Journal of Clinical Investigatioit Vol. 42, No. 9, 1963 REGULATION OF GROWTH AND SEQUESTERING FUNCTION * By HARRY S. JACOB, RICHARD A. MAcDONALD, AND JAMES H. JANDL (From the Thorndike Memorial Laboratory, the Second and Fourth [Harvard] Medical Serv- ices, and the Mallory Institute of Pathology, Boston City Hospital, and the Department of Medicine, Harvard Medical School, Boston, Mass.) (Submitted for publication March 16, 1963; accepted May 31, 1963)

A major function of the reticuloendothelial sys- It is probable, however, that functional splenic tem is to remove defective cells and other par- tissue actually increases, since in long-standing ticulate matter from the blood stream. Red cells hemolytic anemias, such as thalassemia, the spleen that have been severely altered, as by complement- may progressively increase its capacity to destroy fixing or agglutinating antibodies (1, 2), ag- transfused normal red cells (14). Further in- glutinating metals (3), prolonged incubation (4), direct evidence that the of chronic hemo- or severe chemical injury (5), are sequestered lytic disorders may become hyperfunctional may throughout the reticuloendothelial system, much be inferred from the depression in the levels of as are foreign particles such as bacteria, carbon, blood elements other than red cells as splenic or metallic colloids (6, 7). When this is so, the enlargement progresses. This sort of "hyper- liver, with its great blood flow, is the major site splenism" has also been observed in patients with of sequestration, the spleen simply assisting to infections (15) and in animals injected with the extent of its own blood supply. In other in- various macromolecular or particulate materials stances, the spleen, despite its modest blood flow, (16-21). In these examples, the increased load is the dominant sequestering organ. Thus a vari- of particulate or poorly solubilized material to be ety of pathological red cells (8-11) and red cells cleared by reticuloendothelial tissue has stimulated mildly altered by chemical (3-5, 12), physical the spleen to overgrow and hyperfunction. Con- (5), or immune (1, 2) agents are susceptible to versely, the finding that splenic atrophy occurs in selective splenic sequestration and destruction. animals subjected to repeated venesections (22) Comparatively little is known of the participation has led to the suggestion (23) that reducing the of the lungs and in particle clear- load of effete red cells delivered to the spleen ance. The lung does not sequester experimentally may inhibit its growth. To investigate these altered red cells (13), unless the cells are very matters, studies were made in rats of the growth coarsely agglutinated (2). Thus far, the marrow and function of spleens, livers, and spleen auto- has been observed to play only a small role in transplants under a variety of experimental the clearance of experimentally altered red cells "work loads." (13), although it may possibly sequester cells that The ability of transplanted spleen tissue to re- have been very subtly injured. generate was reported almost fifty years ago by The factors that regulate the growth and func- Manley and Marine (24, 25). That such trans- tion of the reticuloendothelial system, as well as plants could in some way function was suspected the relative activities of such component parts as by Perla and Marmorston-Gottesman (26) from the spleen, liver, and bone marrow, are obscure. observations that spleen autotransplants protected That splenic enlargement occurs in chronic hemo- otherwise splenectomized rats from bartonellosis. lytic anemias suggests that spleen growth may be Later, Palmer, Kemp, Cartwright, and Wintrobe stimulated by an increase in its "work load." (27) found that spleen autotransplants in rats Admittedly, some of this increased splenic bulk suppressed the leukocytosis that normally follows may represent trapped red cells and their debris. . In a report (28) preliminary to * This investigation was supported in part by grant this communication, regenerated spleen autotrans- GM-03507-11 from the National Institutes of Health, plants were found capable of sequestering Cr51- Bethesda, Md. labeled red cells that had been coated with an 1476 REGULATION OF THE SPLEEN 1477

FIG. 1. GROSS APPEARANCE OF A SPLEEN AUTOTRANSPLANT. Two views are shown of a fully grown transplant after it had been dissected free of overlying fat and fascia. incomplete antibody. Similarly, in the present ond, "hemisplenectomized" group, only half the parent studies, the capability of reticuloendothelial tissue spleen was removed, the other half remaining in its original site. For comparison, a third group of rats was to trap altered red cells has been utilized for splenectomized without autotransplants, and a fourth quantitating the function of this tissue under vari- group was subj ected to a sham operation. ous experimental conditions. Preliminary reports In those experiments involving the use of Millipore of these findings have been published (28, 29). filters, a transverse section of spleen was aseptically sus- pended in sterile Eagle's (30) tissue culture medium 2 within a cylindrical chamber. The cylinder wall con- MATERIALS AND METHODS sisted of a nylon-reinforced Millipore filter having a pore Preparation of animals. All studies were made on a size of 0.45 A.3 After the ends of the cylinder were Bartonella bacilliformis-free, Caesarean-derived strain of sealed with plastic plugs, it was placed free in the peri- Sprague-Dawley rat (Charles River C. D.).1 Weanling toneal cavity of the original donor rat. At the time of rats, weighing from 50 to 75 g at the time of operation, sacrifice, several weeks later, the filter chamber was em- was were used because of the report (25) that transplanted braced by omentum. The host animal at this time spleen tissue grows best in young animals. Under of normal appearance and size. ether anesthesia and aseptic conditions, a small, weighed, Measurement of the function of reticuloendothelial tis- transverse section of spleen was autotransplanted into suies. At various intervals after transplantation, animals a subcutaneous pocket fashioned in the anterior abdominal were injected by tail vein with 1 ml of a 50%o washed wall. At the time of transplantation, one group of ani- 2 Generously supplied by Dr. A. Martin Lerner. mals underwent total splenectomy, whereas in the sec- 3 Available through the courtesy of Mr. Richard A. 1 Obtained from Charles River Laboratories, Boston, Cotton, Millipore Filter Corporation, Bedford, Mass., as Mass. 4 cm X 5 mm nylon-reinforced EI A filter tubes. 1478 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL

* S . \I .. A+*

'~~~~~~~~Ur | 4 v v > e < ^

...... 17...... *xe<< FIG. 2. SEQUENCE OF HISTOLOGIC CHANGE AFTER SPLEEN TRANSPLANTATION. One week after transplantation, as shown on the left, the bulk of the tissue is necrotic; only a thin rim of mononuclear cells containing some disorgan- ized vascular spaces give evidence of viability. After gradual replacement of the necrotic tissue, normal splenic his- tology is evident in the full-grown 8-week-old transplant seen on the right. suspension of isologous red cells that had been labeled In certain animals, a chronic hemolytic anemia was with 5 ,c of Na2Cr'O,4.4 The red cells were altered produced by the subcutaneous injection 3 times a week either by incubation with an "incomplete" antibody to rat of 5 mg of fi-acetylphenylhydrazine6 per 100 g body red cells prepared in rabbits (13), or by exposure of a weight. In such rats, hemoglobin levels were about 50% 50% cell suspension to an equal volume of 12 mM N-ethyl- of normal during drug administration, and reticulocyte maleimide,5 a sulfhydryl inhibitor (31). In both cases, concentrations averaged about 40%. concentrations of antibody or inhibitor were chosen that gave pronounced splenic but little hepatic uptake of the RESULTS altered cells (12, 13). Three hours after injection, at a time when approximately 50% of the labeled cells had Appearance of spleen autotransplants. Spleen been sequestered, the animals were sacrificed. Cardiac autotransplants survived in over 957% of 150 ani- blood, spleens, livers, kidneys, lungs, femurs, and spleen mals operated on. The gross appearance of a transplants were removed, weighed, and assayed for ra- fully grown transplant in situ is shown in Figure dioactivity in a well-type scintillation counter as de- scribed previously (32). Finally, histologic sections were 1. A well-formed capsule allowed for its easy made from formalin-fixed specimens of spleens, livers, dissection from the surrounding subcutaneous fat. and spleen transplants. Numerous small blood vessels supplied the tissue, The levels of blood cells were measured by standard which thereby lacked a hilum and assumed the techniques (33). Platelets were counted by the method shape of a seminodular, oblate spheroid of deep of Pohle (34); approximately 1,000 platelets in two chambers were counted for each determination. purple color. The sequence of histologic changes occurring after spleen transplantation has been 'Obtained from Abbott Laboratories, North Chicago, reported previously by others (25, 35-38). Ill. 5 Obtained from Schwarz BioResearch, Inc., Mount 6 Obtained from the Matheson Company, East Ruther- Vernon, N. Y. ford, N. J. REGULATION OF THE SPLEEN 1479

Briefly, in the first week following transplantation, SPLEEN TRANSPLANT WEIGHT the tissue underwxenit almost complete necrosis. ACETYLPHENYLHYDRAZINE NON-INJECTED RATS INJECTED RATS 280 At the end of this period, as seen onl the left SPLENECTOMIZED side of Figure 2, a large central necrotic area was 240 surrounded by a thin rim of viable, newly vascu- 200 larized tissue made tip of reticulum and lymphoid 60 I.- P

0 SPLEEN TRANSPLANT FUNCTION 1 SPLENECTOMIZED

FIG. 4. EFFECT OF j6-ACETYLPHENYLHYDRAZINE ADMIN- IS1RATION ON THE WEIGHT (UPPER) AND FUNCTION (LOWER) OF SPLEEN AUTOTRANSPLANTS. In untreated animals (left), the differences between the growth and function of transplants from splenectomized rats (solid circles) and those from hemisplenectomized 0 animals x (open circles) are significant. Repeated administra- tion of j6-acetylphenylhydrazine (right) accentuates I- these differences. Note the threefold increase in trans- plant size in splenectomized rats treated with 8-acetyl- phenylhydrazine (upper right) over that in their nonin- am jected mates (upper left). Each of the points repre- co sents the mean value from twelve animals.

typical Malpighian corpuscles. Characteristic 10 20 30 40 50 60 70 80 90 100 DAYS spleen architecture was evident within 3 weeks. Between 6 and 8 weeks, the transplants reached FIG. 3. EFFECT OF SPLENECTOMY AND HEMISPLENEC- their full growth and histologically resembled TOMY ON THE GROWTH (UPPER) AND FUNCTION (LOWER) normal OF RAT SPLEEN AUTOTRANSPLANTS. After an initial loss spleens (right side of Figure 2). in weight due to necrosis of the original tissue, spleen Growth and function of spleen autotransplants. transplants in splenectomized rats (solid circles) re- Although the histologic appearance of transplants generate rapidly and grow beyond their original weight. from splenectomized and hemisplenectomized ani- Though viable, transplants in animals with half spleens mals was similar, their rate of growth and func- remaining in situ (open circles) show far less growth. tion differed markedly. As shown in As shown in the lower portion, transplants begin to se- the top quester Cr51-labeled red cells coated with an incomplete portion of Figure 3, transplant weight decreased antibody within 2 weeks of surgery to an extent parallel- in both sets of animals for the first 3 weeks; dur- ing their size. Each of the points represents the mean ing this time, necrosis and early regeneration value for two animals. The small number of animals took place. Thereafter, particularly from weeks sacrificed at each time interval precluded any reasonable statistical comparisons. Individual values deviate from 3 to 5, transplants grew rapidly in splenectomized the means portrayed by less than + 15 mg (top) and by animals. In contrast, transplants in hemisple- less than 4 X 10-a ml (bottom), nectomized animals grew but slightly, and in vari- 1480 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL ous experiments attained only W to i the size of plant sequestering function was roughly equal in those from their splenectomized mates. Although the two groups, and reached about one-half that the percentage "take" was the same in both of native spleen and 10 to 30 times that of liver. groups of animals, the final mass of spleen tissue Results similar to those shown in Figure 3 were generated in splenectomized animals eventually obtained when the sequestering function of trans- exceeded the weight of the original transplant by plants was measured with red cells that had been from 1j to 2 times, while in the hemisplenecto- treated with N-ethylmaleimide. A large number mized animals, regeneration ceased before the of animals was employed to permit statistical com- original weight of the tissue transplant had been parisons (Figure 4). As shown in the left upper restored. Histologically, the transplants of sple- portion of Figure 4, 80-day-old transplants were nectomized animals appeared to contain more twice as large in splenectomized as in hemisple- spleen cells than did those of hemisplenectomized nectomized animals. As with sensitized red cells, animals. The lower portion of Figure 3 shows this was associated with a significantly greater that within 2 weeks after surgery, transplants trapping of the treated red cells, shown in the from both groups began to function, as measured left bottom portion of the figure. by their uptake of Cr51-labeled red cells that had Growth and function of spleen autotransplants been coated with incomplete antibody. The abil- in diffusion chambers. From the foregoing, it ity of transplants to trap altered red cells paral- would appear that the absence of parent spleen leled their rate of growth. Thus, a rapid increase tissue stimulates the growth of transplants. To in sequestering function occurred during the phase elucidate the mechanism of this stimulation, of rapid growth of the transplant in splenecto- transplants were isolated from circulating particu- mized rats. On a gram-for-gram basis, trans- late matter, but not from humoral factors, by

A.

*~~~~~~~~~~~.~~-, It. Iop* .

4,~~~~~~~~~~:Ptift.\ 4*~iZ*ii~~~ ~(*~~ 4..'k ' it ~~~~~~~~%.~~~~- *~W~~~, 46,~~~~ -0t4"

A' P , ~

44 J... \ 'it---- FI ITLG4 FSLE ISE FE UORNPATTO NODIFSO HMES fe eea weeksinMiliporechabers,splee autotranslants are vable, but ontain an icreased quntity of firous tissu ashowuderlowmanifcaton(20X)onthelef. n te rgh (70 X, ympocyes(dak ncli),retcu lumcllslighnucei),intat re cels, ad enothillie vaclr hnelLaben REGULATION OF THE SPLEEN 1481

NON-INJECTED ACETYLPHENYLHYDRAZINE- INJECTED 1,000-

900'

SPLENECTOMIZED NO TRANSPLANT 4 ro V)800' CL) z SPLENECTOMIZED WITH TRANSPLANT - 700 In I- To SHAM-OPERATED J

4 -j 600'I. a.

0 0 80 DAYS FIG. 6. EFFECT OF SPLEEN TRANSPLANTS ON PLATELET LEVELS. In noninjected animals (left), the usual elevation of platelets after splenectomy (open circles) is suppressed by full-grown spleen transplants (triangles) this difference was statistically significant (p < 0.025). Simi- larly, the exaggerated thrombocytosis after splenectomy in rats given ,3-acetylphenylhydrazine (right) is inhibited by the presence of transplants (p < 0.025). In animals sham-operated on that were injected with jS-acetylphenylhydrazine and that possessed hyperplastic spleens (right, solid circles), platelet levels are significantly lower than in their noninjected mates sham-op- erated on (left, solid circles), p < 0.001. Each of the points represents the mean value from twelve animals. transplantation within Millipore-filter diffusion suggestion that increased "work" stimulates chambers placed in the peritoneal cavities of a spleen tissue growth and function, experimental group of 20 animals. Absence of radioactivity in groups of rats analogous to those described above these transplants after the injection of Cr5l-labeled were given injections of /8-acetylphenylhydrazine red cells into the host animals verified their ex- 3 times a week, beginning 2 days after surgery, clusion from the circulation. This transplanted to produce a chronic hemolytic anemia. Giant tissue remained viable histologically (Figure 5), spleen transplants were induced by this procedure, and contained normal splenic cellular elements some nearly as large as the parent spleens. The such as lymphocytes, reticulum cells, red blood average weight of such transplants, as shown in cells, fibroblasts, and endothelial-lined vascular the right upper portion of Figure 4, was about spaces (Figure 5, right portion). The splenic 3 times that of transplants in uninjected rats (left architecture was poorly organized, however, most upper portion). By extrapolation, it is estimated of the tissue consisting of fibroblasts (Figure 5, that such transplants would weigh about 100 g in left portion). Though viable, these transplants man. As in the uninjected animals, growth was failed to grow, whether or not splenectomy had most pronounced when the transplants repre- been carried out. sented the sole existing splenic tissue. Although Effect of chronic hemolysis on spleen autotrans- foci of extramedullary hematopoiesis were seen plant growth and function. To assess further the histologically in these enlarged transplants, retic- 1482 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL

uloendothelial hyperplasia was striking, and as portion, Figure 6). Leukocytosis after splenec- shown in the right lower portion of Figure 4, this tomy was similarly suppressed (not shown), increase in transplant size in splenectomized ani- confirming the findings of Palmer and associates mals was accompanied by augmented sequester- (27). As shown on the right side of the figure, ing function. Transplants within Millipore-filter the repeated administration of /3-acetylphenylhy- chambers, thereby denied contact with the circu- drazine led to an exaggerated thrombocytosis in lating red cells injured by /3-acetylphenylhydra- splenectomized animals lacking transplants; but zine, were not stimulated to greater growth by again, in splenectomized animals possessing trans- the hemolytic process. plants, platelets were held to normal levels. By Effect of spleen autotransplants on circulating analogy with their uptake of effete red cells, it is platelet and leukocyte levels. The thrombocytosis probable that these transplants suppress leukocyte that usually follows splenectomy was inhibited by and platelet levels by their capacity to sequester the presence of full-grown spleen transplants (left blood cells. In favor of this interpretation is the

120- WEIGHT z .SPLENECTOMIZED 4 100- (A 2 so- zs;4- w I 60-

z 40- HEMISPLENECTOMIZED

-J 20- 0n (n I I

SPLENECTOM IZED it) FUNCTION

I0

I- HEMISPLENECTOMIZED

-i

1800- 0. PLATELETS 600- 1400- 200- Iz SPLEN ECTOMIZED 1000S HEMISPLENECTOMIZED o 8001 I ~o

I 0 I 2 3 -4 WEEKS AFTER TRANSPLANTATION FIG. 7. SEQUENTIAL CHANGES IN PLATELETS IN RELATION TO THE GROWTH AND FUNCTION OF SPLEEN AUTOTRANSPLANTS. In the second week after trans- plantation, spleen transplants suppress the thrombocytosis (lower) that oc- curs in splenectomized animals (solid circles). At this time, sequestering function (middle) and growth (upper) have just become manifest. Platelet levels in splenectomized rats without transplants remain elevated throughout (not shown), while in transplant-containing animals with half-spleens (open circles) platelets remain normal. Each point represents the mean value from six animals. REGULATION OF THE SPLEEN 1483 30

~25-

W 20-

z N5RS L I. P E L E E N

wa N

W5-~ ~ ~ ~ ~ ~

NORMAL SPLE NOM EGALY SPLENECTOMY

FIG. 8. INTERRELATIONSHIPS BETWEEN HEPATIC AND SPLENIC ERYTHROPLASIA. As compared to normal animals (left), the uptake by the liver of labeled red cells treated with N-ethylma- leimide was nearly reduced by half in animals made splenomegalic by p-acetylphenylhydrazine administration (middle), p < 0.001. On the other hand, hepatic uptake was increased in ani- mals that had been splenectomized 8 weeks previously (right), p < 0.001. Each bar represents the mean value from twelve animals. finding that the increased platelet and leukocyte leukocyte alterations paralleled those observed levels of splenectomized animals were not sup- with platelets. pressed by transplants within Millipore-filter The administration of 8-acetylphenylhydrazine chambers, although it is emphasized that spleen to animals sham-operated on caused hyperplasia regeneration was poorly organized within these of their spleens as it did in those with spleen diffusion chambers. autotransplants. Such spleens were 4 to 5 times The temporal sequence of thrombocytosis after larger than those from noninjected controls. Al- splenectomy, and its suppression by spleen trans- though, in part, this increment in size reflected an plants, is shown in the bottom portion of Figure 7. accumulation of red-cell debris and the appear- Significant depression of the thrombocytosis in ance of extramedullary hematopoiesis, reticulum- animals is not evident until 2 splenectomized cell hyperplasia was evident histologically. In weeks after transplantation, at a time when the these splenomegalic animals sham-operated on, increased weight (Figure 7, upper portion) and depicted at the bottom right of Figure 6, platelet function (Figure 7, middle portion) of the trans- levels were depressed significantly below those of planted tissue have just become manifest. In noninjected rats (bottom left). Leukocyte levels contrast, in splenectomized animals without trans- l)lants, platelet levels remained high during the were similarly affected. Thus, splenic hyper- period of observation (not shown), while in hemi- plasia, induced by an abnormality of red cells, led splenectomized rats with transplants, platelet to a reduction of the levels of platelets and leuko- levels remained normal. In this same study, cvtes as well. Interrelationships between the liver and spleen. 7 In contrast to the other experiments, female rats were As shown in Figure 8, the sequestering function of employed in this study, which may explain the somewhat higher base-line level of platelets. the liver is affected by that of the spleen. Ap- 1484 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL

Ps(0lO (005 (0.02 (0.30 (0.0 0.001 <0005 240 0

220, 0 0 I- z 200 0 0 .

IL0 0 0 0 I-O z T 0 0 0 0 E 160 ILJ Bd!4 l- 0 0 0 I

0 0 0 T 0 l. 0.I 0 (D 'D, JA 0 8

° 8 8=3 0 0 0 8 0 0 0 0 a - 2 4 6 - WEEKS AFTER SPLENECTOMY

FIG. 9. INCREASE IN HEPATIC SEQUESTRATION OF ABNORMAL RED BLOOD CELLS AFTER SPLENECTOMY. After splenectomy, hepatic sequestration of la- beled red cells treated with N-ethylmaleimide gradually increased as com- pared with control animals sham-operated on (represented by 100% on the ordinate). Values from animals sham-operated on are depicted by open symbols; those from splenectomized animals by closed symbols. Mean val- ues are indicated by horizontal bars. Although there was considerable vari- ation in the data, it appeared that sustained increases in hepatic reticuloen- dothelial function were not attained until about 2 months after splenectomy. proximately 10% of sulfhydryl-inhibited red cells the abnormal red cells, rather than differences were sequestered in the livers of normal animals merely in splenic competition for these cells.8 3 hours after injection (left portion); this hepatic The hepatic response to splenectomy was stud- function was nearly reduced by half in animals ied further. As shown in Figure 9, hepatic se- made splenomegalic by prior ,8-acetylphenylhydra- questering ability underwent an irregular in- zine administration in which an increased pro- crease following splenectomy, as compared to that portion of cells was trapped in the spleen (middle observed in animals sham-operated on and in- portion). In contrast, the sequestering ability jected with the same injured red cells. The se- of the liver was markedly increased in animals questering ability of the livers appeared to reach that had lacked spleens for several months (right a maximum some 6 to 8 weeks-after splenectomy. portion). In these studies, over 50% of the Of further interest was the histologic finding that labeled cells were still in the circulation at the Kupffer cells from animals splenectomized and treated with for pro- time of sacrifice. This excess of circulating cells, f8-acetylphenylhydrazine shown previously to be altered in a uniform man- 8 Previous studies (39) have demonstrated that rat ner (12), assures that the liver has a comparable livers removed from animals and assayed for radio- opportunity in each study to sequester labeled activity as in these studies contain 3.5%o of the circulating blood volume. Correction of hepatic radioactivity for cells. Thus, the observed differences in hepatic this intravascular activity would increase the relative sequestration reflect variation in its avidity for differences depicted in Figure 8. REGULATION OF THE SPLEEN 1485

TABLE I the size of the reticuloendothelial system and its Effect of splenectomy on liver and femur functional capacity with respect to cell sequestra- altered red cells* sequestration of tion are governed by the work required of it. Days The finding that transplanted splenic tissue is splenec- tomized RBC uptake, ml X 10-3 stimulated to greater size after splenectomy sup- before No. of sacrifice animals Livert Femursj ports this suggestion and confirms the observation of Marine and Manley (25). In our studies, this 2 6 63.5 i 4.5 SD 2.3 ± 1.4 60 7 77.5 i 11.5 4.2 ±t 1.1 stimulation was associated with an increase in transplant sequestering function and was further * Incubated with 30 jmoles N-ethylmaleimide per ml intensified by the added imposition of a hemolytic red blood cells. t Liver: p <0.05. process. Although not completely excluded by tFemurs: p <0.01. these studies, it seems unlikely that humoral fac- tors are involved in this stimulation. Thus, viable longed periods of time showed much more eryth- spleen autotransplants, which normally grow well rophagocytosis than did similarly treated animals in the peritoneal cavity of rats (38, 39), did not with intact, hyperplastic spleens. In ancillary appreciably increase in size, even in splenectomized studies, it was found that erythrophagocytosis by animals, when isolated from circulating particu- Kupffer cells occurred in all animals, splenecto- late matter in diffusion chambers. Furthermore, mized or not, early in the course of /8-acetylphen- the presence of hyperplastic parent spleens pro- ylhydrazine-induced hemolysis. However, after duced by injections of 83-acetylphenylhydrazine several weeks of hemolysis, erythrophagocytosis did not inhibit transplant growth; indeed, such by Kupffer cells diminished in animals with transplants were larger than those of uninjected spleens (these having become greatly enlarged), animals. Enlargement of spleens and spleen but persisted in animals without spleens. transplants in these studies was due to an increase Another experiment was performed to verify in their number of cells as evidenced by: 1) the the effect of splenectomy on hepatic sequestering general cellular hyperplasia seen histologically and function. Two days before sacrifice, one group 2) the close parallelism between function (as of rats that had been splenectomized for about 8 measured by blood-cell sequestration) and size. weeks was sham-operated on, while another, pre- A possibility not excluded by the present studies viously sham-operated on group was splenecto- is that much of the growth increment in spleen mized. Forty-eight hours later, all of the animals transplants reflects an immigration of cells from were injected with Cr5l-labeled, altered red cells, other sites, a mechanism that would account for a and organ radioactivity was measured as usual. failure of "growth" in diffusion chambers. Pre- The hepatic uptake in animals without spleens sumably the sources of such a possible cellular in- for the longer period exceeded that by the newly flux would be nonsplenic, since existing spleen splenectomized rats; the difference was relatively tissue inhibits rather than abets spleen transplant small but significant (Table I). In addition to growth. the spleen and liver, reticuloendothelial tissue of Although in these experiments splenic hyper- the bone marrow participated to a small extent was induced of in the trapping of abnormal red cells. Such plasia by sequestration f3-acetyl- phenylhydrazine-injured red cells, other kinds of bone marrow activity was also found to be regu- particles presumably may also stimulate spleen lated in relation to the activity of the spleen. As growth and hyperfunction. For example, spleno- seen in Table I, femurs from animals without megaly with some features of "hypersplenism" spleens for 8 weeks trapped nearly twice the num- has been described in lipidoses such as Gaucher's ber of altered red cells as did those from animals in clinical as well as ex- splenectomized 2 days before sacrifice. disease (20. 40), (41) perimental (42) hyperlipemia, and after the ex- perimelntal injection of macromolecules such as DISCUSSION bacterial endotoxins (43), gelatin (17), shellac Effect of "work load" oil reticuloendothelial (44), zymosan (21), polyvinylpyrrolidone (45), growth and function. These studies suggest that and methylcellulose (18, 19). The view that par- 1486 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL ticulate matter may stimulate the growth of retic- thrombocytopenia often seen in chronic hemolytic uloendothelial cells by purely physicochemical states (53, 54), and probably accounts for the rather than immunologic mechanisms is consistent pancytopenias associated with administration of with the deductions of Bilek (45) and of Thor- materials such as methylcellulose (18, 19) and becke and Benacerraf (46). Regardless of mech- zymosan (21 ). anism, the consequence of this stimulation in pa- Interrelations of the reticuloendothelial activity tients with certain chronic hematologic disorders of different organs. The present studies have is that a vicious cycle may ensue. Thus, red-cell demonstrated that the spleen, liver, and to a lesser sequestration would stimulate reticuloendothelial extent, the bone marrow share the sequestration hyperplasia, which in turn would exaggerate fur- of effete red cells in such a way that the activity ther sequestration. Indeed, in many such cases, of one organ is affected by that of the others. interruption of this cycle by removal of the pro- Thus, the reticuloendothelial activity of the liver gressively enlarging spleen has been of marked and marrow is stimulated by the extra work load clinical benefit (10, 14, 20, 47, 48). It may be imposed on them after splenectomy. This com- noted, however, that such a cyclic process, while pensatory response occurs gradually. Its devel- often dangerous, appears to be self-limited. Thus, opment may explain the inappreciable effect of splenic growth secondary to hemolytic processes splenectomy on red-cell life-span in otherwise nor- tends to level off, which is presumably the reason mal individuals and may also, in part, account for that spleen enlargement of itself rarely produces late relapses following splenectomy in various marked anemia (20). This in turn may indicate hemolytic anemias. That hepatic sequestration that the capacity of the spleen to increase its rate of vulnerable red cells does indeed become promi- of cell generation has a limit, as is the case with nent in such relapses after surgical splenectomy the bone marrow (49). This limit may be de- has been documented by radioisotopic techniques termined by the supply of stem cells. (2). The behavior of experimentally produced spleen The compensatory response of distant reticulo- transplants in these studies is analogous to that endothelial cells after splenectomy would seem to of the splenules in the clinical entity splenosis. be imperfect in many situations, possibly because This condition, manifested by a disseminated of an inability to duplicate certain anatomic fea- growth of splenules within the peritoneal cavity, tures peculiar to the spleen. Thus, splenectomy in is associated almost exclusively with previous hereditary spherocytosis leads to permanent re- splenectomy (50). Although splenic trauma be- mission in virtually all cases, although it is not fore or during surgery originally seeds this tis- certain that red-cell survival remains perfectly sue, it would appear that its growth is stimulated normal thereafter (55). In addition, the capacity in a manner analogous to that of spleen trans- to remove pathogens from the circulation would plants. Indeed, one or more unusually large and not appear to have been efficiently transferred probably functional splenules have been reported to other reticuloendothelial tissues in those pa- (51, 52) in the peritoneal cavities of two patients tients developing severe infections after splenec- with hereditary spherocytosis who relapsed after tomy (56-60), although increased propensity to initially beneficial splenectomies. The gradual infection in these patients is denied by some (61, emergence postoperatively of "accessory spleens" 62). On the other hand, Finland (63) has sug- may often reflect such a process. gested that increased susceptibility to fulminant Reticuloendothelial tissue, when stimulated as infection may be manifest for only a short period in these studies to become hyperplastic, was not after splenectomy. Thus, with bacteria, the com- discriminating in its hyperfunction. Overgrowth peinsatory response of the liver may be gradual, of the spleen, induced by its sequestration of ab- as was observed in these studies with effete red normial red cells, significantly depressed the levels cells. of platelets and leukocytes as well. It seems prob- Conversely, a depression of hepatic reticulo- able that depression of these elements was due to endothelial activity was observed in animals with their heightened sequestration in hyperplastic hyperplastic spleens. In rats with 8-acetylphenyl- spleens. This mechanism might explain the hydrazine-induced splenomegaly, a diminished REGULATION OF THE SPLEEN 1487 hepatic uptake of labeled, effete red cells and an in the littoral cells of the liver. It is difficult in inhibition of erythrophagocytosis lby Kupffer cells suich stll(lies to exclude the possibility that inm- wvas noted. By analogy to these findings, it mutnologic mechanisms are responsible for the would seenm reasonable to suspect that removal of cellular proliferation observed. Nevertheless, the a, hyperplastic spleen in humans might entail a l)roliferative response of Kupffer cells to foreign more dangerous reduction of net reticuloendo- particles is similar to the splenic response to the thelial activity, at least temporarily, than removal sequestration of autologous blood cells in these of a normal spleen. Indeed, removal of enlarged studies. spleens from individuals with underlying hema- Accordingly, it is postulated that abnormal cells tological disease has been most commonly asso- or particles on contact with reticulum cells stim- ciated with fulminant infection after splenectomy ulate a biochemical response that engenders mi- (58, 63), whereas no increased propensity to tosis. Such stimulation to cellular division is infection could be demonstrated in individuals reminiscent of that occurring in parthenogenesis with normal-sized spleens after splenectomy for whereby physical, chemical, or pathological (73) traumatic rupture (61). wounding induces cell division. That the growth and function of reticuloendo- Cellular proliferation induced by such agents, thelial tissues are interdependent is further sup- including effete red cells, may be essentially ported by reports that previous or coincident reparative in nature and involve a nonspecific re- splenectomy stimulates hepatic regeneration fol- sponse to injury, rather than being a specific lowing partial hepatectomy (64, 65). Histologi- response to humoral or immunologic stimulation. cally, this enhancement of growth entails mainly Recent studies (74) indicate that the splenic re- hyperplasia of Kupffer cells, which are observed sponse to hemolysis involves the generation of to be actively erythrophagocytic during this stage all cellular elements and more nearly resembles of regeneration (64). Conversely, Stern (66) wound healing than it does a response to a spe- has recently shown that spleens significantly in- cific regulator. In any case, the result of this crease their size and uptake of particulate matter stimulation, when applied to reticuloendothelial following partial hepatectomy; this augmentation tissue, is that the population of active phagocytes continues until full hepatic regeneration has oc- is temporarily and usefully increased. When curred. such stimulation is excessive, marked reticulo- Mechanism of work-load stimulation of reticulo- endothelial hyperplasia ensues with possible path- endothelial growth and function. It seems reason- ologic sequelae. able to conclude that reticuloendothelial cells mul- SUM MARY tiply in direct proportion to the quantity of un- wanted particles reaching them. The mechanism Studies were made in rats of the growth, his- of this stimulatory response remains obscure. tology, and function of reticuloendothelial tissue. Evidence reviewed recently by Karnovsky (67, It was shown that the growth and function of 68) establishes that phagocytosis of itself accel- spleens, livers, and spleen autotransplants, as erates metabolic processes within leukocytes in measured by their size and by their capacity to vitro. By analogy, it may be that a similar stim- sequester Cr51-labeled altered red cells, were reg- ulation of metabolism in reticuloendothelial cells ulated in proportion to the "work" required of in vivo might provide the energy for their mitoses. them. Thus, spleen autotransplants, in animals In this regard, it is of interest that following a lacking other spleen tissue, grew larger and func- short "blockade" period, a single injection of tioned more actively than did such transplants in particulate matter into animals increases their animals possessing spleens. Increasing the "phagocytic mass," at least as measured by the "work load" by imposing a hemolytic process subsequent increased removal rate of circulating produced a hyperplastic response both in spleens particles (69, 70). This rebound phenomenon and in spleen transplants. When spleen tissue has been shown by Kelly and her co-workers (71, in zivo was isolated from particulate matter of 72), using various organic particles, to be asso- the circulation by transplantation within diffusion ciated with an increased rate of cellular division chambers, it remained viable, but did not grow 1488 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL or detectably function despite the stimulus of and J. F. Delafresnaye, Eds. Springfield, Ill., hemolysis. Hyperplasia of splenic tissue induced Charles C Thomas, 1957, p. 52. by chronic hemolysis depressed the levels of all 7. Dobson, E. L. Factors controlling phagocytosis in Physiopathology of the Reticulo-endothelial Sys- blood cell elements, probably through increased tem, B. N. Halpern, B. Benacerraf, and J. F. sequestration. Delafresnaye, Eds. Springfield, Ill., Charles C The induction of hyperplasia in the spleen in- Thomas, 1957, p. 80. hibited reticuloendothelial function in the liver. 8. Young, L. E., R. F. Platzer, D. M. Ervin, and M. J. Conversely, splenectomy led to a compensatory in- Izzo. Hereditary spherocytosis. II. Observations on the role of the spleen. Blood 1951, 6, 1099. crease in hepatic and marrow sequestering func- 9. Jandl, J. H., M. S. Greenberg, R. H. Yonemoto, and tion. W. B. Castle. Clinical determination of the sites These findings seem best explained by the of red cell sequestration in hemolytic anemias. following proposals: 1) particulate matter directly J. clin. Invest. 1956, 35, 842. and locally stimulates the division of reticulo- 10. Watson, R. J., H. C. Lichtman, and H. D. Shapiro. Splenomegaly in sickle cell anemia. Amer. J. Med. endothelial cells and 2) the co-ordinated homeo- 1956, 20, 196. static regulation of the various reticuloendothelial 11. Baker, S. J., E. Jacob, K. T. Rajan, and E. W. Gault. organs is governed simply by the total particulate Hereditary haemolytic anaemia associated with work load. elliptocytosis: a study of three families. Brit. Evidence is discussed suggesting that the pro- J. Haemat. 1961, 7, 210. 12. Jacob, H. S., and J. H. Jandl. Effects of sulfhydryl liferative response of reticuloendothelial cells to inhibition on red blood cells. II. Studies in vsvo. particulate matter is basically reparative. It is J. clin. Invest. 1962, 41, 1514. postulated that the energy for cell division is de- 13. Jandl, J. H., and M. E. Kaplan. The destruction of rived from the biochemical excitation that at- red cells by antibodies in man. III. Quantitative tends phagocytosis. factors influencing the patterns of hemolysis in vivo. J. clin. Invest. 1960, 39, 1145. 14. Smith, C. H., I. Schulman, R. E. Ando, and G. Stern. ACKNOWLEDGMENT Studies in Mediterranean (Cooley's) anemia. I. The authors are grateful for the technical assistance of Clinical and hematologic aspects of splenectomy, Miss Susan C. Bell and Miss Nancye M. Files. with special reference to fetal hemoglobin synthesis. Blood 1955, 10, 582. 15. Jandl, J. H., H. S. Jacob, and G. A. Daland. Hyper- REFERENCES splenism due to infection: a study of five cases 1. Jandl, J. H. Sequestration by the spleen of red cells manifesting hemolytic anemia. New Engl. J. Med. sensitized with incomplete antibody and with me- 1961, 264, 1063. tallo-protein complexes (abstract). J. clin. Invest. 16. Hueper, W. C. Macromolecular substances as patho- 1955, 34, 912. genic agents. Arch. Path. 1942, 33, 267. 2. Jandl, J. H., A. R. Jones, and W. B. Castle. The de- 17. Hueper, W. C. Experimental studies in cardiovascu- struction of red cells by antibodies in man. I. Ob- lar pathology. V. Effects of intravenous injections servations on the sequestration and lysis of red of solutions of gum arabic, egg albumin and gelatin cells altered by immune mechanisms. J. clin. In- upon the blood and organs of dogs and rabbits. vest. 1957, 36, 1428. Amer. J. Path. 1942, 18, 895. 3. Jandl, J. H., and R. L. Simmons. The agglutination 18. Hueper, W. C. Reactions of the blood and organs of and sensitization of red cells by metallic cations: dogs after intravenous injections of solutions of interactions between multivalent metals and the methyl celluloses of graded molecular weights. red-cell membrane. Brit. J. Haemat. 1957, 3, 19. Amer. J. Path. 1944, 20, 737. 4. Jandl, J. H., and A. S. Tomlinson. The destruction 19. Palmer, J. G., E. J. Eichwald, G. E. Cartwright, and of red cells by antibodies in man. II. Pyrogenic, M. M. Wintrobe. The experimental production of leukocytic and dermal responses to immune hemo- splenomegaly, anemia, and leucopenia in albino lysis. J. clin. Invest. 1958, 37, 1202. rats. Blood 1953, 8, 72. 5. Harris, I. M., J. M. McAllister, and T. A. J. Prank- 20. Motulsky, A. G., F. Casserd, E. R. Giblett, G. 0. erd. The relationship of abnormal red cells to the Broun, Jr., and C. A. Finch. Anemia and the normal spleen. Clin. Sci. 1957, 16, 223. spleen. New Engl. J. Med. 1958, 259, 1164 and 6. Benacerraf, B., G. Biozzi, B. N. Halpern, and C. 1215. Stiffel. Physiology of phagocytosis of particles by 21. Gorstein, F., and B. Benacerraf. Hyperactivity of the the R. E. S. in Physiopathology of the Reticulo- reticuloendothelial system and experimental anemia endothelial System, B. N. Halpern, B. Benacerraf, in mice. Amer. J. Path. 1960, 37, 569. REGULATION OF THE SPLEEN 1489

22. De Langen, C. D. Function of the spleen and blood. 40. Mandelbaum, H., L. Berger, and M. Lederer. Acta med. scand. 1943, 115, 271. Gaucher's disease. I. A case with hemolytic anemia 23. Crosby, W. H. Normal functions of the spleen rela- and marked thrombopenia: improvement after re- tive to red blood cells: a review. Blood 1959, 14, moval of spleen weighing 6822 grams. II. Lipid 399. analysis of the Gaucher's spleen. Ann. intern. 24. Manley, 0. T., and D. Marine. The transplantation Med. 1942, 16, 438. of splenic tissue into the subcutaneous fascia of the 41. Berk, M. Secondary hypersplenism with recurrent abdomen in rabbits. J. exp. Med. 1917, 25, 619. gastrointestinal bleeding. Calif. Med. 1953, 78, 518. 25. Marine, D., and 0. T. Manley. Homeotransplantation 42. Stuart, A. E., G. Biozzi, C. Stiffel, B. N. Halpern, and of the spleen in rabbits. and D. Mouton. The stimulation and depression of III. Further data on growth, permanence, effect of reticulo-endothelial phagocytic function by simple age, and partial or complete removal of the spleen. lipids. Brit. J. exp. Path. 1960, 41, 599. J. exp. Med. 1920, 32, 113. 43. Ho, M., and E. H. Kass. Hemolytic anemia in rab- 26. Perla, D., and J. Marmorston-Gottesman. Studies on bits following injection of bacterial endotoxin. Bartonella muris anemia of albino rats. III. The Proc. Soc. exp. Biol. (N. Y.) 1958, 97, 505. protective effect of autoplastic splenic transplants 44. Shen, S. C. Unpublished studies. on the Bartonella muris anemia of splenectomized 45. Bilek, 0. Experimental production of proliferative rats. J. exp. Med. 1930, 52, 131. lesions of the reticulo-histiocytic system in rats 27. Palmer, J. G., I. Kemp, G. E. Cartwright, and M. M. (so-called experimental reticulosis). II. Prolifera- Wintrobe. Studies on the effect of splenectomy tion of the RHS in rats after polyvinylpyrrolidone. on the total leukocyte count in the albino rat. Vnitrni lekairstvi 1961, 7, 1324. Blood 1951, 6, 3. 46. Thorbecke, G. J., and B. Benacerraf. The reticulo- 28. Jandl, J. H. Sequestration of sensitized red cells by endothelial system and immunological phenomena. splenic autotransplants. Clin. Res. 1960, 8, 210. Progr. Allergy 1962, 6, 559. 29. Jacob, H. S., R. A. MacDonald, and J. H. Jandl. 47. Doan, C. A. Hypersplenism. Bull. N. Y. Acad. Med. The regulation of spleen growth and sequestering 1949, 25, 625. function. J. clin. Invest. 1962. 41, 1367. 48. Sprague, C. C., and J. C. S. Paterson. Role of the 30. Eagle, H. The growth requirements of two mam- spleen and effect of splenectomy in sickle cell dis- malian cell lines in tissue culture. Trans. Ass. ease. Blood 1958, 13, 569. Amer. Phycns 1955, 68, 78. 49. Crosby, W. H. The metabolism of hemoglobin and 31. Jacob, H. S., and J. H. Jandl. Effects of sulfhydryl bile pigment in hemolytic disease. Amer. J. Med. inhibition on red blood cells. I. Mechanism of 1955, 18, 112. hemolysis. J. clin. Invest. 1962, 41, 779. 50. Cotlar, A. M., and E. J. Cerise. Splenosis: the 32. Jandl, J. H. The agglutination and sequestration of autotransplantation of splenic tissue following in- immature red cells. J. Lab. clin. Med. 1960, 55, jury to the spleen. Report of two cases and review 663. of the literature. Ann. Surg. 1959, 149, 402. 33. Ham, T. H. A Syllabus of Laboratory Examinations 51. Stobie, G. H. Splenosis. Canad. med. Ass. J. 1947, in Clinical Diagnosis. Critical Evaluation of Labo- 56, 374. ratory Proceedures in the Study of the Patient, 52. Mackenzie, F. A. F., D. H. Elliot, H. H. G. Eastcott, revised edition, L. B. Page and P. J. Culver, Eds. N. C. Hughes-Jones, P. Barkhan, and P. L. Cambridge, Mass., Harvard University Press, 1960. Mollison. Relapse in hereditary spherocytosis with 34. Pohle, F. J. The blood platelet count in relation to proven splenunculus. Lancet 1962, 1, 1102. the menstrual cycle in normal women. Amer. J. 53. Wintrobe, M. M. Clinical Hematology, 5th ed. med. Sci. 1939, 197, 40. Philadelphia, Lea and Febiger, 1961, p. 650. 35. Perla, D. The regeneration of autoplastic splenic 54. River, G. L., A. B. Robbins, and S. 0. Schwartz. transplants. Amer. J. Path. 1936, 12, 665. S-C hemoglobin: a clinical study. Blood 1961, 18, 36. Calder, R. M. Autoplastic spleen grafts: their use in 385. the study of the growth of splenic tissue. J. Path. 55. Jandl, J. H. Hereditary spherocytosis in The Meta- Bact. 1939, 49, 351. bolic Basis of Inherited Disease, J. B. Stanbury, 37. Williams, R. G. The microscopic structure and be- J. B. Wyngaarden, and D. S. Fredrickson, Eds. havior of spleen autografts in rabbits. Amer. J. New York, McGraw-Hill, 1960, p. 1018. Anat. 1950, 87, 459. 56. King, H., and H. B. Shumacker, Jr. Splenic studies: 38. Cameron, G. R., and K. S. Rhee. Compensatory hy- I. Susceptibility to infection after splenectomy per- pertrophy of the spleen: a study of splenic growth. formed in infancy. Ann. Surg. 1952, 136, 239. J. Path. Bact. 1959, 78, 335. 57. Gofstein, R., and S. Gellis. Splenectomy in infancy 39. Crosby, W. H., and N. R. Benjamin. Frozen spleen and childhood: the question of overwhelming in- reimplanted and challenged with Bartonella. Amer. fection following operation. J. Dis. Child. 1956, 91, J. Path. 1961, 39, 119. 566. 1490 H. S. JACOB, R. A. MACDONALD, AND J. H. JANDL

58. Smith, C. H., M. Erlandson, I. Schulman, and G. 67. Karnovsky, M. L. Metabolic basis of phagocytic Stern. Hazard of severe infections in splenectom- activity. Physiol. Rev. 1962, 42, 143. ized infants and children. Amer. J. Med. 1957, 22, 68. Karnovsky, M. L, Metabolic shifts in leucocytes dur- 390. ing the phagocytic event in Biological Activity of 59. Broberger, O., F. Gyulai, and J. Hirschfeldt. Sple- the Leucocyte, Ciba Foundation Study Group No. nectomy in childhood. A clinical and immunologi- 10, G. E. W. Wolstenholme and M. O'Connor, Eds. cal study of forty-two children splenectomized in Boston, Little, Brown, 1961, p. 60. the years 1951-1958. Acta paediat. (Uppsala) 69. Biozzi, G., B. Benacerraf, and B. N. Halpern. The 1960, 49, 679. effect of Salm. typhi and its endotoxin on the 60. Lucas, R. V., Jr., and W. Krivit. Overwhelming in- phagocytic activity of the reticulo-endothelial sys- fection in children following splenectomy. J. tem in mice. Brit. J. exp. Path. 1955, 36, 226. Pediat. 1960, 57, 185. 70. Benacerraf, B., and M. M. Sebestyen. Effect of bac- 61. McKinnon, W. M. P., S. J. Boley, and J. Manpel. terial endotoxins on the reticuloendothelial system. Infection in children following splenectomy for Fed. Proc. 1957, 16, 860. traumatic rupture. J. Dis. Child. 1959, 98, 710. 71. Kelly, L. S., E. L. Dobson, C. R. Finney, and J. D. 62. Laski, B., and A. MacMillan. Incidence of infection Hirsch. Proliferation of the reticuloendothelial in children after splenectomy. Pediatrics 1959, 24, system in the liver. Amer. J. Physiol. 1960, 198, 523. 1134. 63. Finland, M. Serious infections in splenectomized children. Pediatrics 1961, 27, 689. 72. Kelly, L. S., B. A. Brown, and E. L. Dobson. Cell 64. Higgins, G. M., and J. T. Priestley. Experimental division and phagocytic activity in liver reticulo- pathology of the liver. VI. Restoration of the endothelial cells. Proc. Soc. exp. Biol. (N. Y.) liver in white rats after partial removal and splenec- 1962, 110, 555. tomy. Arch. Path. 1932, 13, 573. 73. Stolk, A. Pathological parthenogenesis in viviparous 65. Perez-Tamayo, R., and R. Romero. Role of the toothcarps. Nature (Lond.) 1958, 181, 1660. spleen in regeneration of the liver. An experi- 74. Jandl, J. H., H. S. Jacob, and R. A. MacDonald. mental study. Lab. Invest. 1958, 7, 248. Reticuloendothelial proliferation stimulated by in- 66. Stern, K. Studies on reticuloendothelial function in jured, autologous red cells. Proc. 3rd int. Sym- relation to the growth processes. Ann. N. Y. Acad. posium on Immunopathology. Basel, Benno Sci. 1960, 88, 252. Schwabe, in press.