The response of submandibular gland from the 8-month-old female with an HI titer of 1 : 800 differed significantly from that of the glands of younger adult males with negative HI reactions. Which of the age, sex, or serologic variables was responsible for this response difference cannot be stated. The differences were: (1) a greater tendency for the proliferating epithelial elements to "invade" far out into the gelatin sponge, centrifugally from the explants (figs. 14 and 15); and (2) a greater tendency for the outgrowth to display acinar structures similar to patterns ordinarily displayed by adenocarcinomas (figs. 15, 16, and 17). These acinar structures were present in the absence of any stromal elements and at considerable distances from the explants. They resembled the glandular patterns reproduced by adenocarcinomas cultivated on gelatin sponge by Leighton in some of his original application of matrix methods to tissue culture (78).

Attention was next focused on the response of submandibular glands in early stages of development in embryos. Use was made of the extensive groundwork by Borghese (79, 80) and Grobstein (81, 82) in the culture of submandibular rudiments. Intact submandibular rudiments from C3H/ Bi or (AKR X C3H/Bi)F1 embryos were used at the 13th and 14th day of development; the day of detection of vaginal plug in the mother was counted as day 0. The gross and histologic appearances of these rudiments are shown in figures 18 to 21. Variation in degree of development was noted among the embryos, but, in general, on the 13th day the epithelial portion consisted of a single pear-shaped bud, and on the 14th day it had developed 4 to 15 branches with terminal swellings. The details of development of this organ in culture will not be reviewed here, as they have been accurately documented, and numbered stages have been assigned by Borghese (79, 80) and by Grobstein (81, 82). It is worth noting, however, that in sections of either 13- or 14-day rudiments, the cytologic features of the epithelial and capsular mesenchymal components are remarkably similar. In 14-day rudiments, in particular, proliferation of the epithelium is extremely rapid, as judged by frequency of mitotic figures. If polyoma virus could be assumed to exert its action immediately in rudiments exposed at this time, this action would have to represent a suppression of proliferation rate, as estimated by relative frequencies of mitotic figures in the rudiments and in the tumors in vivo.

Cultures of the intact rudiments were prepared in two ways as indicated in table 3. Those listed as "on or under plasma clot" were prepared by placing the rudiments on the surface of preformed chicken-plasma clot (3 parts chicken plasma and 4 parts chick-embryo extract) in Leighton tubes, or by placing the clotting mixture into the tubes after placement of the rudiments. Those listed as "on plasma clot adjacent to gelatin sponge" were prepared by placing the rudiments on the surface of chicken-plasma clot, adjacent to sponges that had previously been embedded in the clot.

Table 3. Submandibular Rudiment Cultures Studied For Response To Polyoma Virus

Type of culture

Duration of culture period (days)

Age and number of rudiments cultured

13-Day rudiments

14-Day rudiments

Control

Test

Control

Test

On or under plasma clot

9 12 15 17 18 22

1

0 4 0 0 0

2 0 0 0 0 0

0 2 0 6 2 2

0 2 0 6 2 2

Total on clot alone

5

2

12

12

On plasma clot adjacent to gelatin sponge

0 3 0 2 10

0 3 0 2 0

(Contaminated)

3 0 7 4 0

3 0 7 3 0

19

22 27 29 30

2 2 0 0

6

2 2 0 0 7

0

0 7 3 0

0 0 7 4 0

Total on clot plus gelatin sponge

25

16

24

24

Total on both types of culture

30

18

36

36

Rudiments were exposed to virus from the same virus pool, and in the same manner employed with adult-gland fragments in this report. Control rudiments were allowed to stand in medium to which no virus was added, for the same 2-hour period, and all were kept at room temperature in a desiccator jar gassed with 5 percent CO2 and 95 percent air during the exposure period.

Liquid medium, of the composition described previously, was first added to the cultures 16 hours after the rudiments were placed. Both control and test cultures were fed 2 to 3 times weekly, and pH was controlled by gassing the tubes with 5 percent CO2 in 95 percent air.

The results are tabulated in table 2 along with previous data. The 14-day rudiments developed much as described by Borghese (79, 80) and Grobstein (81, 82), regardless of whether they had been infected or not. Likewise, no difference between control and test cultures of 13-day rudiments could be detected in living cultures during the 1st week, though the degree and speed of morphogenesis displayed by these earlier rudiments was less extensive and more variable than for the later ones.

Viral effects first became visible in living cultures after 9 to 14 days. During this interval, necrosis of the mesenchymal cells became obvious, since the cells became dark and granular and eventually lost form. Figures 22 and 23 compare control and test cultures of 13-day rudiments after 17 days in culture under plasma clot, without gelatin sponge. Besides the fibroblastic cells of the mesenchyme, macrophages also underwent necrosis. These, too, became dark and rounded as the cytoplasm filled with brown granules. In the control cultures, swarms of clear, vigorous-appearing macrophages appeared in the rudiments and on the glass-clot interface after the 1st week of culture, and these persisted in good condition as along as the cultures were maintained. In living cultures on or under plasma clot there was no sign of epithelial damage or atypical proliferation at any time, but after the necrosis of the mesenchyme became marked, the epithelial elements were obscured by the granular debris (fig. 23). In cultures using gelatin sponges, the uninfected as well as the infected rudiments were drawn against and into the sponge as clot was lysed laterally. It was therefore more difficult to visualize differences in these until microscopic sections were prepared. However, the relative paucity of macrophages on the glass and in the clot adjacent to the infected rudiments contrasted with the presence of large numbers of healthy-appearing macrophages in comparable locations in control cultures.

In fixed and stained cultures the earliest histologic change was the presence of small foci of necrosis in the capsular mesenchyme, associated with nuclear changes in the mesenchymal cells, similar to the changes previously described in epithelium (39). These changes, without involvement of epithelium and without proliferative features, were seen in 13-day infected rudiments after 9 days in culture (fig. 25). Control cultures showed no similar changes (fig. 24).

Cytolysis and nuclear lesions of the epithelial portions of the rudiments were first seen in 12-day-old cultures of infected 14-day rudiments (figs.

26 and 27). At this time there were also necrosis and nuclear inclusions in the mesenchymal cells, but still no extensive loss of these elements.

After 22 days in culture, infected 13- or 14-day rudiments showed a more extensive necrosis of mesenchyme, as well as epithelial necrosis, and nuclear lesions of both epithelium and mesenchyme (figs. 28 to 31). Although the pattern of the epithelial tree was disrupted by the effects of necrosis, there was still no evidence of a proliferative reaction.

The earliest evidence of a proliferative response that has been observed as yet in infected rudiment cultures occurred in 14-day rudiments after

27 days in culture. Figures 33 to 35 illustrate the morphologic alteration that was seen in comparison with a control culture (fig. 32). The alteration was similar qualitatively to that seen in infected cultures of glands of newborns, but was considerably less voluminous in the rudiments. It is difficult to assess this apparent quantitative difference because of the great differences in the nature of the starting materials. Certainly as yet there has occurred no distant extension of the altered epithelium into the gelatin sponges, as was seen with infected cultures of adult submandibular glands.

To summarize these still incomplete observations on the response of salivary gland to polyoma virus in this tissue-culture system, it can be said that there was a trend from the youngest to the oldest tissue toward an increasingly active proliferative response, whereas there was a trend in the opposite direction with respect to the cytolytic response. The difference between early rudiments and newborn tissue also appeared to be greater than the difference between newborn and adult tissue.