Clinical Science (1985) 69, 681-686
68 1
The calmodulin content of the epidermis in psoriasis
S . MAC NEIL, W . F . G . TUCKER*, R . A. DAWSON, S . S. BLEEHEN*
A N D S . TOMLINSON
Deparrmenr of Medicine, Clinical Sciences Cenrre, Norrhern General Hospital, Sheffield, U.K., and *Deparrmenr of
Dermarology, Royal Hallamshire Hospiral, Sheffield, U.K.
(Received 29 April128 June 1985; accepted 2 J u 1985)
~
Summary
1. The calmodulin content of epidermis was
determined by assay of biologically active and
radioimmunoassayable calmodulin in epidermal
biopsy samples of 16 normal control subjects and
36 patients with psoriasis.
2. Calmodulin levels in the involved epidermis
of patients with psoriasis were significantly greater
than in epidermis of control subjects, with both
methods of calmodulin measurement. Levels of
calmodulin in the uninvolved epidermis were also
elevated but to a lesser degree, achieving statistical
significance only when measured by radioimmunoassay. However, the degree of correlation between
the two measurements of calmodulin was poor
for the patient samples, suggesting that each may
measure a different form of calmodulin.
3. The specificity of the elevated calmodulin
in psoriatic epidermis was investigated by measuring calmodulin in another unrelated tissue. Calmodulin activity in circulating peripheral blood
lymphocytes of seven patients with psoriasis was
similar to that found in the lymphocytes of ten
normal volunteers.
4. The relationship between calmodulin and
the hyperproliferative state of the psoriatic epidermis was investigated. No significant increase in
calmodulin activity was found after mitogen
stimulation of lymphocyte proliferation or after
Sellotape-stripping of the epidermis by a protocol
which has been shown to cause hyperproliferation
of the epidermis.
Correspondence: Dr Sheila Mac Neil, Department of Medicine, Clinical Sciences Centre,
Northern General Hospital, Sheffield S5 7AU,
U.K.
5. Elevated calmodulin in psoriatic epidermis
therefore appears to be a localized phenomenon of
the disease and does not appear to be a consequence of the hyperproliferative state of the
epidermis.
Key words. calmodulin, epidermis, lymphocytes,
psoriasis.
Abbreviation: PHA, phytohaemagglutinin.
Introduction
Psoriasis, a disease characterized by hyperproliferation of the epidermis, affects up to 2%
of the U.K. population. The cause of psoriasis
remains unknown but many biochemical abnormalities have been described in the involved (and
to a lesser extent in the uninvolved) epidermis of
patients with psoriasis. Several of these features
of psoriasis, greatly increased cellular proliferation [ 1,2], abnormal cyclic nucleotide metabolism
[3], increased phospholipase A2 activity [4] and
increased polyamine synthesis [S], are dependent
on the action of the calcium dependent regulatory
protein, calmodulin. Calmodulin is involved in the
regulation of the enzymes governing all of these
different processes [6]. Accordingly, an abnormality in the synthesis, activity or degradation
of calmodulin could be a common denominator
for these diverse biochemical abnormalities in
psoriasis.
Recently, immunoassayable levels of calmodulin
were reported to be grossly elevated (30-fold) in
the involved epidermis in psoriasis. Levels in the
uninvolved epidermis were normal [7].Because of
the potential importance of this finding, we
recently sought to confirm it by measuring the
682
S. Mac Neil et al.
levels of biologically active calmodulin, rather than
immunoassayable
calmodulin, in
psoriatic
epidermis. We found a sixfold increase in the levels
of calmodulin activity present in both the involved
and uninvolved epidermis of 16 patients with
psoriasis compared with the epidermis of 12
control volunteers [8]. It was thought that this
more modest increase in calmodulin might be
attributable to the different assays used as
discrepancies between calmodulin levels measured
by radioimmunoassay and by activity assay have
been reported previously [9]. There is also
evidence that both assays may be underestimates
of the absolute calmodulin content present in a
tissue [ 101.
The aims of the present study were threefold.
Firstly, to explore the differences between the two
reports of elevated epidermal calmodulin in
psoriasis [7, 81 by comparing calmodulin levels
measured by radioimmunoassay and by calmodulin activity assay in a larger group of
patients. Secondly, to investigate the specificity
of this increase in epidermal calmodulin by comparing the calmodulin levels in an unrelated tissue,
peripheral blood lymphocytes, in control volunteers and in patients with psoriasis. Finally, we
wished to investigate whether the increased calmodulin of psoriatic epidermis is related to the
hyperproliferative state of the epidermis. For the
latter, calmodulin levels were measured in control
volunteers’ lymphocytes induced to proliferate
by phytohaemagglutinin and in control volunteer
epidermis induced to proliferate by means of
Sellotape-stripping.
Subjects and methods
Subjects
Epidermal biopsies were obtained from 36
patients with psoriasis (18 male, 18 female, mean
age 44.5 years) and 16 normal controls (eight
male, eight female, mean age 52 years). Etlucal
Committee approval was obtained for this study.
Of the patients, two had erythrodermic psoriasis,
two had severe near-erythrodermic psoriasis
affecting over 90% of the skin surface, two had
stable plaque psoriasis affecting less than 5% of
their skin, and the remainder had extensive
chronic plaque psoriasis covering between 10 and
30% of their skin surface area (estimated by the
rule of nines method). Only patients who had not
had any form of systemic anti-metabolite or
retinoid treatment were entered in the study and
14 patients had been off all treatment for over 1
week before biopsy. The remaining patients had
used topical corticosteroid [ 101 and tar prepara-
tions [ 1 13 before biopsy. None of the patients was
receiving known calmodulin inhibitors. Venous
blood for lymphocyte preparation was obtained
from seven patients with chronic plaque psoriasis
(three male, four female, mean age 37 years), three
of whom had earlier given skin biopsies, and from
10 normal controls (six male, four female, mean
age 34 years).
Skin biopsy sampling
Epidermal shave biopsies approximately 3 mm
in diameter were cut freehand using a scalpel blade
under local anaesthetic from clinically uninvolved
skin and from psoriatic plaques. The forearm was
the selected site in all patients and volunteers.
Superficial scale was routinely removed before
taking biopsy samples since, in pilot studies,
psoriatic scale had been found to contain some
calmodulin activity.
Sellotape-stripping
Epidermal biopsies were taken from six of the
normal control volunteers from the medial aspect
of the left forearm immediately before and 24 h
after removal of the stratum corneum by 20
repeated applications and removal of standard
cellulose tape (Sellotape). After this procedure,
the skin had a glistening appearance indicative of
removal of the stratum corneum. This stimulus
has been shown to result in increased epidermal
proliferation, as evidenced by increased [3H]thymidine uptake into interfollicular epidermal
cells [ l l , 121, with the maximum increase in
mitotic index occurring approximately 24-48 h
post-stripping [ 131. Normal skin and non-involved
psoriatic skin are known to react in a similar
manner [ I l l .
Calmodulin sample preparation
Skin biopsy and lymphocyte samples were
each stored and homogenized in 1 ml of 40
mmol/l Tris-HC1, pH 7.4, with 100 p o l l 1 CaC12,
containing 50 mg/l phenylmethylsulphonyl fluoride
and SOp1/1 pepstatin A, with approximatel-y 20
strokes of a tight glass Dounce homogenizer at
4°C. Aliquots of this homogenate were taken for
protein determination by the method of Lowry
et al. [ 141. (A rapid Coomassie blue protein assay
we had used previously [8] for this purpose we
have since found to underestimate the amount of
protein present without prior solubilization and
neutralization of homogenates.) The remainder
of the homogenate was heated to 90°C for 6 min
Elevated epidermal calmodulin in psoriasis
and denatured protein removed by centrifugation.
Recovery of [3H]calmodulin (Amersham International Ltd, Bucks., U.K.) in the supernatant was
95 k 1.3% (mean f SEM, n = 6) for epidermal
samples homogenized and incubated with [3H]calmodulin before extraction.
Supernatants were stored at -20°C before
assay, which was usually within 7 days for assay
of biologically active calmodulin.
Radioimmunoassays for calmodulin were also
performed in batches on these supernatants [after
appropriate dilution with radioimmunoassay buffer] within 2-3 months of sample preparation.
The stability of calmodulin in stored supernatants was confirmed by comparing calmodulin
activity in seven supernatants assayed within 7
days of storage at -20°C (3.0620.9 pg of calmodulin activity/mg of protein, mean k SEM) with
supernatants assayed after a minimum of 6 months
storage at -20°C (3.27 k 1.27 pg of calmodulin
activity/mg of protein, mean f SEM).
Lymphocyte preparation
683
these lymphocytes (details as previously described
[161).
Assay of biologically active calmodulin
Biologically active calmodulin was assayed as
previously described [17] with a beef heart calmodulin-sensitive phosphodiesterase (Boehringer
Mannheim, London). Assays contained, in a fiial
reaction mixture of 400 A, 40 mmol/l Tris-HC1,
pH 7.0 at 37"C, 4 mmol/l 2-mercaptoethanol,
5 mmol/l MgClz, cyclic [3H]AMP (2 x lo5c.p.m./
tube), 100 pmol/l cyclic AMP, 25 pnol/l CaClz
and calmodulin or epidermal extract as required.
Calmodulin activity was determined at three
dilutions of each sample and samples were usually
assayed in at least two assays within 7 days of
storage. The interassay coefficient of variation for
the assay was 17.7% for one normal epidermal
sample (mean value f SD of 0.34 k 0.06 pg of
calmodulinlmg of epidermal protein) assayed in
seven consecutive assays. The amount of calmodulin activity present was related to the protein
content of the original homogenate. Assays were
performed in triplicate.
Lymphocytes were prepared as in [ 151. Venous
blood (15 ml) was taken from each subject into a
sterile 20 ml plastic tube containing 500 units of
Assay of immunoreactive calmodulin
preservative-free heparin. Approximately 1 g of
Immunoreactive calmodulin was determined by
carbonyl iron powder (G.A.F. Ltd, Manchester,
U.K.) was then added, followed by 5 m l of using a commercially available radioimmunoassay
Dextran 150 in sodium chloride solution (154 (Amersham International Ltd). Supernatants, prepared as described previously, were diluted tenfold
mmol/l) (Fisons plc, Sussex, U.K.). Samples were
(or greater as required) with radioimmunoassay
gently inverted and then placed in a water bath
at 37°C with further inversion of each tube every buffer (125 mmol/l borate buffer, pH 8.4, con5 min until 30 min when samples were removed taining 1 mmol/l EGTA and 75 mmol/l NaC1).
After initial confirmation that epidermal superand placed on a magnetic field for a further 30
min. Supernatant was removed to another tube natants diluted in parallel to the standard curve
and 10 ml of a lymphocyte separation medium for pure calmodulin, samples were assayed at a
(Flow Laboratories, Herts., U.K.) carefully layered single dilution in one assay only. Assays were
underneath and the tubes centrifuged at 800 g performed in duplicate.
for 10 min. Supernatants were discarded and the
deposits washed with sodium chloride solution Statistics
(154 mmol/l) and centrifuged as before. This
Values are expressed as the means k SEM.
washing step was repeated once more and then the
lymphocyte pellet was resuspended in 1 ml of 40 Differences between means were compared by
mmol/l Tris-HC1, pH 7.4, with 100 /.moll1 CaC12, paired, non-paired t-test or by Mann-Whitney Utest as appropriate. Values of P < 0.05 were taken
containing
50 mg/l
phenylmethylsulphonyl
fluoride and 5Opl/l pepstatin A, and stored at as statistically significant. Correlation coefficients
-20°C until assay of biologically active cal- of the relationship between calmodulin levels
measured by radioimmunoassay and by assay of
modulin (usually within 7 days).
For seven of the normal control volunteers, biologically active calmodulin were determined.
lymphocytes were prepared as described and then
cultured for 72 h with an optimally stimulating
Results
concentration of phytohaemagglutinin (PHA)
(Wellcome Diagnostics, Dartford, U.K.) and [3H]- Epidermal calmodulin levels were significantly
thymidine (Amersham International Ltd, Bucks., greater in the involved epidermis of patients with
U.K.) uptake was assessed in parallel cultures of psoriasis compared with levels in the epidermis
S. Mac Neil et al.
684
of normal volunteers, irrespective of the method
of calmodulin measurement (Fig. 1). However, for
the uninvolved epidermis, calmodulin levels
although higher than in control skin achieved
statistical significance only when measured by
radioimmunoassay . The mean value k SEM (n) for
biologically active calmodulin was 0.72 kO.12 (16)
pg of calmodulin/mg of epidermal protein for
normal volunteers, 2.29 f0.38 (36) (P< 0.005 by
Mann-Whitney U-test) for the psoriatic plaque of
36 patients and 1.42 f 0.26 (36) (N.S.) for the uninvolved epidermis. For calmodulin levels
measured by radioimmunoassay calmodulin was
0.82k0.16 (14) for normalvolunteers, 1.97kO.26
(28) (P<0.005)for psoriatic plaque and 1.29*
0.14 (29) (P< 0.05) for the uninvolved epidermis.
'Or
-Biologically
Immunoreactive
active calmodulin
calmodulin
c
.
E?
M
i
v
I
-
--
Psoriatic epidermis
Psoriatic epidermis
FIG. 1. Comparison of calmodulin content in
normal and psoriatic involved and uninvolved
epidermis determined by assay of biologically
active calmodulin and of radioimmunoassayable
calmodulin. Biologically active calmodulin was
determined in samples from 16 normal volunteers
and in the psoriatic plaque of 36 patients and in
the uninvolved epidermis of 36 patients. Radioimmunoassayable calmodulin was determined for
14 of the normal volunteer samples, 28 of the
psoriatic plaques and 29 of the uninvolved epidermis samples. Statistical significance was
determined as described in the Results section.
The correlation between calmodulin levels
measured by activity assay and by radioimmunoassay was examined for nine of the control volunteers and 28 of the patients with psoriasis. Calmodulin levels in control volunteer samples were
similar when measured by radioimmunoassay and
by activity assay, with a good degree of correlation between the two (r = 0.91).
For psoriatic epidermis, however (whether
involved or uninvolved), the values obtained for
levels of radioimmunoassayable and biologically
active calmodulin agreed closely for 23 out of the
55 comparisons possible (correlation coefficient of
r = 0.88), but there was little agreement for the
remainder, radioimmunoassayable levels appearing
either much higher or much lower than calmodulin
activity levels.
There was n o obvious correlation between the
severity or extent of the disease and the amount of
biologically active calmodulin present in the epidermis. For example, for the four patients with
erythrodermic or nearerythrodermic psoriasis calmodulin activity in the plaque was 4.66 l .43 (4)
pg/mg of protein, not significantly greater than for
the other patients sampled. Calmodulin levels were
also similar in the plaque of patients receiving
topical steroids at the time of the study [3.30f
0.80 (10) pg/mg of protein] and in those who had
been without any form of treatment for over 1
week at the time of biopsy I2.89k0.68 (14)pg/
mg of protein].
In an attempt to determine whether elevated
levels of calmodulin are a generalized defect in
psoriasis, calmodulin activity was measured in
peripheral blood lymphocytes (see Table 1).
Calmodulin levels in lymphocytes proved similar
for control volunteers and patients with psoriasis
[0.60 kO.16 (10) pg of calmodulin/mg of lympho-
*
TABLE 1. Epidermal and lymphocyte calmodulin
activity in patients with psoriasis and in normal
volunteers
Results shown are means f SEM ( n ) .
Calmodulin activity
(pg/mg of protein)
Epidermis
Psoriasis patients
Plaque
Uninvolved
Normal volunteers Untreated
2.29+-0.38(36)
1.42 i 0.26 (36)
0.88i0.18 (6)
Post Sellotape- 1.15r0.3 (6)
stripping
Lymphocytes
Psoriasis patients
No additions
0.5420.12 (7)
Normal volunteers No additions
0.60 t0.16 (10)
PHA-stimulated 0.72i0.07 (7)
Elevated epidermal calmodulin in psoriasis
cyte protein for controls and 0.54k0.12 (7) for
patients]. (Of this group of patients with psoriasis,
three had previously given skin biopsy samples and
the calmodulin levels in these were higher than in
normal epidermis.)
In investigating whether increased levels of
calmodulin are a feature of hyperproliferation,
calmodulin activity was determined in PHAstimulated lymphocytes and in Sellotape-stripped
epidermis in normal volunteers. PHA stimulation
of lymphocyte proliferation increased ['HIthymidine uptake by 375 & 77-fold (mean SEM,
n = 7) but did not significantly increase calmodulin activity. Similarly, when Sellotape-stripping
was used in a protocol which has elsewhere been
shown to induce epidermal cell proliferation [ 11131, a paired t-test on six control volunteers
showed no significant increase in calmodulin
activity after this procedure (see Table 1).
*
Discussion
Calmodulin occupies a pivotal role in the regulation of many intracellular processes, several of
which are known to be abnormal in psoriasis
[l-51. An alteration in calmodulin activity could,
therefore, explain many of these apparently unconnected features of psoriasis. Increases in epidermal calmodulin of 30-fold [7] and sixfold [8]
have previously been reported by measurement of
radioimmunoassayable calmodulin and calmodulin
activity respectively, although we now believe
the increased calmodulin levels in the latter report
to be an overestimate based on the use of inappropriate protein assay (see the Subjects and
methods section). In the present study a relatively
modest (two- to three-fold) increase in calmodulin
was found irrespective of whether calmodulin was
measured by radioimmunoassay or by assay of
calmodulin activity. Thus, the discrepancy
between our studies and that of Van de Kerkhof
& Van Erp [7] does not appear to be explained by
the different assay methods used. Moreover, there
are few studies in which calmodulin levels have
been measured by both activity assay and radioimmunoassay. We would suggest that the relatively
poor correlation found between the two for the
patient samples in the present study argues for
caution in comparing these assay values in any
quantitative sense. It is possible that the two
assays measure different forms of calmodulin.
Both may also underestimate the actual amount
of calmodulin present, as suggested by studies
using gel electrophoresis methodology [ 101.
Perhaps more interesting than the degree of
elevation in calmodulin levels in the psoriatic
685
plaque is our finding that calmodulin levels are
also slightly elevated in the clinically uninvolved
epidermis in psoriasis. Van de Kerkhof & Van Erp
[7] found calmodulin levels in the uninvolved
skin to be normal. The uninvolved skin in psoriasis,
while appearing clinically and histologically
normal, can show a number of abnormalities
including subtle ultrastructural changes, an
increase in the free mitotic index [ l , 21 and increased phospholipase A2 levels [4]. Therefore, if a
specific defect is present in psoriasis, it is perhaps
not surprising that it should manifest in both the
involved and uninvolved epidermis.
At present, we cannot explain the discrepancy
between the studies but it is possible that patient
selection and the site of sampling may go some
way towards explaining these results (P. C. M. Van
de Kerkhof, personal communication). The choice
of expressing calmodulin content relative to
tissue DNA [7] rather than to tissue protein [8]
is also unlikely to explain this discrepancy as one
would expect psoriatic plaque to be proportionately DNA-rich compared with uninvolved epidermis, hence calmodulin values expressed relative
to DNA would seem lower in the plaque.
To investigate whether increased calmodulin
activity is specific to the epidermis in psoriasis,
we measured the calmodulin activity present in an
unrelated and accessible tissue, peripheral blood
lymphocytes. There was no significant difference
in calmodulin activity between lymphocytes
prepared from normal volunteers and from
patients with psoriasis. This suggests either that
elevated calmodulin activity in psoriasis may be a
specific defect of the epidermis or that calmodulin
may be elevated in dividing cell populations. Since
calmodulin is known to be particularly important
in mitosis [ 101, a raised calmodulin level might be
expected to occur in hyperproliferative tissue such
as the psoriatic plaque.
The relationship between calmodulin and cellular proliferation was examined in normal lymphocytes induced to proliferate with PHA and in
normal epidermis treated with a regimen of
Sellotape-stripping, which has been shown to
induce epidermal proliferation [ 11-1 31. Neither
procedure caused a proportional increase in
calmodulin over the other cellular protein present.
This is very much in accord with the picture
which is currently emerging of calmodulin being
elevated after cellular transformation but not
under conditions of normal cellular proliferation
[lo]. The calmodulin levels we have found in
psoriatic plaque and in the much less proliferating
uninvolved epidermis are consistent with the
levels reported for several neoplastic tissues and
transformed cells [lo].
686
S. Mac Neil et al.
In conclusion, in the present study we demonstrate that calmodulin levels, determined by either
radioimmunoassay or activity assay, are elevated
in the involved epidermis in psoriasis. Calmodulin
levels in the uninvolved epidermis are also slightly
elevated, although to a lesser degree. This defect
appears specific t o the epidermis and does not
appear to be simply a consequence of the hyperproliferative state of the epidermis. The relevance
of the increased calmodulin to the aetiology of
the disease remains unknown but such an increase
in calmodulin could be responsible for many of
the diverse biochemical abnormalities known to
occur in psoriatic epidermis.
Note added in proof
Since submission of this manuscript similar values
for psoriatic epidermal calmodulin in involved and
uninvolved skin have been reported by Fairley,
Marcelo, Hogan & Voorhees (Journal of Znvestigative Dermatology (1985) 84,195-198).
Acknowledgments
We are grateful to the MRC, the Smith Kline
Foundation, the Psoriasis Association and the
Wellcome Trust for financial support. S.T. is a
Wellcome Trust Senior Lecturer. We thank Dr K.
Gelsthorpe and Mrs L. Cawood for their assistance
with the lymphocyte preparations and culture and
Mr S. Quayle for provision of normal abdominal
epidermis.
References
1. Weinstein, G.D. & Frost, P. (1968) Abnormal cell
proliferation in psoriasis. Journal of Investigative
Dermatology, 50,254-259.
2. Goodwin, P., Hamilton, S. & Fry, L. (1974) The cell
cycle in psoriasis. British Journal of Dermatology,
90,517-523.
3. Marcelo, C.L., Duell, E.A., Stawiski, M.A., Anderson,
T I . & Voorhees, J.J. (1979) Cyclic nucleotides in
psoriatic and normal keratomed epidermis. Journal
of Investigative Dermatology, 12, 20-24.
4. Forster, S., Ilderton, E., Summerly, R. & Yardley,
H.F. (1983) The level of phospholipase A, activity is
raised in the uninvolved epidermis of psoriasis. British
Journal of Dermatology, 108,103-105.
5. Russell, D.H., Combat, W.L., Duell, E.A., Stawiski,
M.A., Anderson, T.F. & Voorhees, J.J. (1978) Glucocorticoid inhibits elevated polyamine synthesis in
psoriasis. Journal of Investigative Dermatology, 7 1,
177-181.
6. Tomlinson, S., Mac Neil, S., Walker, S.W., Ollis, C.A.,
Merritt, J.E. & Brown, B.L. (1984) Calmodulin and
cell function. Clinical Science, 66,497-508.
7. Van de Kerkhof, P.C.M. & Van Erp, P.E.J. (1983)
Calmodulin levels are grossly elevated in the psoriatic
lesion. British Journal of Dermatology, 108, 217-218.
8. Tucker, W.F.G., Mac Neil, S., Bleehen, S.S. & Tomlinson, S. (1984) Biologically active calmodulin levels are
elevated in both involved and uninvolved epidermis
in psoriasis. Journal of Investigative Dermatology,
82,298-299.
9. Chafouleas, J.G., Dedman, J.R., Munjaal, R.P. &
Means, A.R. (1979) Development and application of a
sensitive radioimmunoassay . Journal of Biological
Chemistry, 254,10262-10267.
10. Veigl, M.L., Vanaman, T.C. & Sedwick, W.D. (1984)
Calcium and calmodulin in cell growth and transformation. Biochimica et Biophysica Acta, 114,
84-92.
11. Wiley, H E . & Weinstein, G.D. (1979) Abnormal
proliferation of uninvolved psoriatic epidermis:
differential induction by saline, propranolol and tape
stripping in vivo. Journal of Investigative Dermatology,
73,545-547.
12. traun-Falco, O., Christopher, E. & Kurban, A. (1967)
Abnormes verhalten der epidermalen regeneration bei
patienten mit psoriasis vulgaris. Archives Klinical
patienten mit psoriasis vulgaris. Archiv fuer Klinische
und Experimentelle Dermatologie, 229,276-284.
13. Pinkus, H. (1951) Examination of the epidermis by
the strip method of removing horny layers. 1.
Observations on the thickness of the horny layer and
on mitotic activity after stripping. Journal of Investigative Dermatology, 16, 383-386.
14. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall,
R.J. (1951) Protein measurement with the Folin
phenol reagent. Journal of Biological Chemistry,
193,265-275.
15. Van Rood, J.J., Van Leeuwen, A. & Zweehus, R.
(1970) The 4a and 4b antigens. Do they or don’t
they? In: Hisrocompatability Testing, pp. 93-103.
Ed. Tersaki, P.1. Munksgaard, Copenhagen.
16. Mac Neil, S., Walker, S.W.,Brown, B.L. & Tomlinson,
S. (1982) Evidence that calmodulin may be involved in
phytohaemagglutinin-stimulated lymphocyte division.
Bioscience Reports, 2, 891-897.