Indian Pediatrics 1999; 36:362-367

A.A. Manohar, M. Williamson, H.A. Kamat, G.V. Koppikar and R.H. Merchant*

From the Department of Microbiology, T.N. Medical College and B. Y.L. Nair Charitable Hospital, Mumbai Central, Mumbai 400 008, India and Department of Neonatology*, Nowrosjee Wadia Maternity Hospital, Parel, Mumbai 400 0 I 2, India.

ReprilJt requests: Ms. A.A. Manohar, A/15, Jai-Kirti Society, Turel Pakhadi Road, Malad (West), Mumbai 400 064, India.

Manuscript received: March 6, 1998; Initial review completed: April 3, 1998;
Revision accepted: November 3, 1998.

Objective: To compare the cellular and humoral factors in colostrum from HIV infected and uninfected lactating mothers. Design: Cross sectional study. Setting: Maternity Ward. Methods: Colostrum was collectedfrom 130 mothers (62 HIV seropositives and 68 HIV seronegatives). These colostrum samples were tested for total cell count, cell viability, differential count, phagocytic activity of macro phages, 'T' cell counts, IgA, IgM and IgG levels. Results: There was a statistically significant decrease in the phagocytosis and 'T' cell number (p <0.001) and in the IgA and IgG levels (p<0.05) in the colostrum obtained from HIV seropositive mothers as compared to HIV seronegative ones. Conclusion: Some of the cellular and humoralfactors are reduced in colostrum samples obtainedfrom HIV seropositives as compared to normals.

Key words: Colostrum, HIV, Immunity, Immunoglobulins..

HUMAN milk is the most appropriate and nutritionally balanced protective food for an infant. Breast milk and colostrum contain significant amounts of cell and humoral factors that protect the neonate from a variety of infections(1) including diarrheal diseases, otitis media(2) and respiratory ilInesses(3). Post partum transmission of HIV through breast milk was first reported in 1985(4). Van de Perre et al.(5) postulated that transmission . of HIV I through breast milk could be favored by the presence of infected cells or by deficiency of anti-infective substances in breast milk or both the factors. This study was therefore carried out with the following objectives: (i) to detect the presence of HIV antibodies in colostrum obtained from HIV seropositive women by a modification of the existing serum ELISA: and (ii) to quantitate the cellular components like total cell count, cell viability, differential count, phagocytosis by macroph- ages and 'T' cell count and to estimate the levels of IgA, IgM and IgG from colostrum of. HIV seropositive and seronegative lactating mothers.

Subjects and Methods

Samples of colostrum were collected from 62 HIV seropositive women (study group) and 68 HIV seronegatives (controls). The age group of these mothers ranged from 18-32 years. All the mothers in the control group were clinically normal with no evidence of malnutrition or mastitis. All of them delivered normally at term. All the 62 seropositive mothers were asymptomatic. Of these, 60 were housewives and 2 were commercial sex workers. All delivered at full term and 10 of 62 were delivered by Caesarean section.

Collection, Transport and Processing

Breast milk samples were collected by manual expression into two sterile siliconized glass test tubes. The tubes were then labelled and transported in an ice-box immediately to the laboratory for processing. Approximately 8- I 0 ml of colostrum was collected into each of the sterile glass test tubes. One tube was centrifuged at 1500 rpm for 20 minutes. The fat layer was decanted and supernatant was used for quantitation of immunoglobulins by Single Radial Immunodiffusion. technique (SRID) and for detection of HIV antibodies by modified serum ELISA. Colostrum sample in the second tube was diluted with Phosphate Buffered Saline (PBS, pH 7.4) after marking the original volume of milk and was centrifuged at 1500 rpm for 20 minutes  to defat it. The pelleted cells were washed in PBS and cell volume was readjusted to the original cell volume. This suspension was used for cellular studies.

Method forVetection of HIV Antibodies from Colostrum by Modified Serum ELISA

The routine ELISA procedure was followed for detection of antibodies to HIV I and II in colostrum with some modification. The usual procedure was diluting 5 III of serum in 250 III of sample diluent and from this 100 III was added to the microtiter wells. Since it is known that the amount of HIV antibodies in colostrum will be less, this modification effectively concentrated the sample in the wells. Hence, 100 III of undiluted colostrum was directly added to the wells. The remaining procedure 'was as follows: The microtiter plate with the samples was incubated at room temperature (RT) for 30 minutes, After incubation, plate was. washed 5 times with wash solution and 100 III of freshly prepared per-oxidase conjugate was added to each well and incubated at RT for 30 minutes. Again plate was washed with wash solution and 100 III of freshly prepared tetramethylbenzidine (TMB) solution was added to each well and incubated at RT for 30 minutes. The reaction was stopped by adding stop solution (1 N H2SO4) and the color developed was read on ELISA plate reader at 450 nm. The kit used for the above procedure was Detect HIV, manufactured by Biochem Immunosystem Inc, Montreal, Canada.

The procedure for all the tests carried out was: (A) Total cell count(6):Estimation of total count was done as per the routine method in hemocytometer without diluting the sample; (B) Cell Viability(6): Equal amount of Erythrocin.B dye (0.4%) and cell suspension was taken. The mixture was loaded into the White Blood Cell (WBC) chamber of hemocytometer. The viable cells exclude the dye and appear colorless whereas the dead cells take up the dye and look pink. This test has to be done within 3 minutes otherwise the live cells take up the stain and look pink; (C) Differential count(6): Smears which were pre- pared from defatted milk were fixed and stained with Wright's stain. The slides were then observed under oil immersion lens. About 6-7 ml of cell suspension was subjected . to Ficoll-Hypaque (density 1.077 ± 0.00 1) density gradient centrifugation. The opaque interface containing macrophages and lymphocytes was removed and washed with PBS. The cell suspension thus obtained was used for estimating the phagocytic potential of macrophages and for enumeration of 'T' cells; (D) Phagocytosis (6): The cell suspension was dispensed into small, polystyrene petriplates and incubated at 3TC for 2 hours in 90% moist atmosphere in a candle jar After incubation the supernatant fluid containing non-adherent cells was poured into tubes contain- ing Dulbecco's Modified Eagle Medium (DMEM, pH7.0) for 'T' cell studies. To the adherent cells, opsonized yeasts (Saccharo- myces cerevisiae) were supplied extraneously and the mixture was incubated again at 3TC in 90% moist atmosphere in a candle jar. For this study an optimal period of 40 minutes was taken. During this period, the yeast cells were taken up by the macrophages which adhered to the petriplates. The mean number of yeasts/ cell was calculated which indicated the phagocytic potential of macrophages; (E) T cell count(6): Equal amount of lymphocytes and sheep red blood cells (SRBC's) were mixed and incubated at 3T C for 10 minutes and then left at 40 C overnight in a refrigerator. Next day the lymphocytes were examined and counted under high power lens of microscope. All cells binding three or more SRBC's were considered as 'T' lymphocytes and this was estimated in percentage.

The data obtained were subjected to statistical analysis. Mean:!: SD was calculated for all the variables in both groups. Student's unpaired 't' test (two tail) and Fischer's Z test were applied.

Results

The colostrum samples obtained from 62 HIV seropositive women were positive for HIV I and II antibodies by the modified serum ELISA. From Table I it is clear that amongst all the cellular factors, phagocytic activity of macrophages and 'T' cell counts showed a statistically significant decrease in the study qroup when compared to the controls. The phagocytic potential of macrophages in the control group ranged from 50-90%, the mean being 70.4%, while in the study group it ranged from 45-70%, with the mean of 57.3% (p <0.001). In the colostrum samples obtained from controls, the 'T' cell counts ranged from 43-55%, the mean being 50.5% whereas in the study group it ranged from 25-50% with the mean of 37.6% (p <0.001 ). Cell viability, total and differential cell count did not show a statistically significant difference between both the groups. Macrophages and Donnes corpuscles together constituted 60-70%, neutrophils were around 20-25% and lymphocytes accounted for 10-12% (70-200 cells/cu mm) of the total milk cells in both the groups. The remaining cells were epithelial and smudge cells. Table II shows that IgA was the pre- dominant immunoglobulin followed by IgM and then by IgG. IgA and IgG values snowed a statistically significant decrease in colostrum obtained from study group as compared to controls (p <0.05). There was no significant decrease in IgM levels when both the groups were compared.

Discussion

This study was carried out to compare the cellular and humoral factors in colostrum from HIV infected and uninfected lactating mothers. A modification of existing serum ELISA was used to detect HIV antibodies in colostrum. Using this modification, we could detect HIV antibodies from colostrum of all 62 HIV seropositive women.

Amongst the cellular factors total cell count, cell viability and differential count in colostrum did not show any statistically significant difference between the HIV seropositive and HIV seronegative mothers. Our results in HIV seronegative women are similar to studies done by various workers(1 ,6-11).

There was a statistically significant decrease in phagocytic activity of marcophages in colostrum obtained from HIV seropositive mothers (57%) as compared to seronegatives (70%). This could be because mononuclear phagocytes are known to be infected by HIV and are also responsible for disseminating HIV infection in the host(12). Goldman and Smith( 13) performed phagocytic studies of colostral cells in normal women and reported 10%,60% and 75% phagocytosis at the end of 15, 30 and 60 minutes, respectively. In our study we found 70% phagocytosis in controls at the end of 40 minutes.
 

TABLE I

Results of Cellular Factors in Breastmilk (11=130)

Groups	Total cell count (cells/cu mm)	Cell viability (%)	Differential count (%)				Phagocytosis (%)	'T' cell count (%)
			M	N	L	Other cells
Study	1596.9	89.47	65.73	22.1	11.27	1.12	57.32	37.56
group	± 428.9	± 7.28	± 3.35	± 3.18	± 2.62	± 1.16	± 7.11	± 5.43
(n = 62)
Control	1690.5	92.51	65.82	22.84	]0.65	0.81	70.38	50.50
group (n=6?)	± 464.3	± 3.69	± 3.36	± 3.31	± 2.38	± 1.0	± 7.19	± 3.16
p value	NS	NS	NS	<0.001	<0.001

M = Macrophages; N = Neutrophils; L = Lymphocytes; n = number of samples; NS = Not significant.
 

TABLE II

Results of Humoral Factors ill Breast milk (Il= 130)
 

Groups	IgA (mg/dl)	IgM (mg/dl)	IgG (mgldl)
Study group (n=62)	241.45 ± I 10.39	73.63 ± 45.88	9.69 ± 7.07
Control group (n=68)	289.34 ± 129.15	75.00 ± 47.86	12.62 ± 7.97
P value	p < 0.05	NS	P < 0.05

NS = Not signil1cant; n = number of samples.


'T' cell count was done by E-rosette assay. Instead of using monoclonal antibodies and fluorescent microscope techniques this assay was done since it was easy to perform, repro- ducible, economical and did not require any sophisticated instruments. The percentage of T lymphocytes from colostrum obtained from normal mothers was similar to that obtained by other researchers(13-15).

The SRID technique was used for quantitation of immunoglobulins because it was simpler, economical and easily standardizable. Compared to HIV seronegative women, the levels of all immunoglobulins were decreased in the study group. Belec et al.(16) have studied IgA levels from breast milk of HIV seronegative controls and HIV seropositive mothers. The IgA levels in their study did not show any statistically significant difference between the two groups. In our study, IgA levels in colostrum were higher than the IgG and IgM levels in both groups. These findings are similar to those obtained by earlier workers(17 -19). There is a paucity of literature on such a study of cellular and humoral properties of colostrum obtained from HIV seropositive mothers.

Since the total number of cells are not de- creased but there is a decrease in the number of 'T' cells with concomitant decrease in phagocytic function and immunoglobulin levels, it could be inferred that there is an immunological defect in 'T' cells, macrophages and 'B' cells. This type of immunodeficiency is similar to that observed in blood when the patient is in an asymptomatic stage where al- though the total cells are within normal limits, their functions become progressively deficient resulting in more and more immunodeficiency.

The results of this study indicate that there is a decrease in some of the cellular and humoral factors in the colostrum of HIV seropositive women as compared to seronegatives. Whether these results could be extrapolated into offering advice to HIV infected mothers on additional or only top (non breast milk) feed to their newborns still remains questionable. However, at our centers . we follow the WHO guideline which recommends breast feeding irrespective of maternal HIV status. Our results will serve as baseline data for further larger studies in order to come to a definite conclusion in this regard. We now plan to undertake these cellular and humoral parameters in colostrum and comparing them simultaneously to the same in blood of the. lactating women.

Acknowledgements

We are grateful to Mr. Arekar, Biostatistician for doing the statistical analysis of the data and are indebted to .the Departments of Pediatrics and Gynecology, B.Y.L. Nair Charitable Hospital for allowing us to obtain the milk samples. We also thank the colleagues in the Department of Microbiology for their support. We also express our gratitude to the Nair Golden Jubilee Research Foundation for funding the entire research.