) showed intense expression of cathepsin K
mRNA (Fig. 7A, homogeneously localized), contained pro-
cathepsin K (Fig. 7B, CE5 positive) but negligible levels of
mature cathepsin K (Fig 7C, 10C7 negative). This mature
cathepsin K–negative population of cells also did not show
any cathepsin activity (Figs. 7C and 7D). Equivalent results
were shown in osteoclastoma tissues that contained no
zones of intramembranous ossification (data not shown).
Confirmation of cysteine protease activity
Specificity was addressed by the use of the nonselective
cysteine protease inhibitor E64, the aspartate protease in-
hibitor pepstatin A, the serine protease inhibitor PMSF, and
a selective irreversible inhibitor of cathepsin K SB 240314.
SB 240314 (Fig. 5) and E64 (Fig. 6) potently inhibited
osteoclast cathepsin activity in both osteoclastoma and fetal
bone sections. Pepstatin A (10 ␮M) and PMSF (10 ␮M)
were inactive (Fig. 6). SB 240314 did not inhibit cathepsin
activity in proximal convoluted kidney cells (Figs. 5 and 6).
In contrast, the nonselective cysteine protease inhibitor E64
potently inhibited kidney cathepsin activity (Fig. 6). These
data suggest that the cathepsin activity in the kidney is
cathepsin L.
Cathepsin K processing occurs before osteoclast
attachment and bone resorption and correlates
with enzyme activity detected cytochemically
Sites of osteoclastoma tissue were selected in which
osteoclasts were adjacent to zones of intramembranous os-
sification (osteoclasts apposed to bone were evident but rare
at this site). Osteoclasts that were closest to zones of ossi-
fication in the osteoclastoma tissue showed low expression
Nonresorbing osteoclasts express cathepsin K mRNA
and procathepsin K protein but do not show
cathepsin activity
Serial cryostat sections of human osteoclastoma tissue of the proform of the enzyme (CE5 negative, arrows in Fig.
(blocks from 5 patient samples were selected, either con- 8A), and high expression of the mature form (10C7 positive,

CATHEPSIN K IS PROCESSED INTRACELLULARLY PRIOR TO BONE RESORPTION
483
FIG. 8. Section of osteoclas-
toma that contains zones of in-
tramembranous ossification; se-
rial sections were reacted with an
antibody to (A) procathepsin
(CE5), (B) mature cathepsin K
(10C7), and (C) reacted for ca-
thepsin activity (Z-Phe-Arg-
4M␤NA). Cathepsin K process-
ing (arrows) clearly occurs before
osteoclast attachment and resorp-
tion. (D) Isotype matched nega-
tive control for panel B. (E) An
osteoclast (fetal bone) situated
between bone (arrowheads) al-
ready shows polarized enzyme
activity (Z-Phe-Arg-4M␤NA).
(F) Serial section to panel C. Ca-
thepsin K mRNA expression in
osteoclasts (arrows) adjacent to
bone. Osteoclasts distant from
bone were smaller and are indi-
cated by arrowheads. Section
counterstained with methylene
blue (bar ϭ 100 ␮m).
arrows at top of Fig. 8B). The expression of mature cathep- ment.(4) Ultrastructural studies have shown that osteoclasts
sin K directly correlated with enzyme activity detected do not contain high concentrations of phagocytic structures
cytochemically (Fig. 8C). Polarization of active enzyme such as secondary lysosomes. Instead, hydrolytic enzymes
are found in elements of the exocytic pathway(41) reflecting
was evident in osteoclasts near bone in osteoclastoma (Fig.
8C, arrows) and fetal bone (Fig. 8E). Although the polarity
of cathepsin K activity in Fig. 8C (middle arrow) appears
opposite to the nearby bone (top of figure), this could be
explained by polarity toward bone deeper (and thus hidden)
within the block of osteoclastoma tissue. Cathepsin K
mRNA expression appeared highest in the larger osteoclasts
adjacent to bone (Fig. 8F).
At sites of bone remodeling in sections of osteoclastoma
(Figs. 9A–9C) osteoclasts resorbing bone almost exclu-
sively expressed the mature form of cathepsin K that colo-
calized with intense enzyme activity polarized at the
osteoclast-bone interface. In contrast to nonresorbing oste-
oclasts, cathepsin K mRNA expression was highly variable
in the resorbing osteoclast population, ranging from a neg-
ligible to a moderate signal (Fig. 9D). In sections of fetal
bone reacted for cathepsin K activity, high power views of
osteoclast resorption sites revealed a distinct pattern of
polarization (Figs. 9E and 9F).
a high biosynthetic activity. Cysteine protease(s) plays a key
role in bone matrix degradation.(8–12) A number of groups
have identified a novel cysteine protease, cathepsin K, that
appears to be selective for the osteoclast and plays a pivotal
role in bone matrix degradation.(13,34) Localization studies
showed that although human cathepsin K was expressed
abundantly in osteoclasts, cathepsins B, L, and S were
expressed at very low levels or were not detect-
able.(13,14,16,19) Furthermore, transgenic mice lacking ca-
thepsin L activity had a normal bone phenotype.(42) Cyto-
chemical analysis of cathepsin activity in sections of
cathepsin K knockout mouse bone revealed that, in contrast
to wild-type (ϩ/ϩ) osteoclasts, Ϫ/Ϫ osteoclasts showed no
cathepsin activity, confirming that cathepsin K is the pre-
dominant cysteine protease of the osteoclast (unpublished
observations).
In vitro assays, using isolated human osteoclasts seeded
onto bone chips, have revealed activated cathepsin K in
osteoclast resorption lacunae and trails.(25) In mammalian
cells, cysteine proteases are synthesized as latent precursors,
and can either be secreted as proenzymes or be transported
via mannose-6-phosphate receptors to lysosomal compart-
DISCUSSION
During osteoclast-mediated bone resorption, the organic ments where they can be processed to the mature active
constituents of bone are degraded by lysosomal enzymes, enzyme, to either play an intracellular role or be secreted.(43)
which are secreted directionally into the resorbing compart- The pH underneath the ruffled border of the osteoclast is

484
DODDS ET AL.
ity studies suggest that the neutral pH zone of the resorption
lacunae is the site of cathepsin K and interstitial collagenase
activity,(21,45) but that the activation of procathepsin K oc-
curs in the low pH environment by either cleavage by
another protease or by an autocatalytic process.(20,21,46) To
address this issue, we developed a microcytophotometric
assay to localize and quantify osteoclast cathepsin K activ-
ity in sections of human bone. The distribution of enzyme
activity was then compared with the pattern of immunore-
active pro- or mature cathepsin K protein expression.
Sections of osteoclastoma, which included abundant os-
teoclasts distant from bone formation sites, showed intense
expression of cathepsin K mRNA, high levels of procathe-
psin K, but negligible levels of mature cathepsin K. This
population of osteoclasts showed no cathepsin activity. We
further show that as osteoclasts approach bone (in both
osteoclastoma and fetal bone), there is processing of the
proenzyme to mature cathepsin K that directly correlated
with enzyme activity detected cytochemically. Similarly, in
both bone tissues, polarization of cathepsin K was observed
in osteoclasts before attachment to bone, suggesting that the
activation and polarization of cathepsin K is regulated by
bone and/or local factors. The osteoclasts apposed to bone
(in both fetal and osteoclastoma) almost exclusively ex-
pressed the mature form of cathepsin K and showed intense
enzyme activity, again polarized at the ruffled border. High-
power microscopy of fetal bone revealed a gradient of
activity from the apical surface of the osteoclast to the
ruffled border, consistent with lysosomal trafficking to the
ruffled border. One consistent observation noted in sections
of osteoclastoma tissue was that osteoclasts close (not at-
tached) to bone appeared larger and had higher cathepsin K
mRNA levels when compared with osteoclasts more distant
from bone. In contrast, osteoclasts attached to bone showed
highly variable expression. It remains to be determined
whether this variable expression pattern reflects transcrip-
tional regulation of the cathepsin K gene potentially at
different stages of osteoclast function.
The use of a sensitive cathepsin K cytochemical assay in
conjunction with immunolocalization and in situ hybridiza-
tion confirmed that cathepsin K activation occurs intracel-
lularly, before secretion into the resorption lacunae and the
onset of bone resorption. We further showed that cathepsin
K processing and polarization occurs as the osteoclast ap-
proaches bone, implying that local factors in the microen-
vironment may regulate these events.
FIG. 9. Sections of osteoclastoma that contain zones of (remodeling)
intramembranous ossification; serial sections were immunoreacted with
antibodies (A) CE5, (B) 10C7, (C) reacted for cathepsin activity
(Z-Phe-Arg-4M␤NA), or (D) hybridized with the cathepsin K antisense
probe. Osteoclasts are indicated (arrows and arrowheads). (E) An
osteoclast (arrowheads) resorbing mineralized bone in a section of fetal
bone reveals intense polarized activity at the osteoclast-bone interface.
Note the gradient of activity observed in vesicles spanning from the
apical surface (large arrowheads) to the ruffled border. Processes em-
bedded into the bone also show cathepsin activity (small arrowheads).
Section reacted for cathepsin activity using the Z-Phe-Arg-4M␤NA
substrate. (F) High power view of panel E highlights the osteoclasts
processes (arrowheads; bar ϭ 40 ␮m).
ACKNOWLEDGMENTS
We thank Dr. Richard Lackman from the Jefferson Hos-
pital, Philadelphia, PA, USA, for supplying the osteoclas-
toma tissue and Heather McClung for technical support.
REFERENCES
approximately pH 4.0.(5,44) However, it is speculated that
the pH at the zone closest to bone is more neutral, because
of the buffering capacity of dissolved bone salts.(21) Stabil-
1. Blair HC, Teitelbaum SL, Ghiselli R, Gluck S 1989 Osteoclas-
tic bone resorption by a polarized vacuolar proton pump.
Science 245:855–857.

CATHEPSIN K IS PROCESSED INTRACELLULARLY PRIOR TO BONE RESORPTION
485
2. Baron R, Neff L, Brown W, Courtoy PJ, Louvard D, Farquhar
MG 1988 Polarized secretion of lysosomal enzymes: Co-
osteoclast cathepsin K: Expression, purification, activation,
and substrate identification. J Biol Chem 271:12517–12524.
distribution of cation-independent mannose-6-phosphate re- 21. Bro¨mme D, Okamoto K, Wang BB, Biroc S 1996 Human
ceptors and lysosomal enzymes along the osteoclast exocytic
pathway. J Cell Biol 106:1863–1872.
3. Vaananen HK, Karhukorpi EK, Sundquist K, Wallmark B,
Roininen I, Hentunen T, Tuukkanen J, Lakkakorpi P 1990
cathepsin O2, a matrix protein-degrading cysteine protease
expressed in osteoclasts. Functional expression of human ca-
thepsin O2 in Spodoptera frugiperda and characterization of
the enzyme. J Biol Chem 271:2126–2132.
Evidence for the presence of a proton pump of the vacuolar 22. Garnero P, Borel O, Byrjalsen I, Ferreras M, Drake FH,
H(ϩ)-ATPase type in the ruffled borders of osteoclasts. J Cell
Biol 111:1305–1311.
4. Hall TJ, and Chambers TJ 1996 Molecular aspects of oste-
oclast function. Inflamm Res 45:1–9.
5. Silver IA, Murrills RJ, Etherington DJ 1988 Microelectrode
studies on the acid microenvironment beneath adherent mac-
rophages and osteoclasts. Exp Cell Res 175:266–276.
6. Salo J, Lehenkari P, Mulari M, Metsikko K, Vaananen HK
McQueney MS, Foged NT, Delmas PD, Delaisse J-M 1998
The collagenolytic activity of cathepsin K is unique among
mammalian proteinases. J Biol Chem 273:32347–32352.
23. Votta BJ, Levy MA, Badger A, Dodds RA, James IE, Thomp-
son S, Bossard MJ, Carr T, Connor JR, Tomaszek TA, Sze-
wczuk L, Drake FH, Veber DF, Gowen M 1997 Peptide
aldehyde inhibitors of cathepsin K inhibit bone resorption both
in vitro and in vivo. J Bone Miner Res 12:1396–1406.
1997 Removal of osteoclast bone resorption products by tran- 24. Thompson SK, Halbert SM, Bossard MJ, Tomaszek TA, Levy
sytosis. Science 276:270–273.
MA, Zhaou B, Smith WW, Abdel-Meguid SS, Janson CA,
D’Alessio KJ, McQueney MS, Amegadzie BY, Hanning CR,
Desjarlais RL, Briand J, Sarkar SK, Huddleston MJ, James
CFI, Carr SA, Garnes KT, Shu A, Heys JR, Bradbeer JN,
Zembryki D, Lee-Rykaczewski E, James IE, Lark MW, Drake
FH, Gowen M, Gleason JG, Veber DF 1997 Design of potent
and selective human cathepsin K inhibitors which span the
active site. Proc Natl Acad Sci USA 94:14249–14254.
7. Nesbit SA, Horton MA 1997 Trafficking of matrix collagens
through bone-resorbing osteoclasts. Science 276:266–269.
8. Hill PA, Buttle DJ, Jones SJ, Boyde A, Murata M, Reynolds
JJ, Meikle MC 1994 Inhibition of bone resorption by selective
inactivators of cysteine proteinases. J Cell Biochem 56:118–
130.
9. Debari K, Sasaki T, Udagawa N, Rifkin BR 1995 An ultra-
structural evaluation of the effects of cysteine-proteinase in- 25. Yamashita DS, Dodds RA 2000 Cathepsin K, and the design
hibitors on osteoclastic resorptive functions. Calcif Tissue Int
56:566–570.
10. Delaisse JM, Eeckhout Y, Vaes G 1984 In vitro and in vivo
of cathepsin K inhibitors. Curr Pharm Des 6:1–24.
26. Greenspan A 1991 Sclerosing bone dysplasias—a target-site
approach. Skeletal Radiol 20:561–583.
evidence for the involvement of cysteine proteinases in bone 27. Sarnsethsiri P, Hitt OK, Eyring EJ, Frost HM 1971
resorption. Biochem Biophys Res Commun 125:441–447.
Tetracycline-based study of bone dynamics in pycnodysosto-
11. Lerner UH, Grubb A 1992 Human cystatin C, a cysteine
sis. Clin Orthop 74:301–312.
proteinase inhibitor, inhibits bone resorption in vitro stimu- 28. Cabrejas ML, Fromm GA, Roca JF, Mendez MA, Bur GE,
lated by parathyroid hormone and parathyroid hormone-
related peptide of malignancy. J Bone Miner Res 7:433–439.
12. Delaisse JM, Boyde A, Maconnachie E, Ali NN, Sear CHJ,
Ferreyra ME, Demarchi C, Schurman L 1976 Pycnodysosto-
sis: Some aspects concerning kinetics of calcium metabolism
and bone pathology. Am J Med Sci 271:215–220.
Eeckhout Y, Vaes G, Jones SJ 1987 The effects of inhibitors 29. Meredith SC, Simon MA, Laros GS, Jackson MA 1978 Pyc-
of cysteine proteinases and collagenase on the resorptive ac-
nodysostosis. A clinical, pathological, and ultramicroscopic
tivity of isolated osteoclasts. Bone 8:305–313.
study of a case. J Bone Joint Surg Am 60:1122–1127.
13. Drake FH, Dodds RA, James IE, Connor JR, Debouck C, 30. Elmore SM, Nance WE, McGee BJ, Montmollin ME, Engel E
Richardson S, Lee-Rykaczewski E, Coleman L, Rieman D,
Bartlow R, Hastings G, Gowen M 1996 Cathepsin K, but not
1966 Pycnodysostosis, with a familial chromosome anomaly.
Am J Med 40:273–282.
cathepsins B, L, or S, is abundantly expressed in human 31. Gelb BD, Shi GP, Chapman HA, Desnick RJ 1996 Pycnodys-
osteoclasts. J Biol Chem 271:12511–12516.
ostosis, a lysosomal disease caused by cathepsin K deficiency.
14. Tezuka K-I, Tezuka Y, Maejima A, Sato T, Nemoto K, Ka-
Science 273:1236–1238.
mioka H, Hakeda Y, Kumegawa M 1994 Molecular cloning of 32. Johnson MR, Polymeropoulos MH, Vos HL, Ortiz de Luna RI,
a possible cysteine proteinase predominantly expressed in os-
teoclasts. J Biol Chem 269:1106–1109.
15. Shi GP, Chapman HA, Bhairi SM, DeLeeuw C, Reddy VY,
Francomano CA 1996 A nonsense mutation in the cathepsin K
gene observed in a family with pycnodysostosis. Genome Res
11:1050–1055.
Weiss SJ 1995 Molecular cloning of human cathepsin O, a 33. Saftig P, Hunziger E, Wehmeyer O, Jones S, Boyde A, Rom-
novel endoproteinase and homologue of rabbit OC2. FEBS
Lett 357:129–134.
16. Bro¨mme D, Okamoto K 1995 Human cathepsin O2, a novel
merskirch W, Moritz J, Schu P, von Figura K 1998 Impaired
osteoclastic bone resorption leads to osteopetrosis in cathepsin
K-deficient mice. Proc Natl Acad Sci USA 95:13453–13458.
cysteine protease highly expressed in osteoclastomas and 34. Gowen M, Lazner F, Dodds RA, Kapadia R, Field J, Tavaria
ovary molecular cloning, sequencing and tissue distribution.
Biol Chem 376:379–384.
17. Li YP, Alexander M, Wucherpfennig AL, Yelick P, Chen W,
Stashenko P 1995 Cloning and complete coding sequence of a
M, Bertoncello I, Drake F, Zavarselk S, Tellis I, Hertzog P,
Debouck C, Kola I 1999 Cathepsin K knockout mice develop
osteoporosis due to a deficit in matrix degradation but not
mineralization. J Bone Miner Res 14:1654–1663.
novel human cathepsin expressed in giant cells of osteoclas- 35. Aibe K, Yazawa H, Abe K, Teramura K, Kumegawa M,
tomas. J Bone Miner Res 10:1197–2202.
18. Inaoka T, Bilbe G, Ishibashi O, Tezuka K-I, Kamegawa M,
and Kokubo T 1995 Molecular cloning of human cDNA for
Kawashima H, Honda K 1996 Substrate specificity of recom-
binant osteoclast-specific cathepsin K from rabbits. Biol
Pharm Bull 19:1026–1031.
cathepsin K: Novel cysteine proteinase predominantly ex- 36. Van Noorden CJ, Vogels IM, Smith RE 1989 Localization and
pressed in bone. Biochem Biophys Res Commun 206:89–96.
19. Fukunishi S, Iwata Y, Mort JS 1997 Immunolocalization of
cathepsin K at sites of bone resorption. Trans ORS 22:586.
cytophotometric analysis of cathepsin B activity in unfixed and
undecalcified cryostat sections of whole rat knee joints. J His-
tochem Cytochem 37:617–624.
20. Bossard MJ, Tomaszek TA, Thompson SK, Amegadzie BY, 37. Dodds RA, Connor JR, Drake F, Feild J, Gowen M 1998
Hanning CR, Jones C, Kurdyla JT, McNulty DE, Drake FH,
Gowen M, Levy MA 1996 Proteolytic activity of human
Cathepsin K mRNA detection is restricted to osteoclasts dur-
ing fetal mouse development. J Bone Miner Res 13:673–682.

486
DODDS ET AL.
38. Dodds RA, Gowen, M and Bradbeer JN 1994 Microcytopho- 44. Fallon MD 1984 Biology of the osteoclast. In: Cohn DV,
tometric analysis of human osteoclast metabolism: The lack of
activity in certain oxidative pathways indicates an inability to
sustain biosynthesis during resorption. J Histochem Cytochem
42:599–606.
Fujita T, Potts TJ Jr, Talmage RV (eds.) Endocrine Control of
Bone and Calcium Metabolism. Elsevier, Amsterdam, The
Netherlands, p. 144.
45. Delaisse JM, Vaes G 1992 Mechanisms of mineral solubiliza-
tion and matrix degradation in osteoclast bone resorption. In:
Rifkin BR, Gay CV (eds.) Biology and Physiology of the
Osteoclasts. CRC Press, Boca Raton, FL, pp. 289–314.
39. Rieman DJ, McClung HA, Dodds RA, James IE, Holmes S,
Drake FH, Gowen M 2000 Biosynthesis and processing of
cathepsin K in cultured human osteoclasts (in press).
40. Dodds RA, James IE, Rieman D, Hwang SM, Ahern R, Con- 46. McQueeney MS, Amegadzie BY, D’Alessio K, Hanning CR,
nor JR, Holmes SD, Thompson S, Veber D, Drake FH, Lark
MW, Gowen M 1998 Osteoclast cathepsin K is processed
intracellularly prior to osteoclast attachment to bone. Bone
23:S334.
McLaughlin MM, McNulty D, Carr SA, Ijames C, Kurdlya J,
Jones CS 1997 Autocatalytic activation of human cathepsin K.
J Biol Chem 272:13955–13960.
41. Baron R, Neff L, Louvard D, Courtoy PJ 1985 Cell-mediated
extracellular acidification and bone resorption: Evidence for a
low pH in resorbing lacunae and localization of a 100-kD
lysosomal membrane protein at the osteoclast ruffled border.
J Cell Biol 101:2210–2222.
42. Hart KA, Potts W, Bowyer J, Slynn G, Elston P, Vernon W,
Tucker D, Tolley K, Johnstone D 1996 Transgenic mice lack-
ing cathepsin L activity. J Bone Miner Res 11:S366.
Address reprint requests to:
Dr. Robert A Dodds
Osiris Therapeutics, Inc.
2001 Aliceanna Street
Baltimore, MD 21231, USA
43. Katunuma N 1989 Mechanisms and regulation of lysosomal Received in original form September 24, 1999; in revised form
proteolysis. Rev Biol Cell 20:35–61. September 8, 2000; accepted October 4, 2000.