Leukocyte adhesion deficiency in a Dutch Holstein calf: A case with a clear-cut family history

Veterinary Itntnunology and Itntnunopathology, Elsevier Science Publishers B.V.. Amsterdam

295

37 ( 1993) 295-308

Leukocyte adhesion deficiency in a Dutch Holstein calf: A case with a clear-cut family history W.E. Bernadinaa, A.J. Duitsb, H.C. Kalsbeek’, Th. Weming’, L. Elving” and G.H. Wentink’

W. Leiboidd,

“Institute of Infectious Diseases and immunology, Department oJItnmmtology, Faculty ojk’eterinary Medicine, Yalelaan 1, 3584 CL, Utrecht, Netherlands bDeparttnetlt ofExperimental Immunology, University Hospital, Utrecht, Netherlands ‘Department qflarge Animal Medicine and Nutrition, Faculty oJVeterinary Medicine, Utrecht, Netherlands dItntt~unology Unit, Veterinary School, Hannover, Germany ‘Department of Herd Health and Reproduction, Faculty of Veterinary Medicine, Utrecht. Netherlands (Accepted 15 October 1992)

ABSTRACT Bemadina, W.E., Duits, A.J., Kalsbeek, H.C., Wensing, Th., Leibold, W., Elving, L. and Wentink, G.H., 1993. Leukocyte adhesion deficiency in a Dutch Holstein calf: A case with a clear-cut family history. Vet. Imtnunol. Immunopathol., 37: 295-308. A leukocyte adhesion deficiency characterized by recurrent (predominantly bacterial) infections, lack of extravascular polymorphonuclear leukocyte(PMN) and pus formation has been described first in humans and then in dogs, and recently also in cattle. Because of important clinical similarities, a unitary explanation for the leukocyte adhesion deficiency (LAD) syndrome in mammals is proposed, inasmuch that an intrinsic leukocyte defect (i.e. mutations in genes encoding the common CD18 subunit), is thought to cause the disease. However, thus far, the hallmark of such intrinsic leukocyte defects, notably their heritability (or familial incidence), has not (yet) been unequivocally demonstrated. This is the first report to describe the occurrence of four Dutch bovine LAD (BLAD) cases with the clearest familial clustering observed to date. The diagnosis was based on the clinical features of very poor thriving, in general, of the calves, hypemeutrocytosis without appreciable left shift, and the absence of PMN CD 11a, or CD 11b, or CD 1 I c using monoclonal antibodies (mAb) and/or Concanavalin A binding activity of PMN lysates in immunoblots. Interestingly, a familial clustering was observed also for below-normal PMN CD1 Ic expression. Thus, a cow with low CD1 Ic expression (50.4%) and delivering three of the study BLAD calves, also had a healthy descendant with low (44.9%) PMN CD I 1c expression. These findings suggested the possibility that both subnormal expression and lack of PMN CD 11 expression are inheritable factors in cattle. Furthermore, a large prospective study using the present mAb for selecting relatives expressing the complete spectrum (0 to 2 90°h) of PMN Correspondence to: W.E. Bernadina, Institute of Infectious Diseases and Immunology, Department of Immunology, Faculty of Veterinary Medicine, Yalelaan 1, 3584 CL, Utrecht, Netherlands.

@ 1993 Elsevier Science Publishers

B.V. All rights reserved 0165-2427/93/$06.00

W.E. BERNADIN.4

296 CD I I /CD I S csprcssion would crcatc a comprchcnsivc study role ofgcnctic factors and ofsurvival straw& in BLr\D.

population

for understanding

ET .AL.

both

~hc

ABBREVIATIONS BLAD. bovine lcukocytc adhesion deficiency; oxide: LAD. leukocyte adhesion dclicicncy: monoclonal antibodies: PMN, polymorphonuclcar

BSA. bovine serum albumin: DMSO. DimcthylsullLFA-I. lymphocyte function-rclatcd antigen-I, m.\b. Icukocytc: SDS. sodium dodccyl sulfntc.

INTRODUCTION

Adhesion of adequate numbers of mammalian phagocytes to endothelial cells at the site of infection is critical for the mammal’s subsequent clearance of the infection (Smith et al., 1989; Lawrence et al., 1990). In this respect several families of adhesion molecules have been shown to be involved in phagocyte emigration at inflammatory sites (reviewed by Springer, 1990). Severe, occasionally fatal, recurrent infections despite marked persistent leukocytosis have been described in children, dogs and calves (Anderson et al., 1985: Giger et al., 1987; Kehrli et al., 1990). This potentially devastating syndrome, termed either granulocytopathy syndrome or leukocyte adhesion deficiency (LAD) is further characterized by lack of pus formation and markedly decreased adhesion-dependent functions of white cells in vitro. The underlying pathogenesis reflects an intrinsic leukocyte defect causing the leukocytes of LAD patients to have deficient or no expression of some closely related surface glycoproteins termed leukocyte integrins. These include LFA1 (lymphocyte function-related antigen- 1)) Mac- 1 and p 150,95, (Anderson and Springer, 1987). Each leukocyte integrin comprises and cy and a p subunit, associated in an ~vj32 structure, the /32 subunit (CD1 8) being identical for all members of the subfamily. The molecules LFA- 1, Mac- 1 and p 150,95 differ in their cy subunits, designated CD 11 a, CD1 1b and CD 1 lc respectively, using immunological analysis. In man, intrinsic leukocyte defects are usually heridable and characteristically have a familial incidence. Many human and all animal cases of LAD reported so far appear to support this contention (Crowley et al., 1980; Springer et al., 1984; Giger et al., 1987; Kehrli et al., 1990). To date, mutations in genes encoding the common p2 subunit (CD 18) are thought to be the underlying cellular defect causing the LAD syndrome in man and probably also in animals (Kishimoto et al., 1987; Kehrli et al., 1990). In recent years there has been a marked increase in the use of important semen from prominent American Holstein bulls for improving milk production in Dutch Holstein cows. To the same end, enormous numbers of ‘prime’ American embryos have been imported into the Netherlands over the past decade. Given the substantial inbreeding with some bulls occurring in the

LEUKOCYTE

ADHESION

297

DEFICIENCY

USA (Young et al., 1988), and given the occurrence of (heridable) bovine LAD (BLAD) in the American Holstein population (Kehrli et al., 1990), BLAD was expected to appear in the Netherlands. Therefore, when a Dutch Holstein calf presentedclinically with ‘poor thriving’, recurrent bacterial infections, and constant leukocytosis without appreciable left shift, this study was started to address the questions of whether this calf had BLAD and whether a family history of this diseasecould be established. Thus, we could then, more reliably, select a comprehensive group of animals with a view to examining (a) defenseand/or (b) the role of genetic factors in BLAD. This report describesthe first ever caseof BLAD in the Netherlands, and additionally demonstrates that the affliction as well as PMN CD 11/CD 18 heterodimer expressionin general, may show a familial pattern. MATERIALS AND METHODS Case report

Calf B, a Dutch Holstein specimen was studied. The calf born on 2 February 1991, to its foster mother was the result of embryo transplantation (see Fig. 1 for grossfamily tree of Calf B ). The calf was referred to the University’s Clinic for Large Animal Medicine and Nutrition (hereafter referred to as Clinic) with the following history. Following birth, Calf B never really thrived

Amencan

Holstein-Frlesian

Bull

(Sire

of) h

Fig. 1. Schematic representation of the family tree of calves in the Netherlands with BLAD. (0 ), animals diagnosed as BLAD-affected based on clinical signs and either immunofluorescent FCM, or Con A-binding in immunoblots; (0 ), established healthy carriers of the BLAD gene as based on the evidence that BLAD is due to,a recessivesimple gene defect [ 61; (t), calveson which euthanasia was done owing to continuous poor health conditions. NS, not studied; ET, embryo transplantation.

29s

W.E. BERN.4DIN.4

ET -\L.

well, according to the owner. The calf had recurrent infections and responded poorly to apparent appropriate antibiotic therapy. Further clinical presentations included, a dull skin, retarded growth compared with Sister A (see Fig. 1) and some other not well-defined complaints (summarized by the calf’s owner as ‘poor-thriving’). At entry into the Clinic, Calf B had extreme leukocytosis (white blood cell (WBC) count 587.8 x 10’ ,& ’ ). The calf was given sustained appropriate antibiotic and Clenbuterol (Ventipulmin, Boehringer, Ingelheim, Germany) therapy. Blood test results obtained thereafter indicated the occurrence of a leukocytosis, characterized by no appreciable left shift and persistence during presumably infection free periods. Calf B was kindly donated to the Clinic for performing the present studies which were approved and in accordance with the guidelines for clinical research. White blood cell counts and differentiation Blood was obtained by jugular venipuncture and collected into EDTA-containing tubes. Total WBC counts were determined using a Sysmex cell counter. Blood smears were prepared for differential WBC analysis, stained by MayGrunwald/Giemsa, and 100 cells were differentiated into neutrophils, eosinophils, basophils, lymphocytes and monocytes. Total numbers and relative proportions (expressed as percentages) of each cell population per /ll of blood were estimated. Source and preparation

qfpolvmorphonuclear

leukocytes

In each test PMN from Calf B were studied simultaneously with leukocytes from control specimens. The controls included one or more of the following animals: B’s mother, B’s Sister A, B’s Brother C, two calves, put down owing to recurrent infections and persistent leukocytosis (PMN stored at - 70” C for further analysis because of lack of the below mAb at that time), and six to eight age-matched controls, kept at one of the University’s farms or at farms harbouring B’s relatives. For the preparation of PMN, venous blood (jugular vein) was collected in heparin and processed within 2 h of collection as follows. Heparinized blood, diluted 1: 1 in PBS supplemented with heparin was layered onto Ficoll-Isopaque (sp.gr. 1.078) and centrifuged at 2000 rpm for 45 min at room temperature. Thereafter, the cells in the plasma-Ficoll interface were isolated, the remaining supematant was aspirated and the RBC-PMN pellet was washed three times with PBS, followed by lysis of red blood cells (RBC). Lysis of RBC was obtained by incubation ( 15 min, 4” C), in 0.92% ammonium chloride supplemented with KI-ICOJ ( 1 .O g l- ’ ) and EDTA (37.2 m l- ’ ), pH 7.4. The resulting PMN preparations usually contained more than 97% viable granulocytes (neutrophils and eosinophils) as determined by May-Grunwald

LEUKOCYTE

ADHESION

DEFICIENCY

299

Giemsa staining, and dextran blue exclusion, respectively.PMN were counted and adjusted to a concentration of 2 x 10’ cells ml- ’ of either RPM1 or RPM1 supplemented with 10% fetal calf serum (FCS ). Monoclonal antibodies mAb used for the analysis of the isolated neutrophils by flow cytometry (FCM, seebelow) included a mAb (R15.7) to CD 18 (Entman et al., 1990); a mAb (IL-A99) to CD1 la; a mAb (IL-Al 5) to CD1 lb (Splitter and Morrison, 1991); and a mAb (NAM4) to CD 11c (Splitter and Morrison, 1991) . mAb R15.7 was kindly provided by R. Rothlein (Boehringer Ingelheim, Ridgefield, CT) whereasthe mAb IL-A 15, IL-A99 and NAM4 were gifts from N. MacHugh (International Laboratory for Research on Animal Diseases (ILRAD ), Nairobi, Kenya) and J. Letesson (Facultes Universitaires NotreDame de la Paix, Namur, Belgium), respectively. A mAb to human CD 18 and a mAb to human CD 11a, both cross-reactingwith its bovine counter-part were usedby one of us (W.L. ) for analysis of Calf B’s PMN in a parallel study performed in Germany. Specificity of each of the mAb R 15.7, IL-A99, IL-A 15,and NAM4 was verified in an immunoprecipitation assay as described elsewhere (Splitter and Morrison, 1991), using in the assay: (a) untreated mAb; (b) mAb following their incubation with both PMN and PMN lysates (50 fig per well) from healthy calves to indiscriminately remove antibody activity to CD1 1 molecules (or CD 18 molecules, as applicable). Westernimmunoblots of neutrophil lysates Mac- 1 analysis (CD 11b ) was performed in neutrophil lysates as described (Kehrli et al., 1990). Briefly, neutrophils (all stored at - 70°C) from clinically healthy Holstein (0.4-0.6 x lo6 PMN), and from sick specimensas well as from Calf B (2.4- 10x 1O6PMN ) were solubilized using 1% (Nonidet ) NP40 (Sigma, St Louis, MO) lysis buffer. The neutrophil lysatestogether with prestained molecular weight markers (Rainbow Markers, Pharmacia, Sweden) were loaded on a SDS(Sodium dodecyl sulfate)-4% polyacrylamide stacking gel. The antigens were separated (overnight) by SDS-Polyacrylamide gel electrophoresis (SDS-PAGE) on an SDS-8% polyacrylamide gel, and then electroblotted ( 1 h) onto nitrocellulose paper using the Trans-Blot (BIORAD ) apparatus. Following blocking of the nitrocellulose sheetwith 1% gelatin and sequential incubation with Con A (Concanavalin A)-biotin ( 10 pg ml- ’ ), and anti-biotin conjugated to horseradish peroxidase (v/v, 1: 3000)) colour was produced using 0.1% 0-dianisidine (Sigma) with 0.05% sodium nitroprusside and 0.0 15% H202 Mac-l is the main Con A-binding polypep-

300

W.E. BERN4DIN.A

tide in the 140-200 kDa region of SDS-PAGE-separated trophil lysates as prepared above.

polypeptides

ET AL.

of neu-

Immttnofluorescence Bovine PMN (2-5 x 10’ cells per test) were incubated with mAb (appropriately diluted in PBS containing 1% BSA (Bovine serum albumin ) and 0.1% NaN,) for 30 min at 4°C. After washing, cells were incubated with FITClabeled F( ab’)? fragments of goat anti-mouse Ig (H&L) antiserum (Tago, Burlingname, CA) for 30 min at 4°C. The cells, washed twice, were fixed with paraformaldehyde ( 1% v/v), and fluorescence quantitated using a FACStar (Becton & Dickinson). Fluorescence data of 5000 PMN are expressed as percentage fluorescent cells corrected for background fluorescence (incubated with conjugate only). In addition, antigen expression was quantitated as mean fluorescence intensity of the positive fluorescent cells expressed as mean channel number, plotted on a logarithmic scale. Assay for homotypic

adhesion

Homotypic adhesion of bovine PMN was performed using a procedure developed for human PMN (Rothlein and Springer, 1986). Briefly, PMN preparations were dispensed in appropriate aliquots ( 1 x 1O5 cells) into duplicate wells in 96-well plates (Nunc, Roskilde, Denmark). Phorbol myristate acetate (PMA) (Sigma, St Louis, MO) at a concentration of 20 ng ml- ’ in DMSO (Dimethylsulfoxide), was then added to each well. After 5, 10, and 15min incubation at 37°C samples were taken and both the free PMN and the PMN that were aught in aggregates were counted. At least 100 PMN were examined per sample. The data are expressed as percent aggregation, i.e. the percentage of PMN that were in clusters of five cells or more. Homotypic adhesion inhibition experiments were performed by incubating healthy PMN with IL-A99 (and R15.7 as a positive control) for 10 min at 37°C followed by addition of PMA and then performance of the assay as above. RESULTS

Haematologic

findings

Blood was collected from Calf B for approximately for either total WBC, and/or WBC composition. The these blood tests are presented in Table 1. Note that varied with time, but that lowest WBC numbers were

8 months and analysed most relevant results of total WBC from Calf B still three to four times

LEUKOCYTE

TABLE Blood DaG

j/l5 d/l8 4129 j/I5 6104 6117 6116 8106 8119 I?/15

ADHESION

DEFICIENCY

301

I test data’

of Calf

B in the course

WBC (X lo’/II-‘)

S7.8 60.5 29.1 26.6 34.3 38.6 32.3 76. I 55.8 38.2

of the study

Percent

(%)

Bas

Eos

PMN-B

PMN-S

Lymph

0 0 0 0 0 0 0 0

0 0 0 I 4 3 0 I

0 3 0 I 0 0 I 0

93 90 73 66 76 68 57 7s

7 6 27 32 20 29 32 ‘I

60

38

Monoc.

I

‘Normal ca11lc values for white blood cells (WBC) =5-10x IO’&’ ofblood. basophils (Bas).=j%,, cosinophils (Eos).
the normal average. At no time-point leukocytosis was accompanied by any left shift. It was noted that the leukocytosis was mainly caused by a mature neutrophilia with, at times, an eosinophilia and variable lymphocytosis. Note also that persistent leukocytosis in Calf B was hardly ever attributable to an obvious monocytosis.

Proper-ties oJ‘monoclona1

antibodies

used

Restricted specificity of mAb IL-A99 for the bovine CD 11 a/CD 18 heterodimer (LFA- 1 ), was reconfirmed in this study by its ability to inhibit homotypic aggregation of bovine PMN (not shown). Furthermore, immunoprecipitation of bovine PMN lysates with R 15.7, and IL-A99, and IL-A 15 as well as with NAM4 was performed. Pre-incubation (two times) of each of these mAb with either bovine PMN or PMN lysates (coated to microtiter plates) completely prevented the appearance of both the (approximately) 165 kDa and 97 kDa bands, a function consistent with the activity of anti-CD1 1 and anti-CD I8 mAb, respectively. Both the aggregation inhibition results and the results obtained in the immunoprecipitation assay corresponded well to the mAb’ cy unit (CD1 I ) reactivity data, reported previously (Entman et al., 1990; Splitter and Morrison, 1991), or (for IL-A99) communicated to us personally ( MacHugh, ILRAD ).

302

W.E. BERNADlN.-\

Neutrophil

CD1 I h/CD1 8 heterodimcr

(Mac- 1) upression

ET .AL.

studies

Initial studies were directed at measuring Mac- 1 contents of Calf B’s PMN using the property of Con A to react with Mac-l as (approximately) a 165 kDa band on immunoblots of neutrophil lysates. PMN from Calf B (kept at - 70°C) had no detectable (if any) Mac-1 on their surfaces as indicated by the following experimental finding. Lysates of Calf B PMN, even when prepared from 10x 1Oh, produced no Mac-1 band in the test. In contrast, the lysates prepared from 0.4-0.6x 10” PMN (kept at -70°C) of healthy agematched controls, all produced a clear Mac- 1 band (Fig. 2, Pane1 A). It was noted that the PMN lysates of two animals with persistent leukocytosis on which euthanasia was done because of untreatable health problems, had no detectable Mac-l (Fig. 2, Pane1 B), thus indicating, in retrospect, that these calves had BLAD. Results identical to those presented here were observed in three separate experiments. Homotypic

adherence studies

In order to investigate the adherence properties of the affected calf’s PMN, a PMA-induced homotypic adherence assay was performed. In the presence of PMA control PMN showed rapid homotypic adherence. In contrast, no significant homotypic adherence was observed when PMN of the affected calf were used (Fig. 3). Similar results were observed in two repeats of the experiment. B

A

20091-

+ .-J

69-

c id

iii ‘tI..

.P

1

8

c 1

2

34

5

6

9

ab

c

d

e

Fig. 2. Mac-l contents of bovine PMN lysates as determined with Con A blots. Panel A shows the Mac-l ‘profiles’ obtained with lysates prepared from two concentrations of PMN from Calf B (lane I. 2.4X IO6 PMN; Lane 2. IOX IO6 PMN) versus seven clinically normal cattle (Lanes 3-9. 0.4-0.6X IO6 PMN). Panel B shows the Mac-l ‘protilcs’ of two calves which wcrc killed because of BLAD-like health problems during life ( IO x I O6 PMN. Lanes a and b: Lane c. 7 x IO” PMN as used for Lane b) vs. two clinically healthy cows (0.4~ IOh PMN. Lanes d and c). Molecular weight standards (in kDa) arc also indicated. PMN lysatcs wcrc clcctrophorcscd on 8% gels under reducing conditions.

LEIIKOCYTE

ADHESION

303

DEFICIENC\

10

5 Time

15

(minutes)

Fig. 3. PMA-induced homotypic adhcrcnce of control and patient bovine PMN. Patient calf (cross-hatched columns) and control calf (solid columns) PMN were cultured in the presence of PMA and scored for homotypic adhesion after incubation for 5. 10, and IS min at 37°C. Data represent mean I? SD of duplicate dctcrminations.

Bovine neutrophil

LFA-1,

Mac-l

andpl50,95

expression

Binding of a panel of mAb directed against bovine CD 1 la, CD 11 b, CD 11 c and CD1 8 to PMN was analyzed by FCM to determine the expression of members of the CD 11 /CD 18 family on normal and patient calf PMN. The

fluoresence

intensity

(log1 0)

Fig. 4. Ccl1 surface expression of CD I I b, CD I Ic and CD 18 on control and patient calf PMN. Esprcssion ofCDI I b. CD1 Ic and CD18 on PMN from control (-) and affcctcd calf (“‘) b) How cytometry using staining with conjugalc only (---). or with murinc m.-\b to the scparatc CD I I /CDI 8 hctcrodimcrs. Panels C. B, and A show the protilcs obtained using anti-CD1 1b mAb (IL-.Al5). and anti-CD1 Ic m.-\b (N.\MJ). Lind anti-CDIS mAb tRl5.7). rcspcctivcly. The/I subunit as well as the (Y subunits of Mac-I and ~150.95 arc lacking from the PMN of the nffcctcd calf.

W.E. BERNAD1N.A

304

ET .AL.

profiles obtained using IL-A 15 (anti-CD I 1b), NAM4 (anti-CD 11 c), and R 15.7 (anti-CD 18 ) are shown in Fig. 4, Panels C, B and A, respectively. Thus the calculated mean percentage of healthy control bovine PMN expressing CD1 lb, CD1 lc and CD1 8, in this order, was found to be 86.7%, 78.0%, and 89.9% whereas for the affected calf these values were 0.4%, O.l%, and 2.1%, respectively. Similarly, the mAb to human CD1 8 or CD1 la failed to bind to any PMN ( 52%) from Calf B, but did bind to equal numbers of PMN from normal controls (including two also used in this study) as did mAb R 15.7 and IL-A99 (data not shown ).

Family history studieson PMN CD1 I /CD1 8 heterodimer cypression Neutrophil CD 11 c expression using FCM and/or PMN ing Con A binding procedure were determined for a panel of (nearly) all near of kin to Calf B and control animals sults indicate that four calves (notably B, and 368, and

Neutrophil CDI l/CD18 nine animals near of kin Animal to Calf

and

kinship

B’

hctcrodimer to Calf B Health status

csprrssion’

ofCalf

WBC’ (X 103&‘)

B. three

healthy

CD

I I /CD

Mac- 1 contents usof cattle consisting (Table 2). The retwo calves marked

cattlc

no kin

I8 csprcssion

NAM4(anti-CDI Control I (none) Control II (nom) Control III (none) B (=calfB) A ( 100% ) c (100%) M ( = mother) 1000 (=aunt) 368 (75%) 365 (75%) MR=369 t (75%)

(75%)

t (75%) ‘Data PMN

healthy health) healthy rec. inf healthy del. health healthy health) rec. inf healthy healthy euthanJ euthan’

ND 8.6 46.2 8.6 ND 6.9 36.5 8.6

to lack of bovine ND, not done.

CD I I /CD

9.4 2 so.o* L loo*

I I-reactive

mAb

Con-\-binding ++ ++ ++ ND ND ND ND ND

50.4 I.2 64.’ 44.9 ND’

ND ND

ND5 ( - ) of CD I I b by immunoblotting Ic m.Ab (NAMJ) by FCM.

‘WBC values as determined or *deduced ( = WBC counting paquc procedure) at day of evaluation of the blood sample. ‘See Fig. I for both the animal symbols used and the familial ‘Euthanasia (euthan) done on the animals due to recurrent to therapy. ‘Owing health:

Ic)

0.1 69.’ 71.5 71.5

‘0.0

are presented as prcscnce ( + + ). or abscncc lysates, or as % PMN reactive with anti-CD1

using

75.0’ 73.9 71.7

7.9

to Cal B. and

following

I’MN

relationships infections

(xc.

at the time

ofsample

proccdurc

isolation

by Ficoll-lso-

of tcsl cattlc. inf ) and poor taking.

dcl. health,

rcsponsc dclicatc

on

LEllKOCYTE.~DHESION

DEFICIENC\.

305

(t ) had a hyperleukocytosis with no left shift (not shown) in conjunction with no PMN CD1 1 expression, i.e. the BLAD syndrome. These four calves were born to two halfsisters following the mating of each with the same bull (their uncle, see Fig. 1 ). Note that Calf C, a full brother of Calf B (with BLAD) has a delicate health, a substantial leukocytosis, but a CD1 lc expression on its PMN similar to that of its mother and Sister A. In contrast, Cow 1000, a clinically healthy (presumed) carrier of the BLAD gene has a much lower PMN CD1 1 expression. Note also that a healthy son (MR=369) born to Mother 1000 has a PMN CD 11 expression of approximately 60% that of normal healthy controls. DISCUSSION

The present investigation was stimulated when a Dutch calf with a family tree indicating strong inbreeding within a BLAD-gene-carrying family presented clinically with ‘LAD-like’ symptoms. To date, the potentially devastating BLAD syndrome has unequivocally been diagnosed in the USA (Kehrli et al., 1990) and the Netherlands (this report). Retrospective studies (Kehrli et al., 1990) however, suggested an earlier occurrence in the USA (Hagemoser et al., 1983) and Japan (Takahashi et al., 1987) of apparent BLAD cases. Although many findings in the present study were consistent with already reported data, several other findings were not. Five new observations were made. Firstly, the well-documented (Springer et al., 1984; Giger et al., 1987; Kehrli et al., 1990) hallmark of the illness, i.e. hyperneutrocytosis without appreciable left shift was demonstrated for 8 months, in the affected calf. Thus we obtained information for the first time on the longevity of the neutrocytosis generated in vivo in an animal with the BLAD affliction. Additionally, it was observed that the hyperleukocytosis persisted during fever-free and apparently infection-free periods. Consonant observations have also been reported from humans and from the only dog with LAD described so far, thus underlining the unitary explanation of the disease. Secondly, we observed a total lack of leukocyte integrin expression on the PMN of Calf B using mAb to CD 18 and the separate CD 1 la, CD 1 lb, and CD 11 c/CD 18 heterodimers. Given the unequivocally proven restricted specificity of the mAb used, it may then be concluded that lack of leukocyte integrin expression observed was not an artefact of either PMN isolation or PMN lysate preparation. In interpreting these data, however, one should consider the possibility that because of the underlying CD1 8 gene defect(s), conformationally changed cy subunit (CD 1 1 ) molecules, no longer detected by the mAb used, may have been produced. Whether this was the case in the calf studied, is unknown. Based on simple logic however, it is very unlikely that the gene defect would have caused such alterations in (assumedly expressed) CD 1 1/CD 18 molecules for these to be undetectable using four different mAb.

306

W.E. HEKNADIN..\

ET :\L.

Furthermore, our Mac-l analysis data seemed to confirm that there was indeed no detectable CD 1 I, notably CD 1 1b on PMN of the calves with BLAD described here. Thirdly, the homotypic adhesion assay enabled us to determine that adhesive function of the PMN from Calf B correlated to lack of leukocyte integrin expression on its PMN. Studies using, for example. adherence to Petri dishes to evaluate PMN capacity for emigrating from blood (an established integrindependent process (Anderson and Springer, 1987 ) are unable to support this conclusion because of failure of the system to distinguish between integrinpromoted and serum fibronectin-promoted (Kerr et al., 1983) adhesion of PMN. In contrast, because PMN must express LFA- 1 for homotypic adhesion to occur (Marlin and Springer, 1987; Makgoba et al., 1988), homotypic adhesion assays are able to specifically detect leukocyte integrin molecules. Using this method we have obtained data indicating that the affected calf’s PMN lacked functionally active leukocyte integrin molecules. Although the present study could not establish whether Calf B indeed had aberrant amounts of extravascular PMN, our homotypic adherence data, together with the calf’s persistent mature neutrophilia suggest that emigration of PMN from the blood did not occur. Fourthly, in contrast to the finding of Hagemoser et al. ( 1983) showing that in bovine granulocytopathy the number of monocytes increases with time, we hardly detected any (obvious) monocytosis during the &month evaluation period of our BLAD-affected calf. The functional relevance of this phenomenon, if any, is presently unknown. In considering that Calf B’s persistent leukocytosis predominantly consisted of a mature neutrophilia with, at times, variable eosinophilia and lymphocytosis, our data suggest a possible emigration of monocytes from the circulation. It has previously been demonstrated in man that monocytes (in contrast to PMN) may emigrate through endothelial cells (Rice et al., 1990) by a CD 1 1 /CD 1&independent adhesion mechanism. Whether this also occurred in the BLAD case presented here is unknown. In vivo. or postmortem studies on transendothelial emigration of labeled CD 1 1 /CD 1S-deficient monocytes will be necessary to address this question. Fifthly, we observed an unequivocal familial incidence of the BLAD syndrome (see Fig. 1, and Table 2 ). In man, intrinsic cellular defects (e.g. enzymes essential to cell metabolism and molecular abnormalities in cellular expression products) generally are inheritable disorders. With respect to human LAD, a familial incidence of the affliction has been observed, albeit that this was not invariably so. In bovine LAD there is (strictly speaking) at present only strong circumstantial evidence to indicate that BLAD also follows a familial pattern. That is, Kehrli et al. ( 1990) while charting all BLAD cases reported world-wide, very convincingly tracked down a common ancestor bull in the pedigrees of all diseased calves. As five of the nine affected calves in

LEUKOCYTE

ADHESION

DEFICIENCY

307

this construct were half-siblings, a concept of BLAD as a diseaseshowing a familial pattern was apparently established. However, this concept becomes blurred when eachreported BLAD caseis analyzed for familial incidence. For example, 100% of the American calves and up to 30% of the Japanesecalves with BLAD do not have a family history of the disorder (Kehrli et al., 1990). Although each of live JapaneseHolstein cows that were mated with the same bull have had one sibling that has been affected, such data are, in our view, still insufficient to confirm the statement that BLAD follows a familial pattern. Obviously, if BLAD, a recessively inherited disease,is to occur in five calves born to five different mothers, theselatter animals must all be obligate carriers of the BLAD gene. This question, we believe, has remained inadequately answeredby the Kehrli et al. ( 1990) report. The present report is the first to demonstrate clearcut familial BLAD. That is, in four calves that, in terms of interrelationships, were full-siblings (n=3) and a 75%-siblings ( n = 1). Thus our ‘epidemiologic data’ seemto justify, we believe, an emphasis on an intrinsic leukocyte defect as an explanation of the BLAD syndrome. Unexpectedly, a familial clustering was also observed for cattle with ‘belownormal’ PMN CD 11/CD 18 molecule expression.For example, Cow 1000, a clinically healthy (obligate) carrier of the BLAD gene,with low PMN CD 11c expression (50.4%) had a clinically healthy son MR=369 with a similar CD 11c expression (44.9%) on its PMN (Table 2). Further evaluation studies of CD 11/CD 18 heterodimer expressionare neededto determine whether the results would distinguish healthy calves from calves predisposedto BLAD development. In man, such distinction has been claimed for the LAD syndrome (Anderson et al., 1985). Although Kehrli et al. ( 1990) reported the presenceof remarkably similar amounts of neutrophil surface CD 18 among putative heterozygotesfor the BLAD trait, our, as yet, few data seem to warn against the universality of this contention. In conclusion, the clinical and other similarities amongst the calves with BLAD describedhere have clearly reconfirmed the unitary explanation of the leukocyte adhesion disorder. Given also the important similarities between human LAD and our bovine LAD cases,the latter appear excellent models for understanding species survival strategies in LAD (or in BLAD, as applicable). REFERENCES Anderson, D.C. and Springer, T.A., 1987. Leukocyte adhesion deficiency: an inherited defect in the Mac-l, LFA-1, and pISO, glycoproteins. Annu. Rev. Med., 38: 175-194. Anderson, D.C., Schmalstieg, F.C., Finegold, M.J., Hughes, B.J., Rothstein, R., Miller, L.J., Koh, S., Tosi, M.F., Jacobs, R.L., Waldrop, T.C., Goldman, A.S., Shearer, W.T. and Springer, T.A., 1985. The severeand moderate phenotypes of heridable Mac- 1, LFA- 1 deficiency. Their quantitative definition and relation to leukocyte dysfunction and clinical features. J. Infect. Dis., 152: 668-689.

308

W.E. BERNADINA

ET AL.

Crowley, C.A., Curnutte, J.T., Rosin, R.E., Andre-Schwartz, J., Gallin, J.I., Klempner, M., Snijderman, R., Southwick, F.S., Stossel,T.P. and Babior, B.M., 1980. An inherited abnormality of neutrophil adhesion: its genetic transmission and its associations with a missing protein. N. Eng. J. Med., 302: 1163-I 168. Entman, M.L., Youker, K., Shapell, S.B., Siegel, C., Rothlein, R., Dreyer, W.J., Schmalstieg, F.C. and Wayne Smith, C., 1990. Neutrophil adherence to isolated adult canine monocytes: evidence for a CD lldependent mechanism. J. Clin. Invest., 85: 1497-1506. Giger, U., Boxer, L.A., Simpson, P.J., Lucchesi, B.R. and Todd, III, R.F., 1987. Deficiency of leukocyte surface glycoproteins MO 1, LFA- 1, and Leu MS in a dog with recurrent bacterial infections. Blood, 69: 1622- 1630. Hagemoser, W.A., Roth, J.A., Lofstedt, J. and Fagerland, J.A., 1983. Granulocytopathy in a Holstein heifer. J.A. Vet. Med. Assoc., 183: 1093-1094. Kehrli, M.E., Schmalstieg, F.C., Anderson, D.C., van der Maaten, M.J., Hughes, B.J., Ackermann, M.R., Wilhelmsen, C.L., Brown, G.B., Stevens, M.G. and Whetstone, C.A., 1990. Molecular definition of the bovine granulocytopathy syndrome: Identification of deficiency ofthe Mac-l (CD1 Ib/CD18) glycoprotein. Am. J. Vet. Res., 15: 1826-1836. Kerr, M.A., Falconer, J.S., Bashey, A. and Swanson Beck, J., 1983. The effect of C3 levels on yeast opsonization by normal and pathological sera: identification of a complement-independent opsonin. Clin. Exp. Immunol., 54: 793-800. Kishimoto, T.K., Hollander, N., Roberts, T.M., Anderson, D.C. and Springer, T.A., 1987. Heterogenous mutations in the /3subunit common to LFA- 1, Mac- 1, and p 150,95 glycoproteins cause leukocyte adhesion deficiency. Cell., 50: 193-202. Lawrence, M.B., Smith, C.W., Eskin, S.G. and McIntire, L.V., 1990. Effect of venous shear stresson CD1 8-mediated neutrophil adhesion to cultured endothelium. Blood, 75: 227-237. Makgoba, M.W., Sanders, M.E., Ginther Lute, G.E., Dustin, M.L., Springer, T.A., Clark, E.A., Mannoni, P. and Shaw, S., 1988. ICAM-I a ligand for LFA-I dependent adhesion of B, T and myeloid cells.Nature, 331: 86-88. Marlin, S.D. and Springer, T.A., 1987. Purified intercellular adhesion molecule-l (ICAM- ) is a ligand for lymphocyte function-associated antigen- 1 (LFA-I ). Cell, 55: 8 13-8 19. Rice, G.E., Munro, J.M. and Bevilacqua, M.P., 1990. Inducible cell adhesion molecule 110 (INCAM- 10) is an endothelial receptor for lymphocytes: A CD 1 I /CD1 g-independent adhesion mechanism. J. Exp. Med., 171: 1369- 1374. Rothlein, R. and Springer, T.A., 1986. The requirement for lymphocyte function-associated antigen-l in homotypic leukocyte adhesion stimulated by phorbol ester. J. Exp. Med., 163: 1132-l 149. Smith, C.W., Marlin, S.D., Rothlein, R., Toman, C., Anderson, D.C., 1989. Cooperative interactions of LFA-1 and Mac-l with intercellular adhesion molecule-l in facilitating adherence and transendothelial migration of human neutrophils in vitro. J. Clin. Invest., 83: 20082017. Splitter, G. and Morrison, W.I., 1991. Antigens expressed predominantly on monocytes and granulocytes: identification of bovine CD 11b and CD1 lc. Vet. Immunol. Immunopathol., 27: 87-90. Springer, T.A., 1990. Adhesion receptors in the immune system. Nature, 346: 425-434. Springer, T.A., Thompson, W.S., Miller, L.J., Schmalstieg, F.C. and Anderson, D.C., 1984. Inherited deficiency of the Mac-l, LFA- 1, p 150,95 glycoprotein family and its molecular basis. J. Exp. Med., 160: 1901-1918. Takahashi, K., Miyagawa, K., Abe, S., Kurosawa, T., Sonoda, M., Nakade, T., Nagahata, H., Noda, H., Chihaya, Y. and Isogai, E., 1987. Bovine granulocytopathy of Holstein Friesian calves and heifers. Jpn. J. Vet. Sci., 49: 733-736. Young, C.W., Bonczek, R.R. and Johnson, D.G., 1988. Inbreeding of and relationship among registered Holsteins. J. Dairy Sci., 7 1: 1659- 1666.