use of densitometric analysis for interpreting hsv serologies based on western blot

159

JVM (10657

Use of densitometric analysis for interpreting HSV serologies based on Western blot

Rhoda L. Ashley and Julie ~ilitoni

Departmenr of Laboratory Medicine, University of ~~shj~g~o~, Seatrle. WA. U.S.A

(Accepted 8 July 1987)

Summary

Western blot assays have been described for detecting antibodies to herpes simplex virus (HSV). A predominance of antibody binding to either the HSV-1 or the HSV-Z-cont~ning blot has been reported to indicate infection with HSV-1 or HSV- 2, respectively. By densitometry, differential binding of total antibody on HSV-1 versus HSV-2 strips can be expressed as a ratio. To determine the clinical corre- lation of these ratios, sera from 81 patients with culture-proven oral or genital herpes were tested by Western blot (WB) and densitometry. Binding ratios ac- curately identified patients with primary HSV-2 infections. However, ratios on sera with HSV-1 antibody or dual antibody status showed considerable overlap. Den- sitometry was shown to amplify and clarify the band corresponding to the HSV-2 specific glycoprotein gG-2 and was useful, in this respect, for detecting HSV-2 antibody in the presence of HSV-1 antibody. Sera from 52 patients with asymptomatic HSV-1, HSV-2 or dual infection were also tested by WB. Typing results were confirmed by cross-adsorption of sera (“adsorption blot assay”). Ratios of HSV-2 to HSV-1 binding were higher in asymptomatic versus symptomatic patients with dual antibody (P<O.Ol). Ratios for those with HSV-1 or HSV-2 antibody types were not affected by disease expression,

Herpes simplex virus; Serology; Western blot ---_..-- -__..__ --______

Supported in part by grant AI-20381

Correspondence fo: R.L. Ashley, Department of Laboratory Medicine, University of Washington, Se- attle, Washington, U.S.A.

0166-0934/87/$03.50 0 1987 Elsevier Science Publishers B.V. (Biomedical Division)

160

introduction

Western blot (WB) technology has proven useful for the purpose of definitively typing serum antibodies to HSV (Bernstein et al.. 1985a: Sherlock et al., 1986).

Results of WB appear to be more accurate than assays such as microneutralization or complement fixation which are based on biologic activity (Bernstein et al., 1984,

1985b; Lee et al., 1985; Sherlock et al., 1986). These studies have also shown that type specificity can usually be determined from HSV Western blot assays (WBA} by the predominance of apparent antibody binding to either the HSV-i-c~)ntaining “blot” (HSV-1 serotype) or the HSV-~-containing “blot“ (HSV-2 serotype). Den- sitometric analyses have been suggested as a means of defining objective criteria

for the presence of HSV-1 only, HSV-2 only, or both HSV-I and HSV-2 antibodies (Bernstein et al., 19XSb; Lentinen et al., 1985). However. few sera from patients with well defined clinical illness have been studied using densitometry.

This study was designed to establish accurate densitometric criteria for type specific HSV antibodies utilizing sera from patients with culture-proven oral or genital herpes.

Methods

Antigen prepur~t~~F1 To prepare antigen for WBs, confluent monolayers of human embryonic tonsil

cells (HET) were infected with HSV-I (E-115) or HSV-2 (333). When cytopathic effect was noted in >80% of the monolayer, cells were rinsed, then scraped into cold PBS and centrifuged at 4°C for 10 min at 500 x g. Pellets were resuspended in 2x SPS (0.125 M Tris, pH 6.8, 20% glycerol, 8% sodium dodecyl sulfate, 10% 2-mercaptoethanol, 0.025% Bromophenol Blue) before freezing in 250 ~1 ali-

quots. To prepare antigen for the immunosorbent for the adsorption blot assay, HSV-

1 or HSV-2-infected cells were harvested by agitation with sterile glass beads, then pelleted and suspended in 15 ml per bottle of coupling buffer (0.02 M NaH,PO,. 0.15 M NaCi, 1% NP40, and 0.01% thimerosal adjusted to pH 8.3 with NaOH). After sonication, cells were rocked overnight at 4°C and antigen was added to pre- swollen and washed cyanogen bromide-activated sepharose 4B (Sigma, St. Louis. MO) as described in detail previously (Sherlock et al., 1986). Briefly, the mixture was held overnight at 4”C, then centrifuged at 400 x g for 5 min and incubated with 0.2 M glycine (pH 8) for 2 h at room temperature. The sepharose was then washed twice each time with coupling buffer and acetate buffer (0.5 M NaCl in 0.1 M sodium acetate, pH 4.0) alternately, then 3 times with TST (0.9% NaCl and

0.01% thimerosol in 50 mM Tris, pH 7.5).

Preparation of Western blots HSV-infected cell lysates were thawed and then boiled for 2 min before loading

onto 1.5 mm discontinuous 9% poiyacrylamide gels containing 0.1% SDS and cross-

161

linked with 2.6% bis-acrylamide (Ashley et al., 1985). Proteins were electro-

phoresed at 20 mA for 90 min, in a Hoefer Mighty Small II vertical slab gel sys-

tem. Gels were equilibrated for 30 min in transfer buffer (25 mM Trizma base, 192

mM glycine, 20% methanol and 0.05% SDS). Proteins were transferred to nitro- cellulose (NC) at 200 mA for 2 h in a Hoefer Transphor chamber. The NC was then cut into 3 mm wide strips and stored at 4°C.

Western blot ussay NC strips were rocked in PBS-Tween (0.05% Tween 20 in PBS) for 30 min, then

incubated overnight at room temperature with human serum diluted 1:lOO in Blotto (5% nonfat dry milk, 0.01% thimerosal, 0.01% antifoam A in PBS). Each serum sample was incubated with a NC strip with HSV-1 antigens and a NC strip with HSV-2 antigens. Strips were washed 3 times for 5 min each with PBS-Tween, then rinsed with PBS. Strips were rocked for 70 min at room temperature with perox- idase conjugated anti-human IgG (Boehringer Mannheim, Indianapolis, IN) diluted 1:1650. Strips were washed 3 times with PBS-Tween then rinsed with TBS (Tris buffered saline; 500 mM NaCl in 20 mM Tris base, pH 7.5). The strips were

stained with 4-chloro-1-naphthol substrate containing solution A (20 ml cold methanol and 60 mg 4-chloro-1-naphthol) and solution B (60 l.~l cold 30% H,O, and 100 ml TBS) which were prepared and mixed just before use. After 10 min strips were rinsed with water and allowed to dry.

Densitometry Densitometry was performed on a Hoefer GS300 scanning densitometer with a

Cole Parmer Model 8376-20 recorder. The tracings were cut out and weighed. The mass of the tracing from the HSV-2 antigen strip was divided by the mass of that from the HSV-1 antigen strip to establish the “2:l” ratio.

Adsorption blot assay Each serum sample was diluted 1:20 in TST and 400 l.~l was added to each of

two 1.5 ml microcentrifuge tubes. One ml of prepared HSV-1-sepharose immunosorbent was added to one tube and 1 ml of HSV-2-sepharose immunosorbent was added to the other. After rocking overnight at 4”C, mixtures were centrifuged for 30 s. Supernatants were decanted and adjusted to 2 ml with Blotto.

Adsorbed sera (1 ml) were reacted with NC strips containing HSV-1 antigens and HSV-2 antigens, respectively. Strips were developed as described above.

Results

Serum sources Sera from 15 patients with culture-proven primary HSV-1 genital infections were

tested; 10 were convalescent sera drawn 14-40 days after onset of symptoms and 5 were drawn more than 2 months post-onset. In addition, 9 sera from culture-

proven cases of HSV-1 oral infections were tested. Convalescent sera from 39 patients with culture-proven HSV-2 infections were tested, 14 from patients with primary HSV-2 genital infections and 25 from patients with recurrent genital HSV-2

episodes. Of these 25 patients, 10 had been followed since their primary genital

episode, and 15 presented with recurrent episodes. Nine patients with first episode non-primary HSV-2 genital infection were also

studied. These patients all had HSV-2 cultured from genital lesions and had HSV-

1 antibody in their acute phase sera as determined by adsorption blot. All scro- converted to IHSV-2 and had HSV-1 plus HSV-2 antibody in convalescent sera drawn 1.5-44 days after onset of symptoms. Sera were also analyzed from 9 patients with both HSV-1 and HSV-2 antibodies who presented with culture-proven recurrent HSV-2 genital infections.

Sera from 52 patients without signs or symptoms of HSV infection were also

typed by WBAr 15 had HSV-f antibody, 19 had HSV-2 antibody, and I8 had both HSV-1 and HSV-2 antibodies. These serotypes were all confirmed by adsorption

blot.

Densitometry rutias: culture-proven HSV-I

Densitometry ratios were expressed as the total density of the antibody on the HSV-2 strip divided by the total density of the antibody on the HSV-1 strip (“‘2: 1” ratio). The 2:l ratios for patients with culture-proven HSV-1 oral infections ranged from 0.13-0.88 (median = 0.31: Fig. 1). Ratios ranged from 0.23-1.57 (median = 0.69) for those with primary genital infections (Fig. 1). The 2:l ratios for the 10 sera drawn close to the primary episode (Fig. I; closed circles) did not differ from the 5 sera drawn 58-155 days post-onset (Fig. I, open circles) with ranges of 0.2LtI.57 and 0.28-l SO, respectively.

Densitometry ratios: culture-proven HSV-2

Sera from patients with proven HSV-2 primary genital episodes (,z==14) were easily distinguished from sera with HSV-I antibodies by densitometry ratios (Fig. 1). The median ratio for this group was 4.02 (range 1.82-5.81). The 2.5 culture- proven recurrent genital herpes patients also had high ratios (2.2Ck11.53; median = 3.4). Ratios in this group were slightly lower in patients presenting within 1 yr of their primary episode (2.2tM.47; median= 2.97; Fig. 1, closed circles) than for

those who had histories of greater than 1 yr (2.74-11.53; median = 3.48; Fig. I, open circles). These sera were thus easily typed by predominance of antibody binding to the HSV-2 blot.

Densitometry ratios: dual antibodies To analyze the effect of HSV-2 seroconversion on densitometry ratios of sera

which also contain HSV-1 antibody. 9 sets of sera (acute and convalescent) from patients with culture-proven non-primary first episode (“initial”) genital HSV-2 infection were subjected to WBA and adsorption blot assays. It was anticipated that HSV-2 seroconversion in HSV-1 seropositive patients would be evident by a shift in densitometry ratio. All 9 pairs of sera were run on blots derived from the same

163

11

9

7

5.8

5.4

5

2.6

SYMPTOMATIC ASYMPTOMATIC

0

il a

2.2 8 l

1.8 . l

* 1.4

8

. :

.

L

t . l :

.

1 -v-

i T IF-,

I 1

l-W-1 HSV-2 HSV-

l-2

PATIENT GROUPS

Fig. 1. Densitometry ratios from sera from patients with symptomatic or asymptomatic HSV infec-

tions. Sera were analyzed by WB and all blots were subjected to dens~tometry. Ratios of the antibody

binding to the HSV-2 blot to the antibody binding to the HSV-I blot were calculated for each serum.

o = values from patients who were seen 2 or more months after a documented primary first genital

episode with HSV-1 or who had histories of at least 1 yr of recurrent genital HSV-2. The serostatus of

each patient as determined by WB and confirmed by adsorption blot assay, is given at the base of each

bar. All symptomatic patients had positive cultures for HSV-1 or HSV-2 at the time serum was drawn. Asymptomatic patients had neither positive cultures nor histories of oral or genital infections.

preparative gels and all 9 showed an increased ratio between acute and convalescent sera (Table 1). The extent of the change was not related to the length of time between the acute and convalescent sera. As a group, the 9 convalescent sera from patients with inital HSV-2 infections had 2:l ratios ranging from 0.34-1.66; median = 0.72 (Fig. 1).

Nine sera from HSV-2 culture-proven recurrent episodes were also found to have both HSV-1 and HSV-2 antibodies by WBA and adsorption blot assays. Densi-

TABLE 1

Effect of HSV-2 scroconveraion on densitometry ratios in patients with pre-euistlng HSV-1 antibody

Patient No. dav\ between acute ,’ and cow. sern

Densitomctry ratio

acute cone

Change in

ratio

28

12

37 30

22 11 20

17

I5

0.31 0.39

I .o.? 0.56 0.21 0.23 0.24 0.8X

0.45

0.77

I .3x I M, 0.8X 0.33 0.30 0.47

1.25

0.54

0.31 0.99

0.63 0.32 0. I3 0. Ih

0.23 0.37

0.09

Acute and convalescent (conv.) sera from patients with first episode non-primary genital HSV-2 infec-

tions were suhjectcd to WB and densitumetry. Results on all 18 sera were contirmed by adsorption blot

assay. All acute sera had HSV-I antibody only \vhile all convalescent sera had HSV-I plus HSV-2 an-

tibody.

tometry ratios on these sera were very similar to those obtained from the convalescent sera from non-primary first episode genital HSV-2 patients (Fig. 1). Ratios ranged from 0.33-1.48 (median = 0.71).

In summary, analyses of patients with culture-proven HSV infections indicated that type common antibody responses were typical of HSV-1 patients (resulting in relatively high 2: 1 ratios, in some cases). In contrast. patients with culture-proven HSV-2 primary genital infections had predominantly type-specific antibody responses reflected in at least twice as much staining of the HSV-2 strip as the HSV- 1 strip. Ratios of those with culture-proven HSV-2 genital herpes and dual antibody were indistinguishable from ratios from patients with primary HSV-1 infec-

tions.

Asymptomatic infections Sera were tested from 52 patients with neither symptoms nor histories of herpes

infections; 15 had HSV-1 antibody, 19 had HSV-2 antibody, and 18 had dual antibody by adsorption blot. The 2:l ratios for the asymptomatic HSV-1 group ranged from 0.24-1.07 (median 0.50; Fig. 1). These values were very similar to those obtained with symptomatic patients with genital or oral HSV-1 infection (Fig. 1).

The 19 sera from patients with asymptomatic HSV-2 infections had ratios ranging from 2.37-12.6 (median 4.79). As a group, these values were comparable to those obtained with sera from patients with symptomatic genital HSV-2 infections.

In contrast, sera from 18 asymptomatic patients with dual antibody had higher 2:l ratios than did symptomatic patients with first episode genital HSV-2 and dual antibody: 0.76-5.04 (median 1.44) compared with 0.34-1.66 (median 0.72; Table 2). Asymptomatic patients also had significantly higher ratios than those with dual antibodies and symptomatic recurrent genital HSV-2 infections (Table 2).

165

TABLE 2

Densitometry ratios of sera with dual antihody (HSV-1 and HSV-2)

Clinical presentation

Asymptomatic

Genital HSV-2

first episode

recurrent enisode

n Ratio Median

1X 0.76-5.04 1.44

Y 0.34- 1.6h” 0.72 Y 0.33-1.4X” 0.71

I’ P<O.Ol by Wilcoxon rank sum test for comparison between symptomatic and asymptomatic patients.

. l :. : . : .-*.:: .

. : . : : .

~

:

:*.

U”*** . . . . l .* . . . . . . . . Fig. 2. Distinguishing HSV-1 from dual antibody by detection of the HSV-2 specific gG-92 peak. Sera

with HSV-1 antibody (...) or HSV-1 and HSV-2 antibody (-) were reacted with HSV-1 and HSV-2

WBs and 2: 1 densitometry ratios were calculated: 0.87 for the HSV-1 serum and 0.95 for the dual an-

tibody serum. The gG-92 peak is clearly apparent on the tracing from the serum containing HSV-2

antibody.

Serotypirzg by densitometry ratio Sera with only HSV-2 antibody had sufficiently high 2: 1 ratios to distinguish this

serotype from HSV-1 or dual antibody serotypes (Fig. 1). However, ratios from many sera with dual antibody were indistinguishable from the ratios of sera with

HSV-1 serotypes. For example, 11 of 39 (28%) sera with HSV-1 serotypes had ratios between 0.5 and 1.60; a range which included 67% (24136) of sera with dual

antibodies. Densitometry proved useful in a qualitative sense by detecting gG-2, an HSV-

2 type specific glycoprotein which migrates at approximately 92 kDa (Marsden et

al., 1984). Tracings from 2 patients, 1 with culture-proven HSV-1 oral herpes (dotted line) and 1 with culture-proven HSV-2 initial genital herpes and prior HSV- 1 antibody (solid line) are shown in Fig. 2. The HSV-1 patient had a 2:l ratio of 0.87 while the HSV-2 patient had a 2:l ratio of 0.95. When tracings from the HSV- 2 blots are compared, the gG-2 band is clearly evident only on the blot reacted with serum from the patient with HSV-2 disease.

Discussion

WB analysis provides a means of definitively serotyping HSV antibodies. Den- sitometry analyses of sera from patients with culture-proven herpes infections confirmed that, in most cases, HSV-specific antibodies bind preferentially to fixed, denatured homotypic proteins. However, sera with HSV-1 antibodies were diffi- cult to distinguish from sera with dual antibodies on the basis of 2:l ratio alone. For example, 6 of the 15 sera from patients with primary genital HSV-1 infections had ratios which fell above 1 .O, a value reflecting equal staining on the HSV-2 and HSV-I blots. The possibility exists that at least some of these sera contained HSV- 2 antibody as well as HSV-1 antibody resulting in spuriously high ratios. However. these patients appeared to be true primary cases as their acute phase sera were seronegative by WBA. Further. the sera were confirmed as “HSV-1 only” by adsorption blot assay. Thus, it appears unlikely that unrecognized dual antibody status accounts for the overlap in values between “HSV-1 only” sera and sera with

dual antibody. HSV-2 specific sera were more “type-specific” than HSV-1 sera as reflected in

high 2:l ratios with little overlap between HSV-2 ratios and ratios derived from dual infections. WB analyses of cross-adsorbed rabbit sera identified 14 HSV-2 and 12 HSV-1 proteins with type-specific epitopes (Bernstein et al., 1985a). Adsorbed human sera have also been shown to react in a type-specific manner with a larger subset of HSV-2 proteins than HSV-1 proteins (Bernstein et al., 1985a), a finding which our data support.

To clarify the parameters of dual antibody sera further, 2: 1 ratios were obtained on acute and convalescent sera from patients with prior HSV-1 antibody and first episode genital HSV-2 infections. Densitometry ratios changed comparatively little after HSV-2 seroconversion in patients with pre-existing HSV-1 antibody (Ta- ble 1). The most likely explanation for the small change in ratio is that patients

167

mounted an anamnestic response to type-common HSV antigens thus boosting their

titers of antibodies to HSV-1. This phenomenon has been described with other types of assays for HSV antibody (McClung et al., 1976; Smith et al., 1972). Inspection of the blots from acute and convalescent sera support this hypothesis: both HSV- I and HSV-2 blots from convalescent sera were markedly increased in total staining compared with the respective blots from acute sera.

Detecting HSV-2 specific antibody in the presence of HSV-1 antibody is crucial for accurate seropidemiologic studies or for clinical diagnosis of past HSV-2 in-

fection. While our data suggest that densitometry ratios alone cannot suffice to

identify sera with dual antibody, the technique is important in amplifying the HSV- 2 gG band (Marsden et al.. 1984). This protein functions as a sensitive marker for HSV-2 antibody regardless of the presence of co-existing HSV-I antibody (Ashley

et al., 1984; Lee et al., 1985). Sera from patients without histories or symptoms of genital herpes were also

analyzed: 15 had HSV-1, 19 had HSV-2 and 18 had dual antibody. Patients with asymptomatic HSV-1 or HSV-2 infections were indistinguishable from those with the respective symptomatic infections with respect to binding ratios. However, sera from patients with asymptomatic dual infections had strikingly higher 2:l binding ratios than those sera from patients with symptomatic HSV-2 infections and dual antibodies. Symptomatic HSV-2 patients, therefore, tend to have relatively more antibody binding to the HSV-I strip. Active infection may result in higher titers

of cross-reactive antibodies because of the repeated presentation of epitopes which are type common.

Judicious application of densitometry can aid in the accurate interpretation of WBs and, when applied as an aid for identifying HSV-2 gG, may prove especially useful in cases of early HSV-2 seroconversion in the HSV-1 seropositive patient. Identification of other predominantly type-specific proteins and elucidation of the time course of the immune response to these proteins will result in further refine- ment of this technique.

References

Ashley. R., Minson, A. and Corey, L. (1984) Human antibody response to a newly described herpes

simplex virus type 2 (HSV-2) glycoprotein. G92. 24th Interscience Conference on Antimicrob. Agents

Chemother. p. Yh.

Ashley. R.L.. Benedetti. J. and Corey, L. (1985) Humoral immune response to HSV-1 and HSV-2

viral proteins in patients with primary genital herpes. J. Med. Virol. 17. 15_%166.

Bernstein, D.I.. Lovett. M.A. and Bryson. Y.J. (1984) Serologic analysis of first-episode nonprimary herpes simplex virus infection. Am. J. Med. 77, 10551060.

Bernstein, D.I., Bryson. Y.J. and Lovett, M.A. (1985a) Antibody response to type-common and type- unique epitopes of herpes simplex virus pol~peptides. J. Med. Viral. IS, 251-263.

Bernstein. D.I.. Garraty, E., Lovett. M.A. and Bryson. Y.J. (1985h) Comparison of Western blot analysis to microneutralization for detection of type specific herpes simplex virus antibodies, J. Med.

Virol. 15. 223-230.

Lee, F.K.. Coleman. R.M.. Pereira, L., Bailey. P.. Tatsuno. M. and Nahmias, A. (1985) Detection of herpes simplex virus type 2-specific antibody with glycoprotein G. J. Clin. Microbial. 22. 641-643.

Lentinen. M., Koivisto. V.. Lentinen. T., Paavoncn, J. and Lcininkki, P. (1985) Immunoblotting and

enzyme-linked immunosorbent assay analysis ol serological responses in patients infected with herpes

simplex virus types 1 and 2. Intervirology 24, 18-25.

Marsden. H.S., Buckmaster, A.. Palfreyman. J.W., Hope, R.G. and Minson. A.C. (1984) Charactcr-

ization of the 92000 dalton glycoprotein induced by herpes simplex virus type 2. J. Virol. SO. 547-54.

McClung, H.. Praleep, S. and Rawls. W.E. (1976) Rclativc concentrations in human sera of antibodies

to cross-reacting and specific antigens of herpes simplex virus type5 1 and 2. Am. J. Epidemiol. 104.

192-201.

Sherlock. C.H.. Ashley. R.L.. Shurtlcff, M.L.. Mack. K.D. and Corey, L. (1986) Type specificity 01

complement-fixing antibody against herpes simplex virus type 2 ACT-4 early antigen. J. Clin. M-

crobiol. 24, 1093-1097.

Smith, J.W.. Adam. E., Melnick. J.L. and Rawls. W.E. (1972) Use of Cr release test to demonstrate

patterns of antibody response in humans to herpesvirus types I and 2. J. Immunol. 109. SS-CS64.

use of densitometric analysis for interpreting hsv serologies based on western blot

Documents