Further evidence that sperm nuclear proteins are necessary for embryogenesis.

Abstract

We have recently presented evidence that the structural integrity of the mouse sperm nuclear matrix may be necessary for the proper unpackaging of sperm DNA for participation in embryogenesis. It is likely that the sperm nuclear matrix contributes to the organisation of the sperm DNA and its disturbance can seriously damage the paternal genome or its expression. In this work, we confirm our previous data and further suggest that even very subtle changes in the sperm nuclear structure may have a significant impact on embryo development. As reported previously, dithiothreitol (DTT) in the presence of an ionic detergent, ATAB, destabilized the nuclear matrix as measured by the halo assay, and oocytes injected with these nuclei failed to develop. We also discovered that omitting the protease inhibitor PMSF from the buffers used to extract spermatozoa prevented sperm injected into oocytes from participating in development. The organization of DNA into loop domains by the nuclear matrix in these nuclei appeared normal, as measured by the halo assay. Oocytes injected with sperm nuclei that had been washed with ATAB in the presence of phenylmethylsulphonyl fluoride (PMSF) but in the absence of DTT resulted in live births. Neither DTT treatment nor the absence of PMSF would be expected to disrupt the integrity of the paternal DNA. The data therefore suggest that even very subtle alterations in the structural proteins of the nucleus are enough to deprive sperm DNA of the ability to contribute to embryonic development.

Full text

Page 1

(Yanagimachi, 1994). It is known that mouse sperm
heads isolated from tails are not capable of entering the
oocyte by normal physiological means, but they can
nevertheless activate the oocyte and participate fully in
normal embryo development if injected microsurgi-
cally into the oocyte (Kuretake et al., 1996; Kimura et al.,
1998). This approach has allowed us to focus more
specifically on the genetic contribution of the mouse
sperm nucleus to embryo development. Previous work
has demonstrated that at least in the mouse all compo-
nents that are required for the full participation of the
paternal genome in embryogenesis reside within the
sperm nucleus and the perinuclear materials (Kimura
et al., 1998; Perry et al., 1999).
Most researchers believe that the only truly essential
component in the sperm head for embryogenesis is the
DNA itself, and that all other sperm components
Introduction
We have been interested in determining the compo-
nents of the sperm head that are essential for full devel-
opment of the embryo. The prime function of the sper-
matozoa is to deliver the paternal genome to the egg,
and to initiate development by activating the oocyte,
leading to the development of the normal embryo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Zygote 8 (February), pp 51–56. © 2000 Cambridge University Press Printed in the United Kingdom
All correspondence to: W. Steven Ward, PhD, Division of
Urology, Robert Wood Johnson Medical School, 1 RWJ Pl.,
MEB-588, New Brunswick, NJ 08903, USA. Tel: +1 (732) 235 8
73. Fax: +1 (732) 235 6042.
1Division of Urology, Robert Wood Johnson Medical School,
New Brunswick, New Jersey, USA.
2Department of Anatomy and Reproductive Biology, Univer-
sity of Hawaii Medical School, Honolulu, Hawaii, USA.
Further evidence that sperm nuclear proteins are necessary for
embryogenesis
W. Steven Ward1, Hidehumi Kishikawa2, Hidenori Akutsu2, Hiroko Yanagimachi2
and Ryuzo Yanagimachi2
Division of Urology, Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA and Department of Anatomy
and Reproductive Biology, University of Hawaii Medical School, Honolulu, Hawaii, USA
Date submitted: 25.7.99. Date accepted: 25.8.99
Summary
We have recently presented evidence that the structural integrity of the mouse sperm nuclear matrix
may be necessary for the proper unpackaging of sperm DNA for participation in embryogenesis. It
is likely that the sperm nuclear matrix contributes to the organisation of the sperm DNA and its dis-
turbance can seriously damage the paternal genome or its expression. In this work, we confirm our
previous data and further suggest that even very subtle changes in the sperm nuclear structure may
have a significant impact on embryo development. As reported previously, dithiothreitol (DTT) in the
presence of an ionic detergent, ATAB, destabilised the nuclear matrix as measured by the halo assay,
and oocytes injected with these nuclei failed to develop. We also discovered that omitting the protease
inhibitor PMSF from the buffers used to extract spermatozoa prevented sperm injected into oocytes from
participating in development. The organisation of DNA into loop domains by the nuclear matrix in these
nuclei appeared normal, as measured by the halo assay. Oocytes injected with sperm nuclei that had been
washed with ATAB in the presence of phenylmethylsulphonyl fluoride (PMSF) but in the absence of DTT
resulted in live births. Neither DTT treatment nor the absence of PMSF would be expected to disrupt the
integrity of the paternal DNA. The data therefore suggest that even very subtle alterations in the struc-
tural proteins of the nucleus are enough to deprive sperm DNA of the ability to contribute to embryonic
development.
Key words: Embryogenesis, ICSI, Paternal genome, Sperm DNA, Sperm nucleus

Page 2

evolved solely as the means of transporting the
parental DNA into the oocyte. This idea was supported
by the successful development of mice from oocytes
that had been injected with the nuclei of round sper-
matids (Ogura et al., 1994; Kimura & Yanagimachi,
1995b; Sato et al., 1998). The chromatin structure of the
round spermatid is very different from that of sperma-
tozoa, being much less compact and bound to histones
rather than protamines (Balhorn et al., 1984; Ward &
Coffey, 1991; Green et al., 1994). Furthermore, it has
been established that naked DNA injected into oocytes
will initiate the formation of nuclear-like structures,
and that the exogenous DNA will subsequently
undergo some DNA replication (Blow & Laskey, 1986;
Laskey et al., 1993). This suggests that if it were possi-
ble to inject the DNA from a single spermatozoon into
an oocyte, without any nicks or breaks, the oocyte
would form a pronuclear-like structure and at least ini-
tiate DNA replication. But whether such oocytes
would complete development to a live mouse is still an
open question.
Is the DNA the only component in the sperm
nucleus that is necessary for development, or is
there some important and necessary information in the
chromatin structure? The sperm nucleus contains
one level of DNA organisation that is independent
of protamines and histones: the organisation into loop
domains 20–100 kp in size by the nuclear matrix
(Nadel et al., 1995; Ward & Zalensky, 1996). In
somatic cells, this organisation is associated with
DNA replication (Pardoll et al., 1980; Vogelstein et
al., 1980) and transcription of RNA (Gerdes et al.,
1994; Getzenberg, 1994; Chang et al., 1995; Kim et al.,
1996). It is therefore possible that the loop domain
organisation of DNA by the nuclear matrix is im-
portant in the regulation of transcription and DNA
replication of the pronucleus. This type of organisation
can be visualised by extracting the histones (from
somatic cells) or protamines (from spermatozoa)
from isolated nuclei, and staining with a fluorescent
DNA binding dye, such as ethidium bromide (Vogel-
stein et al., 1980; Ward et al., 1989). Nuclei prepared
in this manner have a halo of fluorescence surround-
ing the nucleus, representing the DNA extending
from the nuclear matrix as loops. We have previ-
ously demonstrated that the mammalian sperm
nuclear matrix is extremely sensitive to extraction
conditions. Ionic detergents such as SDS protect
the sperm nuclear matrix from dissipation upon
subsequent extraction of the protamines with 2 M
NaCl and 10 mM dithiothreitol (DTT) (Ward et al.,
1989), while extraction with non-ionic detergents
such as Triton X-100 or NP-40 results in complete dis-
sipation of the sperm nuclear matrix (Ward & Coffey,
1989, 1991).
Since it is technically impossible at present to
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
52
W.S. Ward et al.
isolate chromosome-sized DNA molecules completely
intact for injection, we cannot perform the experi-
ment described above to answer the question of
whether the DNA alone is sufficient for the paternal
genome to participate in embryogenesis. However, we
have recently presented evidence that an intact sperm
nucleus is necessary for normal embryogenesis to
occur. We also demonstrated that treatment of mouse
spermatozoa with the ionic detergent ATAB were
capable of participating fully in embryogenesis when
injected into oocytes (Ward et al., 1999). However,
when 2 mM DTT was included in the original ATAB
extraction buffer, the nuclear matrices became un-
stable, and none of the oocytes injected with such
sperm nuclei developed. These data suggested that
a stable sperm nuclear matrix was important for nor-
mal embryogenesis. In this work, we confirm our
previous results, and extend our observations by
testing the hypothesis that more subtle variations in
the proteinaceous nuclear structure can also affect
the participation of the paternal genome in embryo-
genesis.
Materials and methods
Treatment of mouse spermatozoa with ATAB
The epididymal spermatozoa from one mouse (B6D2
F1) were collected, and suspended in 2 ml of nuclear
isolation media (NIM: 121.6 mM KCl, 7.8 mM
Na2HPO4,1.4 mM KH2PO4, 0.1% polyvinyl alcohol, 10
mM EDTA, with or without 1.0 mM PMSF; all supplied
by Sigma, St Louis, MO) supplemented with an ionic
detergent, 0.5% ATAB (mixed alkyltrimethylammo-
nium bromide; supplied by Sigma), and 2 mM dithio-
threitol (DTT; supplied by Sigma), and the pH adjusted
to 8.2 by adding drops of 1 M KOH. The sperm sus-
pension was occasionally vortexed, then centrifuged at
700 g for 5 min. The sperm pellet was washed in 10 ml
of NIM without ATAB or DTT, split into two aliquots
and centrifuged again. For each experiment, one sperm
pellet was used to test for halo formation (gross stabil-
ity of the nuclear matrix: see below) and the other
sperm pellet was used for intracytoplasmic sperm
injection (ICSI).
Halo assay for gross nuclear matrix stability
The pelleted sperm (heads and tails were all separated)
obtained as described above were resuspended in 100 µl
of 2 M NaCl, 25 mM Tris, pH 7.4, with 2 mM DTT. The
heads were incubated in the salt extraction buffer for 10
min on ice, then stained with 100 µg/ml ethidium bro-
mide to stain the DNA, and examined by fluorescence
microscopy. Sperm heads with stable nuclear matrices

Page 3

remain intact under these conditions, and the DNA
forms a halo of fluorescence surrounding the nuclei.
Sperm heads with unstable nuclear matrices decon-
dense completely, and no visible structure remains.
Preparation of oocytes
B6D2F1 females, 8–10 weeks old, were induced to
superovulate by consecutive injections of 5 IU eCG and
5 IU hCG 48 h apart. Oocytes, collected from oviducts
14–15 h after hCG injection, were freed from cumulus
cells by treatment with 0.1% bovine testicular
hyaluronidase (300 USP units/ng; ICN Pharmaceuti-
cals, Irvine, CA) in Hepes-buffered CZB medium
(Hepes-CZB) (Kimura & Yanagimachi, 1995a). The
oocytes were washed thoroughly and kept in CZB for
up to 2 h at 37 °C under 5% CO2in air. Oocytes to be
injected with spermatozoa were transferred into a
droplet (6 µl) of Hepes-CZB under mineral oil (Squibb
and Sons, Princeton, NJ) in an operation dish on the
microscope stage (Kimura & Yanagimachi, 1995a).
Sperm head injection into oocytes
One pellet of sperm heads prepared as above was
resuspended in 100 µl of NIM with 6% (w/v) poly-
vinylpyrrolidone (PVP, Mr360000; ICN Pharmaceuti-
cals, Irvine, CA) and a single sperm head was used for
ICSI. A single sperm nucleus was aspirated into the
injection pipette attached to the micromanipulator and
injected into an oocyte according to the method
described by Kimura & Yanagimachi (1995a) except
that all operations were carried out at room tempera-
ture (about 25 °C) instead of 16–17 °C. Approximately
10 min after sperm injection the oocytes were trans-
ferred into Ca2+-free CZB containing 10 mM SrCl2and
incubated for 1 h to activate them (Bos-Mikich et al.,
1997). Unlike fresh spermatozoa or Triton- treated
sperm heads (Kimura & Yanagimachi, 1995a; Kuretake
et al., 1996), ATAB + DTT-washed sperm nuclei could
not activate oocytes when injected into oocytes, so the
oocytes had to be activated. We used Sr2+as the acti-
vating agent. Sr2+-treated oocytes were incubated for
about 5 h in CZB medium, and examined with an
inverted microscope. Those with two distinct pro-
nuclei and the second polar body were considered
normally activated and fertilised.
Embryo transfer to foster mothers
Morulae/blastocyst embryos arising from normally
fertilised eggs were transferred to uteri of ICR (albino)
female mice which had mated with vasectomised
males of the same strain 3 days previously (Hogan et
al., 1994). Caesarian sections were performed on day
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Sperm nuclear proteins are necessary for embryogenesis
53
19.5 post coitum and pups with black eyes were raised
by foster mothers (albino CD-1).
Electron microscopy
Sperm heads were fixed in 2% glutaraldehyde in
cacodylate buffer, embedded in Spurr (Ladd Research
Industries, Burlington, VT), and thin-sectioned for
transmission electron microscopy.
Results
Gross stability of the sperm nuclear matrix
As described in the Introduction, mouse sperm nuclei
treated with ATAB + PMSF form halos when
subsequently extracted with 2 M NaCl and 10 mM
DTT to extract the protamines (Ward et al., 1999). In
these nuclei the nuclear matrix appears to be grossly
intact, since the organisation of DNA into loop
domains is still apparent. The inclusion of DTT
caused the nuclei to decondense completely when
subsequently extracted with salt. In the present work
we repeated these results, and added a second vari-
able: the protease inhibitor PMSF (Table 1). As in our
previous work, we found that the presence of DTT
destabilised the nuclear matrix. Interestingly, the
presence of PMSF in the original extraction buffer
had no effect on halo formation. This suggested
either that PMSF had no effect on sperm nuclear struc-
ture, or that the effect was too subtle to be detected by
the halo assay.
Table 1. Gross stability of sperm nuclear matrices after
pretreatment with various extraction buffers
Extraction buffer Nuclear matrix stability
ATAB + PMSF
ATAB
ATAB + DTT + PMSF
ATAB + DTT
Halos (stable)
Halos (stable)
Decondensed (unstable)
Decondensed (unstable)
ATAB, mixed alkyltrimethylammonium bromide; PMSF,
phenylmethylsulphonyl fluoride; DTT, dithiothreitol.
Electron microscopy
We have previously published electron micrographs of
mouse spermatozoa treated with ATAB and PMSF,
with and without DTT (Ward et al., 1999). In this work,
we examined the structure of sperm nuclei treated
under both conditions without PMSF (Fig. 1). Nuclei
after all four treatments look similar: the nuclei are
very condensed, and the perinuclear theca has been
almost completely removed.

Page 4

Sperm injection
For every experiment in which oocytes were injected,
half the sperm preparation was examined for the sta-
bility of the sperm nuclear matrix by the halo assay and
the other half was used for injections. In this way, the
correlations between sperm nuclear structure and
embryonic development could be assessed. The results
for the ICSI experiments for development to blastocyst
stage are illustrated in Fig. 2. For all four treatments of
spermatozoa used, only one allowed the spermatozoa
to participate in embryogenesis: washing with ATAB
in the presence of PMSF and in the absence of DTT. The
differences are striking: 63% of the eggs developed to
blastocysts when injected with sperm nuclei treated in
this manner, while only 1–7% of the eggs developed
when injected with sperm nuclei in all other treat-
ments. Table 2 summarises the results of those
embryos transferred to foster mothers that resulted in
live offspring. Again, only those oocytes that were
injected with spermatozoa treated with ATAB in the
presence of PMSF and the absence of DTT had the abil-
ity to develop to live young.
Discussion
The results presented here confirm our previous find-
ing (Ward et al., 1999) that sperm nuclei that do not
have stable nuclear matrices as measured by the halo
assay are not capable of participating in embryogene-
sis. Oocytes that are injected with these nuclei do not
develop normally, even though the DNA is most likely
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
54
W.S. Ward et al.
intact. The data support our previous suggestion that
the sperm nucleus requires additional factors for
embryogenesis in addition to the genetic information
encoded within the DNA. We have now demonstrated
that a much more subtle modification of the sperm
nuclear proteins brought about by the omission of the
protease inhibitor PMSF during the nuclear isolation
prevents sperm nuclei from participating in embryo-
genesis. This suggests that even very subtle alterations
in the sperm nuclear structure have broad effects on
embryogenesis.
These data have caused us to reconsider the widely
accepted notion that the only function of the sperm
nucleus is to deliver a genetically intact package of
paternal DNA to the oocyte. In these experiments the
DNA has been delivered intact to the oocyte. Further-
more, it is clear that the oocyte has access to the pater-
nal DNA since 47–82% of the oocytes injected with
sperm nuclei under all conditions tested develop at
least to the 2-cell stage (Table 2), and virtually all the
sperm nuclei injected formed pronuclei (data not
shown). This demonstrates that the sperm DNA is
completely unpackaged, and accessible to the develop-
ing embryo. As with the previous study, treatment of
sperm nuclei with the ionic detergent ATAB did not
prevent them from participating in development, but
treatment with ATAB plus DTT did. DTT alone would
not be expected to damage DNA. A recent study has
suggested that pretreatment of spermatozoa with DTT
actually improves the efficiency of fertilisation by
sperm injection into human oocytes (Rho et al., 1998).
Thus, it is DTT in the presence of ATAB that deprived
sperm nuclei of their competence to participate in
embryogenesis, suggesting that a modification of a
protein component, not DNA damage, is responsible
for this loss of competence.
In this study we also demonstrated that the more
subtle alteration in sperm protein structure that results
from omission of the protease inhibitor PMSF renders
sperm nuclei incompetent. PMSF is not necessary for
Figure 1 Electron micrographs of mouse sperm nuclei.
Mouse sperm nuclei were treated with ATAB (A) or ATAB +
DTT (B), both without PMSF. ATAB, mixed alkyltrimethyl-
ammonium bromide; DDT, dithiothreitol; PMSF, phenyl-
methylsulphonyl fluoride.
Figure 2 Development of oocytes to the blastocyst stage.
Abbreviations as in Fig. 1.

Page 5

the gross stability of the sperm nuclear matrix, as mea-
sured by the halo assay (Table 1), but it is necessary for
sperm nuclei to participate fully in embryogenesis
(Table 2). PMSF would be expected to protect nuclear
proteins from autodigestion by endogenous proteases,
thereby stabilising proteins, not DNA, so this is further
evidence that a proteinaceous element is present in the
mouse sperm nucleus that is important for keeping
DNA intact for embryogenesis. This protein(s) is
apparently not removed by extraction with the ionic
detergent ATAB.
These results support the hypothesis that there are
elements other than the DNA in the sperm nucleus that
are essential for keeping the sperm nucleus competent
to participate fully in embryogenesis.
Acknowledgements
This work was supported by NIH grants HD-34362 to
R.Y. and HD-28501 to W.S.W.
References
Balhorn, R., Weston, S., Thomas, C. & Wyrobek, A.J. (1984).
DNA packaging in mouse spermatids: synthesis of prota-
mine variants and four transition proteins. Exp. Cell Res.
150, 298–308.
Blow, J.J. & Laskey, R.A. (1986). Initiation of DNA replica-
tion in nuclei and purified DNA by a cell-free extract of
Xenopus eggs. Cell 47, 577–87.
Bos-Mikich, A., Whittingham, D.G. & Jones, K.T. (1997).
Meiotic and mitotic Ca2+oscillations affect cell composi-
tion in resulting blastocysts. Dev. Biol. 182, 172–9.
Chang, K.S., Fan, Y.H., Andreeff, M., Liu, J. & Mu, Z.M.
(1995). The PML gene encodes a phosphoprotein associ-
ated with the nuclear matrix. Blood 85, 3646–53.
Gerdes, M.G., Carter, K.C., Moen, P.T., Jr. & Lawrence, J.B.
(1994). Dynamic changes in the higher-level chromatin
organization of specific sequences revealed by in situ
hybridization to nuclear halos. J. Cell Biol. 126, 289–304.
Getzenberg, R.H. (1994). Nuclear matrix and the regulation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Sperm nuclear proteins are necessary for embryogenesis
55
of gene expression: tissue specificity. J. Cell. Biochem. 55,
22–31.
Green, G.R., Balhorn, R., Poccia, D.L. & Hecht, N.B. (1994).
Synthesis and processing of mammalian protamines and
transition proteins. Mol. Reprod. Dev. 37, 255–63.
Hogan, B., Beddington, R., Costantini, F. & Lacy, E. (1994).
Manipulating the Mouse Embryo, pp. 173–81. Cold Spring
Harbor: Cold Spring Harbor Laboratory Press.
Kim, M.K., Lesoon-Wood, L.A., Weintraub, B. D. & Chung,
J.H. (1996). A soluble transcription factor, Oct-1, is also
found in the insoluble nuclear matrix and possesses silenc-
ing activity in its alanine-rich domain. Mol. Cell. Biol. 16,
4366–77.
Kimura, Y. & Yanagimachi, R. (1995a). Development of nor-
mal mice from oocytes injected with secondary spermato-
cyte nuclei. Biol. Reprod. 53, 855–62.
Kimura, Y. & Yanagimachi, R. (1995b). Mouse oocytes
injected with testicular spermatozoa or round spermatids
can develop into normal offspring. Development 121,
2397–405.
Kimura, Y., Yanagimachi, R., Kuretake, S., Bortkiewicz, H.,
Perry, A.C. & Yanagimachi, H. (1998). Analysis of mouse
oocyte activation suggests the involvement of sperm peri-
nuclear material. Biol. Reprod. 58, 1407–15.
Kuretake, S., Kimura, Y., Hoshi, K. & Yanagimachi, R. (1996).
Fertilization and development of mouse oocytes injected
with isolated sperm heads. Biol. Reprod. 55, 789–95.
Laskey, R.A., Mills, A.D., Philpott, A., Leno, G.H., Dilworth,
S.M. & Dingwall, C. (1993). The role of nucleoplasmin in
chromatin assembly and disassembly. Phil. Trans. R. Soc.
Lond., Ser B. 339, 263–9; discussion 268–9.
Nadel, B., De Lara, J. & Ward, W.S. (1995). Structure of the
rRNA genes in the hamster sperm nucleus. J. Androl. 16,
517–22.
Ogura, A., Matsuda, J. & Yanagimachi, R. (1994). Birth of nor-
mal young after electrofusion of mouse oocytes with
round spermatids. Proc. Natl. Acad. Sci. USA 91, 7460–2.
Pardoll, D.M., Vogelstein, B. & Coffey, D.S. (1980). A fixed
site of DNA replication in eucaryotic cells. Cell 19, 527–36.
Perry, A.C., Wakayama, T. & Yanagimachi, R. (1999). A novel
trans-complementation assay suggests full mammalian
oocyte activation is coordinately initiated by multiple,
submembrane sperm components. Biol. Reprod.
747–55.
Rho, G.J., Kawarsky, S., Johnson, W.H., Kochhar, K. & Bet-
60,
Table 2. Development of mouse oocytes after sperm injection
No. (%) development in vitro 72–76
h after sperm injection
Sperm extraction
buffer
No. of oocytes
injected
No. of oocytes
survived
>8- cell MorulaBlastocystLive
offspring
ATAB + PMSF
ATAB
ATAB + PMSF + DTT
ATAB, DTT
92
71
81
150
64 (70%)
56 (79%)
70 (86%)
120 (80%)
3 (5%)
46 (82%)
52 (74%)
62 (52%)
21 (33%)
9 (16%)
16 (23%)
51 (43%)
40 (63%)
1 (2%)
2 (3%)
7 (6%)
7 (11%)
0
0
0
Abbreviations as in Table 1.