Laura A. Vogel, PhD**School of Biological Sciences, Illinois State University; ‡Department of Psychology, Illinois
State University, Normal; †Millennium Pain Center, Bloomington, Illinois, U.S.A.
� Abstract: Patients with chronic pain are often chal-lenged with depression stemming from the long-term psy-chophysiological effects of their condition. Consequently,patients with chronic pain are often treated with morphine,which can induce immunosuppression, along with an anti-depressant. The antidepressant citalopram (CTP; Sigma-Aldrich Chemical, St. Louis, MO, U.S.A.) is a serotoninreuptake inhibitor that is reported to have immunomodu-latory effects. Thus, we investigated whether CTP adminis-tration impacted immunity in morphine-treated animals.Adult mice were pretreated for 7 days with either saline orCTP (10 or 30 mg/kg intraperitoneal injections twice daily),followed by subcutaneous implantation of a 25 mg mor-phine pellet for 48 hours. Spleen, thymus, and lymph nodeswere harvested to analyze total cell numbers, relative lym-phocyte populations, and lymphocyte function. In thisstudy, CTP had no effect on either total cell counts or lym-phocyte populations in the thymus. However, in the spleen,total splenocyte numbers in all CTP-treated animals dis-
played an increasing trend over saline-treated animals.Interestingly, although more cells were found in the spleen,distribution of splenic lymphocyte populations did notdiffer between treatments. Despite no increase in total cellnumber, a high dose of CTP (30 mg/kg) resulted in a signifi-cantly higher B cell population in the lymph nodes, while Tcell and NK cell numbers were not different. CTP did notsignificantly reverse morphine-induced weight loss orsplenic B cell antibody secretion in vitro. Additionally, CTPtreatment demonstrated a slight but not significantincrease in both splenic B and T cell mitogen-induced pro-liferation in vitro. In summary, CTP may have a specificpotential in the attenuation of morphine’s immunosuppres-sive effect by enhancing splenocyte numbers and lymphnode B cell populations. �
As early as 1898, Cantacuzene described the immuno-suppressive effect of opium as an inhibitor of leukocytephagocytosis in guinea pigs. More recently, chronicopioid usage has been linked to an increased risk ofopportunistic infections in the form of salmonella,
Address correspondence and reprint requests to: Laura A. Vogel, PhD,School of Biological Sciences, Illinois State University, Normal, IL 61790-4120, U.S.A. E-mail: [email protected].
Submitted: October 1, 2008; Revision accepted: December 18, 2008DOI. 10.1111/j.1533-2500.2009.00259.x
Despite this correlation, the mechanism by whichopioids induce immune suppression has yet to be fullycharacterized.
There are lines of evidence indicating that opioidsmay have a direct effect on immune function, as studieshave shown that opioids can act on opioid receptorsfound on immune cells.5,6 However, additional evidencepoints to a more potent indirect mechanism through theactivation of opioid receptors in the central nervoussystem (CNS) and subsequent signaling via the sympa-thetic branch of the peripheral nervous system. Thus,the CNS and the immune system may have a bidirec-tional link, where actions in the CNS could affect theimmune system and vice versa, via the sympatheticnervous system.7–9 Indeed, Pruett et al.10 described anindirect interaction between morphine and immune sup-pression by demonstrating that incubating antibody-producing cells with morphine in vitro did not elicit thesame effects on decreased antibody production com-pared to in vivo morphine delivery. Additionally, Fechoet al.9 have shown that splenocytic proliferative abilitymay be restored with the use of a b-adrenoceptorantagonist, implicating the indirect mechanism of actionvia the sympathetic nervous system as a site ofmorphine-induced immune suppression.
Serotonin (5-HT) has been implicated in the descend-ing pathway of pain modulation.11 Specifically, opioidanalgesia, which is mediated in the CNS, may interactwith serotonergic neurons.12 For instance, morphine-induced analgesia was attenuated upon treatment with a5-HT receptor agonist and was reversed by the additionof a serotonin precursor, 5-hydroxytryptophan(5-HTP).13 Similar to mu-opioid analgesia modulation,kappa-opioid analgesia is also attenuated by administra-tion of a tryptophan hydroxylase inhibitor and recoverswith 5-HTP.14 Additionally, endogenous opioid analgesiais inhibited with 5-HT(1) receptor agonists.15 Finally,mu-opioid and delta-opioid analgesias were reduced inmice lacking central serotonergic neurons, while kappa-opioid analgesia was completely abolished comparedwith those in wild-type animals.16
In a clinical setting, patients with chronic pain areoften diagnosed with depression and treated withantidepressants such as serotonin reuptake inhibitors inconjunction with opioid analgesics. A particularly inter-esting recent line of evidence indicates a role of seroto-nin in immune mediation because the monoaminereuptake inhibitor tramadol increased immune functionby increasing phytohemmaglutinin (PHA)-induced T
cell proliferation as well as NK cell activity.17,18 There-fore, we wished to elucidate the effects of combinedmorphine and antidepressant therapy on various indicesof immune function through the use of the antidepres-sant citalopram (CTP; Sigma-Aldrich Chemical, St.Louis, MO, U.S.A.), a selective serotonin reuptakeinhibitor (SSRI). We hypothesized that CTP wouldattenuate the immunosuppressive effects of morphinetreatment in mice. We found that morphine-treatedanimals, compared with placebo-treated animals, lost asignificant number of cells from the spleen and thethymus. Among animals administered with morphine,those that were also treated with CTP showed a dose-dependent trend for increasing cell numbers in thespleen. In addition, both splenic B and T cell prolifera-tion followed an increasing trend after treatment with30 mg/kg CTP in comparison to both saline and10 mg/kg CTP-treated animals. In the lymph node, Bcell populations were found to be significantly highest in30 mg/kg CTP-treated animals vs. saline- and 10 mg/kgCTP-treated animals. Our data suggest that CTP exhib-its a dose-dependent rescue of morphine-induced immu-nosuppression, specifically with B cells from the spleenand especially the lymph node.
METHODS
Murine Model of Morphine-InducedImmunosuppression
A mouse model of immunosuppression was induced viachronic morphine administration. C57BL/6 female miceaged 6 to 8 weeks were purchased from the NationalCancer Institute and housed in the animal care facility atIllinois State University on a 12-hour light–dark cycle.Mice were provided with food and water ad libitum.
Animals were anesthetized with isoflurane, whereanesthesia was induced at 4% isoflurane and 0.6% O2
in an inhalation anesthetic chamber and maintained at2% to 3% isoflurane through a nose cone. A 1-cmincision was made in between the region of the scapulaeafter being shaved and disinfected with a povidone-iodine solution. After creating a subcutaneous pocket,we implanted the mice with a 25-mg morphine pellet(National Institute on Drug Abuse, Bethesda, MD,U.S.A.), after which the skin was sutured with inter-rupted stitches. All animals were euthanized 48 hoursafter pellet implantation. All animal handling andexperimental manipulations were carried out underapproved institutional animal care and use protocols atIllinois State University.
196 • nguyen et al.
Drug Administration
A CTP solution was prepared by dissolving CTP in0.9% sterile saline and administered to animals viatwice-daily intraperitoneal injections at 0 mg/kg (salineonly), 10 mg/kg, or 30 mg/kg.19–21 Animals were pre-treated with one of the three doses for 7 days, andinjections continued through the day of morphine pelletimplantation to the end of the study (for a total of 9days of CTP administration).
Single-Cell Suspensions
After euthanasia, spleens, thymuses, and lymph nodes(axillary and inguinal) were harvested aseptically. Mousespleens, thymuses, and lymph nodes were crushedbetween the frosted ends of two sterile glass slides incomplete RPMI media (cRPMI; RPMI-1640 tissueculture media [Cellgro, Manassas, VA, U.S.A.], 10%fetal bovine serum [Hyclone, Logan, UT, U.S.A.], 2 mMglutamine, 50 mM 2-mercaptoethanol, 100 units/mLpenicillin, 100 mg/mL streptomycin) to produce single-cell suspensions. Red blood cells (RBCs) in spleens werelysed by using a lysis buffer (0.144 M ammonium chlo-ride and 0.017 M tris, pH 7.2). Cells were counted byusing a hemocytometer and trypan blue viability dye.
Flow Cytometry
Spleen, thymus, and lymph node cells (1 ¥ 106 cells) werelabeled separately with a mixture of fluorochrome-conjugated antibodies against T cells (anti-CD4-PE, BDBiosciences, San Jose, CA, U.S.A.), B cells (anti-B220-CyChrome, BD Biosciences), or NK cells (anti-NK1.1-FITC, BD Biosciences). Briefly, cells were washed withHanks’ Balanced Salt Solution (Cellgro, Manassas, VA,U.S.A.) containing 0.1% bovine serum albumin (SigmaAldrich, St. Louis, MO, U.S.A.) and labeled with themixture of fluorochrome-conjugated antibodies for 15minutes on ice in the dark. After a second wash, cells wereanalyzed for the population levels of T, B, and NK cells byusing FACS analysis (BD FACSCalibur and CellQuestPro software, BD Bioscience, San Jose, CA, U.S.A.).
Proliferation Assay
Measurement of the proliferative abilities of spleen cellswas performed by using tritiated thymidine. Cells fromthe previously prepared splenic cell suspensions werecultured at 2 ¥ 105 cells/well in cRPMI for 48 hoursat 37°C in a 5% CO2 incubator on a 96-well plate.Mitogens (either 20 mg/mL lipopolysaccharide [LPS] or2 mg/mL conconavalin A; Sigma Aldrich, St. Louis, MO,
U.S.A.) were also added to the medium in separate wellsto induce proliferation of B cells or T cells, respectively.After this initial incubation period, cells were pulsedwith tritiated thymidine (3H-thymidine; 1.0 uCi/well;MP Biomedicals, Solon, OH, U.S.A.) for an additional24-hour period at 37°C in a 5% CO2 incubator. Cellswere then harvested by using a multiwell harvester(PHD Cambridge Technology). Briefly, cells and super-natant were washed with water through the harvester.Only incorporated 3H-thymidine-labeled DNA was col-lected onto glass-fiber filter disks. Additional waterwashes ensured complete removal of unincorporatedthymidine. A liquid scintillation counter (BeckmanLS6500, Beckman Coulter, Fullerton, CA, U.S.A.) wasused to measure levels of radioactivity in counts perminute (cpm) from collected filter disks. Backgroundvalues of thymidine incorporation of unstimulated cellswere subtracted from mitogen-induced proliferation.
Antibody Secretion
Some spleen cells were immediately cultured in 96-wellplates (2 ¥ 105 cells/well) at 37°C in a CO2 incubatoreither with LPS (10 mg/mL) or with cRPMI media onlyfollowing tissue harvest. Cell cultures were maintainedfor 6 days post harvest, and then cell supernatants werecollected for enzyme-linked immunosorbent assay(ELISA). ELISA plates (Costar Corning, Lowell, MA,U.S.A.) were coated with 10 mg/mL goat antimouse IgH + L unlabeled antibody (Southern Biotechnology, Bir-mingham, AL, U.S.A.). Cell supernatants were used at a1:5 dilution, while the secondary antibody of goat anti-mouse Ig H + L-HRP (Southern Biotechnology) wasused at a 1:1,000 dilution. Detection was measured witha substrate detection kit (Zymed, Zymed/Invitrogen,Carlsbad, CA, U.S.A.) according to the manufacturer’sprotocol, and plates were read at 405 nm on a standardplate reader (Bioteck EL800, BioTek Instruments,Winooski, UT, U.S.A.). Background values (media only)were subtracted from LPS-induced antibody secretionsamples.
Statistical Analysis
A Student’s t-test was used to assess significant differ-ences between morphine and placebo treatment groups.Differences between treatment groups were assessed byusing a randomized block design under analysis of vari-ance, where the source of variation is day and fixedeffects of treatment for all CTP experiments. Computa-tions were carried out by using the mixed model analysisprocedure in SAS (Version 9.1; SAS Institute, Cary, NC,
Citalopram Enhances B Cell Numbers • 197
U.S.A.). Statistical significance was defined as P < 0.05.All data are expressed as mean 1 SEM.
RESULTS
Morphine Reduces Spleen Weight and TotalSplenocyte Numbers
To induce immune suppression, animals were treatedwith morphine via subcutaneous pellet implantation for48 hours.22 To assess morphine’s effect on the immunesystem, spleen and thymus were harvested, and tissueweights were measured in relation to total body weight.Although the thymus did not exhibit weight differencesbetween morphine treatment and placebo treatment(data not shown), spleen weight per body weight wassignificantly (P < 0.05) reduced with morphine treat-ment (Figure 1A). Likewise, total cell numbers in the
spleen was significantly reduced (P < 0.05) when treatedwith morphine compared with placebo (Figure 1A).Morphine, however, did not affect distribution of Bcells, T cells, or NK cells in the spleen, as relative per-centages were not significantly different between mor-phine and placebo (Figure 1B). Thus, morphinesignificantly reduces the number of splenic lymphocytes.
CTP Does Not Reverse Morphine-Induced WeightLoss in Mice
Morphine-treated mice have been observed to have adrastic reduction in weight22 compared with theirplacebo counterparts, indicating that weight loss is symp-tomatic of morphine treatment (data not shown). There-fore, we examined whether CTP had an effect on bodyweight in this model. Following treatment, all animalswere weighed before tissue harvest. Figure 2A depicts theproportional change of the final body weight based onthe body weight at the beginning of the study for eachanimal. The proportional range for each treatment groupdid not vary greatly from one another. In Figure 2B,mouse body weights 48 hours after morphine implanta-tion were, on average, saline (16.37 1 0.70, n = 8);10 mg/kg CTP (15.6 1 0.22, n = 5); 30 mg/kg CTP(19.13 1 1.75, n = 4). Therefore, although animalstreated with 30 mg/kg CTP maintained the highestaverage weight (P > 0.05), the rate of weight change wassimilar to animals treated with 0 mg/kg or 10 mg/kgCTP.
CTP Increased Total Cell Numbers in theSpleen But Not in Other Lymphoid Organs
One indicator of morphine-induced immunosuppres-sion is the reduction of absolute number of spleen cellsand thymus cells in animals treated with morphine com-pared with placebo22 (Figure 1). Thus, we wanted toexamine CTP’s effect on total cell numbers in lymphoidorgans after morphine treatment. Cells from the spleen,thymus, and lymph nodes were collected from single-cellsuspensions and counted from animals receiving mor-phine and either 0 mg/kg, 10 mg/kg, or 30 mg/kg CTP.The total number of cells in the thymus and lymphnodes did not change with treatment of CTP (Figure 3).However, in the spleen, treatment with 30 mg/kg CTP(26.41 ¥ 106
1 4.31 ¥ 106 cells, n = 4) produced anonsignificant (P < 0.175) trend of increased numberof cells compared with treatment with saline(19.41 ¥ 106
1 3.23 ¥ 106 cells; n = 8) (Figure 3). Thus,CTP at the doses tested did not reverse morphine-
Figure 1. Implantation of morphine pellets significantly reducesspleen weight and lymphocyte numbers, but relative percent-ages of lymphocyte populations are not affected. Animals wereeuthanized 2 days after morphine or placebo pellet implanta-tion. (A) The mean spleen weight per unit body weight is shownwith SEM (n = 3; P = 0.005) on the left axis and the mean totalsplenocytes with SEM on the right axis (n = 3; P = 0.0006). (B)Percentages of T, B, and NK cells were determined by stainingsplenocytes with appropriate monoclonal antibodies and byusing flow cytometry to distinguish positive and negative popu-lations. No significant difference in splenocyte redistribution wasobserved by Student’s t-test analysis (P > 0.05). CTP, citalopram.
198 • nguyen et al.
induced cell loss in the thymus or lymph nodes but mayslightly enhance splenic cell number.
B Cell Populations in Lymph Nodes Were Increasedwith CTP Treatment
Because morphine has been shown to affect lymphocyteactivity, it was of interest to determine if CTP preferen-tially acted upon any lymphoid cell populations. Popu-lations of B cells, T cells, and NK cells were analyzed byflow cytometry for each harvested tissue. In the spleen,no significant differences (P > 0.05) were found amongB cell populations for all treatment groups (Figure 4).Likewise, average splenic T cell populations and averageNK cell populations were also found to be similar acrossall treatment groups (Table 1). Similarly, the thymus
was found to have levels of B cells, T cells, and NK cellsthat were not significantly different (P > 0.05) across alltreatment groups (Table 1).
Lymph nodes, however, were found to have a sig-nificant increase (P < 0.05) of B cell populations in30 mg/kg CTP-treated animals over saline-treatedanimals (Figure 5). Animals treated with 10 mg/kg CTPdisplayed similar B cell levels as their saline counterparts.However, T cell and NK cell populations in lymph nodeswere found to be similar, respectively, across all treat-ment groups. Collectively, these results indicate that thehigher dose of CTP can rescue B cell populations but notT cell and NK cell populations in lymph nodes. Addition-ally, CTP treatment resulted in little effect in the spleenand thymus lymphocyte population distributions.
CTP May Increase Splenic B andT Cell Proliferation
In addition to lymphocyte numbers, CTP could alsoaffect individual cell function. In fact, tramadol has beenreported to partially restore morphine-induced suppres-sion of PHA responses in humans undergoing surgery.17
Spleen cells were cultured for 48 hours with the appro-priate mitogen and then pulsed with tritiated thymidinefor an additional 24 hours to measure levels of B cell andT cell proliferation. LPS was used to stimulate B cellproliferation, while concanavalin A (Con A) stimulatedT cell proliferation. A liquid scintillation counter was
Figure 2. Citalopram (CTP) does not reverse morphine-inducedweight loss. Animals were weighed at the beginning and con-clusion of the study while receiving saline or CTP (10 mg/kg or30 mg/kg i.p. injection twice a day, days 0–9) and morphine(25 mg pellet, s.c. implantation, days 7–9). Proportion of finalweight to beginning weight for each individual is shown in A.Average mouse weights (g) at the time of harvest were similar(P > 0.05): saline (16.37 1 0.70, n = 8); CTP 10 mg/kg (15.6 1 0.22,n = 5); CTP 30 mg/kg (19.13 1 1.75, n = 4). Data are expressed asmean 1 SEM and shown in B.
Figure 3. The number of spleen cells may be enhanced withcitalopram (CTP) treatment. Spleen, thymus, and lymph nodeswere harvested following CTP and morphine treatment. Single-cell suspensions were made, and the total number of viable cellswas counted. In saline-treated animals (n = 8), splenic cell countswere 19.41 ¥ 106
1 3.23 ¥ 106 cells, while 30 mg/kg CTP-treatedanimals produced a higher but not significant (P < 0.175) numberof cells (26.41 ¥ 106
1 4.31 ¥ 106 cells, n = 4). Data are expressedas mean 1 SEM.
Citalopram Enhances B Cell Numbers • 199
used to measure levels of radioactivity in cpm. Treatmentwith mitogens increased proliferation of both B cells andT cells over media alone for all CTP treatment groups(data not shown). Proliferation of B cells after 30 mg/kgCTP treatment was increased, although not significantly
(P < 0.1), over saline-treated animals (Figure 6). Treat-ing animals with 10 mg/kg CTP resulted in similar B cellproliferation levels as their saline counterparts. Similarto B cells, proliferation of T cells was found to have anincreased although not significant trend (P < 0.1) in30 mg/kg CTP animals over saline animals. Treatinganimals with 10 mg/kg CTP resulted in similar T cellproliferation levels as those in animals treated withsaline. Thus, CTP may produce a slightly positive effecton lymphocyte proliferative abilities in the spleen.
CTP Has No Effect on Splenic CellAntibody Secretion
Another measure of cell function is in vitro antibodysecretion by mitogen-stimulated B cells. Because mor-phine has the ability to suppress B lymphocyte antibodyproduction,23 we wished to explore whether CTP couldrecover antibody secretion. Spleen cells were culturedalong with the mitogen LPS to induce antibody produc-tion. Antibody levels were measured by ELISA; there-fore, measurements were recorded as units ofabsorbance (nm). Across all three treatment groups(saline, 10 mg/kg CTP, 30 mg/kg CTP), levels of
Figure 4. Citalopram (CTP) does notchange splenic lymphocyte popula-tions during concurrent treatmentwith morphine. Following treatment,cells from a single-cell suspension fromthe spleen were treated to an ammo-nium chloride tris buffer to lyse redblood cells. Cells were labeled with amixture of T cell antibodies conju-gated to phycoerythrin (PE; anti-CD4-PE), B cell antibodies conjugated toCyChrome (anti-B220-CyChrome), andNK cell antibodies conjugated tofluorescein isothiocyanate (FITC; anti-NK1.1-FITC) analyzed by flow cytom-etry. Lymphocytes were gated andanalyzed for fluorescence. (A) Flowcytometry plots representative of oneexperiment in each treatment group.(B) Average percentages of B cell, Tcell and NK cell populations in thespleen. Saline (n = 8); 10 mg/kg CTP-treated animals (n = 5); 30 mg/kg CTP-treated animals (n = 4).
Table 1. Analysis of B Cell, T Cell, and NK CellPopulations in the Spleen, Thymus, and Lymph Nodes
Flow cytometry was carried out as in Figure 4. Data are expressed as mean 1 SEMand represent morphine + saline (n = 8), morphine + 10 mg/kg CTP (n = 5), andmorphine + 30 mg/kg CTP (n = 4).* denotes P < 0.1.§ denotes P < 0.05.CTP, citalopram.
200 • nguyen et al.
antibody secretion were similar (Figure 7; P > 0.05).Overall, antibody secretion was unchanged after CTPtreatment, indicating that CTP had no functional effecton B cell antibody production.
DISCUSSIONBesides its analgesic properties, morphine is knownto produce immunosuppressive effects, thus increas-ing the risk of opportunistic infections in patients.The purpose of our study was to determine whetherthe antidepressant CTP can reverse morphine-inducedimmunosuppression in mice. In humans, treatment withCTP has been reported to increase body weight.24,25
However, gross measurements of body weight in ourmouse model indicate that treatment with CTP did nothave an effect relative to saline. Although all mice wereobserved to have a dramatic decrease in body weightfollowing morphine implantation, the proportion of thefinal body weight over the original weight was found tobe similar across all treatment groups. Additionally, theaverage body weights for each treatment group were notsignificantly different.
The most profound effect of treatment with CTPbefore and after inducing immunosuppression was asignificant increase in the percentage of B cells in thelymph node. Treatment with 30 mg/kg CTP resulted inthe highest levels of B cells over animals that wereadministered saline. Surprisingly, animals that receivedthe lower dose of CTP had similar B cell levels as theirsaline counterparts. The total number of cells in thelymph node was found to be similar for all treatmentgroups, suggesting that an increase in total cell numberwas not responsible for the increase in the number of Bcells with 30 mg/kg CTP treatment. No significant dif-ferences were found for T cell or NK cell populationdistributions in the lymph node, indicating that the Bcell increase was not simply the result of loss of non-lymphoid cells in the lymph node. If the effect wassimply the consequence of loss of non-B cells, then wewould expect the relative percentages of NK and T cellsto increase as well. Thus, our results suggest that CTPmay have an effect specifically on B cells and not on Tcells or NK cells. At this time, it is unclear whether theappearance of B cells in the lymph nodes is the result of
Figure 5. Citalopram (CTP) treatmentincreases B cell populations in thelymph node but not T cell populations.Following treatment, single-cell sus-pensions were prepared and labeledsimilar to spleen cells with the excep-tion of red blood cells lysis. (A) Flowcytometry plots representative of oneexperiment in each treatment group.(B) Average percentages of B cell, T cell,and NK cell populations in the lymphnodes. Saline-treated animals (n = 8)had B cell populations of 21.36 1 2.84%cells; 10 mg/kg CTP-treated animals(n = 5) had 23.81 1 3.56% cells; finally,30 mg/kg CTP-treated animals (n = 4)had a significant increase (*denotesP < 0.05) in B cell populations (38.49 14.16% cells) over saline animals. Dataare expressed as mean 1 SEM.
Citalopram Enhances B Cell Numbers • 201
increased migration from other lymphoid compartmentsor increased local proliferation.
Treatment with CTP at either dose produced noeffect on B cell, T cell, or NK cell populations ortotal cell numbers in the thymus. Thus, CTP,in conjunction with morphine, does not have anyimmune effect on thymic cells. Finally, cell number inthe spleen was found to increase in a dose-dependentmanner with CTP treatment. Although the differenceswere not significant (P > 0.05), the increasing numberof splenic cells as CTP dosage increased may indicatethat CTP may have the ability to stimulate spleen cellproliferation. Likewise, proliferation of splenic T cellsand B cells after mitogen stimulation was higher,although not significant (P > 0.05), than that of theirsaline counterparts. However, relative percentages ofsplenic B cells, T cells, as well as NK cells were nearlyequivalent across all three drug treatment groups, sug-gesting that CTP did not preferentially affect a single-cell population. These results were surprising as weexpected CTP to act similarly to tramadol because tra-madol has been shown to increase PHA- and ConA-induced T cell proliferation.17,26 However, when tra-madol is administered in animals under a chronic con-striction pain model, T cell proliferation wassuppressed.27 There remains a possibility that neuro-pathic pain mechanisms differ based on the type ofpain, which may consequently affect the pharmacologyof analgesics that are administered to patients withpain.
Morphine-induced immunosuppression can nega-tively affect lymph node, spleen, and thymus cell num-bers.6 Additionally, subcutaneous injection of morphinecan affect splenic lymphocyte proliferative function butnot mitogen-induced proliferation of B cells and T cellsderived from mesenteric lymph nodes.28 This may indi-cate that morphine has no effect on B cell and T cellfunction derived from the lymph node despite adverseeffects on the spleen. In addition, the route of adminis-tration may also influence morphine’s effect on lymphnodes as we administered morphine via subcutaneousimplantation. Our studies suggest that there may be aslight effect of CTP on total lymphocyte numbers fromlymph nodes (19 1 3 million control treated vs. 26 1 4million 30 mg/kg CTP treated; Figure 2). Therefore, itmay be possible that, although morphine affects lympho-cyte cell numbers in both spleen and lymph nodes, theunderlying cause for increased B cell populations in thelymph node may be attributed to CTP’s differential effecton B lymphocytes from the lymph node vs. the spleen.
Figure 6. Proliferation of B and T cells in the spleen is not sig-nificantly affected by citalopram (CTP) treatment. Single-cellsuspensions prepared from spleen were cultured for 48 hourswith either LPS or Con A and then pulsed with tritiated thymi-dine (3H-Tdr) for 24 hours. Cells were harvested by using a cellharvester, and radioactivity was measured by liquid scintillation.Mitogens are lymphocyte specific: LPS stimulates B cell prolif-eration, Con A stimulates T cell proliferation. Treatmentwith 30 mg/kg CTP (n = 4; B cell: 4099.6 1 107.5 cpm; T cell:5048.5 1 377.8 cpm) results in an increased although not signifi-cant (P < 0.1) trend for both splenic B and T cell proliferation oversaline-treated animals (n = 8; B cell: 3761.6 1 218.9 cpm; T cell:3978.8 1 571.3 cpm). Data are expressed as mean 1 SEM. LPS,lipopolysaccharide.
Figure 7. Levels of splenic B cell antibody secretion are similarafter citalopram (CTP) treatment compared with saline alone.Spleen cells (2 ¥ 105 cells/well in a 96-well plate) were culturedand stimulated with LPS for 6 days. An ELISA was performed oncell supernatants to detect the amount of antibody secreted bythe spleen cells. Antibody production for all three treatmentgroups (saline, 10 mg/kg CTP, 30 mg/kg CTP) was similar(P > 0.05). Absorbance readings for saline-treated animals (n = 8)were 1.00 1 0.09. Treating animals with 10 mg/kg CTP (n = 5)resulted in readings of 1.37 1 0.09, while treatment with30 mg/kg CTP (n = 4) resulted in readings of 1.08 1 0.29. Dataare expressed as mean 1 SEM. LPS, lipopolysaccharide; ELISA,enzyme-linked immunosorbent assay.
202 • nguyen et al.
Thus, further studies are needed to examine the possibil-ity of an effect on lymph node-lymphocyte proliferationwhen co-treated with morphine and CTP.
Because chronic morphine treatment has beenshown to negatively affect levels of immunoglobulinsIgM and IgG,29 we wished to examine whether CTPcould reverse morphine’s effect on antibody produc-tion. Functional analysis of B cells was determined bymeasuring levels of antibody secretion, which werefound to be similar across all treatment groups.Thus, CTP is somewhat effective in modulatingcellular and humoral immunity because of evidencesuggesting some immunoenhancement of lymphocyteproliferation as well as B lymphocyte levels.
Citalopram is an SSRI that is administered as aracemic mixture of S(+)-enantiomer (escitalopram) andR(–)-enantiomer (R-citalopram).30 The mode of actionfor CTP is through the 5-HT transporter where it blocksreuptake of 5-HT into the presynaptic terminal. CTP hasa higher potency (IC50 is <10 nmol/L) of 5-HT reuptakeinhibition than many other SSRIs, including fluoxetine.31
Compared with many other SSRIs, CTP is highly selec-tive for 5-HT reuptake inhibition, while fluoxetine, acommon antidepressant, inhibits both 5-HT and norepi-nephrine reuptake.
Serotonin (5-HT) is a neurotransmitter that is knownto affect mood and appetite. In addition, 5-HT may alsoaffect the immune system. 5-HT transporters have beenfound on human lymphocytes,32 murine splenic mac-rophages33 as well as rat blood lymphocytes.34 Addition-ally, serotonin’s immunomodulatory effects have beenreported to be both positive and negative. Immunoen-hancing effects of 5-HT include an increase of NK cellactivity,35–37 increased B-cell proliferation,38 mediationof T-cell differentiation via dendritic cells and T-cellactivation,39,40 and differentiation of monocytes intodendritic cells.41 In particular, chronic CTP admini-stration has been shown to increase splenic T cell pro-liferation and anti-inflammatory cytokine production ofIL-10.42 Conversely, short-term treatment with fluoxet-ine has resulted in reduced T lymphocyte proliferation.43
Additionally, high doses of 5-HT have an inhibitoryeffect on B lymphocyte and T lymphocyte proliferation,whereas lower doses of serotonin did not show a stimu-latory effect.44 Thus, it appears that the type of antide-pressant and its dose have a differential effect onimmune modulation.
Our study demonstrates that CTP has a dose-dependent effect on certain immune parameters whencoadministered with morphine. High doses of CTP exert
the most effect, in particular, among B cell populationsin the lymph node. Additionally, our data suggest thatCTP has a slightly positive effect on cellular immunityby increasing total T cell number in the spleen as well asslightly increasing T cell proliferation. Further studiesare needed to examine whether other doses as well asroutes of administration (eg, localized treatment) of thisSSRI may increase its effect on lymphocyte number andfunction.
ACKNOWLEDGEMENTS
We would like to thank the National Institute of DrugAbuse for their donation of morphine pellets. This studywas funded by the Millennium Pain Center.
REFERENCES
1. Roy S, Wang J, Kelschenbach J, Koodie L, Martin J.Modulation of immune function by morphine: implicationsfor susceptibility to infection. J Neuroimmune Pharmacol.2006;1:77–89.
2. MacFarlane AS, Peng X, Meissler JJ, et al. Morphineincreases susceptibility to oral Salmonella typhimurium infec-tion. J Infect Dis. 2000;181:1350–1358.
4. Friedman H, Newton C, Klein TW. Microbial infec-tions, immunomodulation, and drugs of abuse. Clin MicrobiolRev. 2003;16:209–219.
5. Szabo I, Rojavin M, Bussiere JL, Eisenstein TK,Adler MW, Rogers TJ. Suppression of peritoneal macrophagephagocytosis of Candida albicans by opioids. J PharmacolExp Ther. 1993;267:703–706.
6. Wang J, Charboneau R, Balasubramanian S, BarkeRA, Loh HH, Roy S. The immunosuppressive effects ofchronic morphine treatment are partially dependent on corti-costerone and mediated by the mu-opioid receptor. J LeukocBiol. 2002;71:782–790.
7. Vallejo R, de Leon-Casasola O, Benyamin R. Opioidtherapy and immunosuppression: a review. Am J Ther.2004;11:354–365.
8. Saurer TB, Carrigan KA, Ijames SG, Lysle DT. Sup-pression of natural killer cell activity by morphine is mediatedby the nucleus accumbens shell. J Neuroimmunol. 2006;173:3–11.
9. Fecho K, Maslonek KA, Dykstra LA, Lysle DT. Evi-dence for sympathetic and adrenal involvement in the immu-nomodulatory effects of acute morphine treatment in rats. JPharmacol Exp Ther. 1996;277:633–645.
10. Pruett SB, Han YC, Fuchs BA. Morphine suppressesprimary humoral immune responses by a predominantly indi-rect mechanism. J Pharmacol Exp Ther. 1992;262:923–928.
Citalopram Enhances B Cell Numbers • 203
11. Millan MJ. Descending control of pain. Prog Neu-robiol. 2002;66:355–474.
12. Tao R, Auerbach SB. Opioid receptor subtypesdifferentially modulate serotonin efflux in the rat cen-tral nervous system. J Pharmacol Exp Ther. 2002;303:549–556.
13. Nemmani KV, Mogil JS. Serotonin–GABA interac-tions in the modulation of mu- and kappa-opioid analgesia.Neuropharmacology. 2003;44:304–310.
14. Vonvoigtlander PF, Lewis RA, Neff GL. Kappaopioid analgesia is dependent on serotonergic mechanisms. JPharmacol Exp Ther. 1984;231:270–274.
15. Song B, Chen W, Marvizon JC. Inhibition of opioidrelease in the rat spinal cord by serotonin 5-HT(1A) receptors.Brain Res. 2007;1158:57–62.
16. Zhao ZQ, Gao YJ, Sun YG, Zhao CS, Gereau RW4th, Chen ZF. Central serotonergic neurons are differentiallyrequired for opioid analgesia but not for morphine tolerance ormorphine reward. Proc Natl Acad Sci USA. 2007;104:14519–14524.
17. Sacerdote P, Bianchi M, Gaspani L, et al. The effectsof tramadol and morphine on immune responses and painafter surgery in cancer patients. Anesth Analg. 2000;90:1411–1414.
18. Gaspani L, Bianchi M, Limiroli E, Panerai AE, Sac-erdote P. The analgesic drug tramadol prevents the effect ofsurgery on natural killer cell activity and metastatic coloniza-tion in rats. J Neuroimmunol. 2002;129:18–24.
19. Fredricson Overo K. Kinetics of citalopram in testanimals; drug exposure in safety studies. Prog Neuropsychop-harmacol Biol Psychiatry. 1982;6:297–309.
20. Laakso A, Palvimaki EP, Kuoppamaki M, SyvalahtiE, Hietala J. Chronic citalopram and fluoxetine treatmentsupregulate 5-HT2c receptors in the rat choroid plexus. Neu-ropsychopharmacology. 1996;15:143–151.
21. Thrybom T, Rooth P, Lindstrom P. Effect of serotoninreuptake inhibitor on syndrome development in obese hy-perglycemic mice (Umea ob/ob). Metabolism. 2001;50:144–150.
22. LeVier DG, McCay JA, Stern ML, et al. Immuno-toxicological profile of morphine sulfate in B6C3F1 femalemice. Fundam Appl Toxicol. 1994;22:525–542.
23. Vassou D, Bakogeorgou E, Kampa M, Dimitriou H,Hatzoglou A, Castanas E. Opioids modulate constitutiveB-lymphocyte secretion. Int Immunopharmacol. 2008;8:634–644.
24. Dannon PN, Iancu I, Lowengrub K, et al. A natural-istic long-term comparison study of selective serotoninreuptake inhibitors in the treatment of panic disorder. ClinNeuropharmacol. 2007;30:326–334.
25. Flechtner-Mors M, Jenkinson CP, Alt A, Adler G,Ditschuneit HH. Metabolism in adipose tissue in response tocitalopram and trimipramine treatment–an in situ microdialy-sis study. J Psychiatr Res. 2008;42:578–586.
26. Sacerdote P, Bianchi M, Manfredi B, Panerai AE.Effects of tramadol on immune responses and nociceptivethresholds in mice. Pain. 1997;72:325–330.
27. Tsai YC, Won SJ. Effects of tramadol on T lympho-cyte proliferation and natural killer cell activity in rats withsciatic constriction injury. Pain. 2001;92:63–69.
29. Hashiguchi S, Morisaki H, Kotake Y, Takeda J.Effects of morphine and its metabolites on immune functionin advanced cancer patients. J Clin Anesth. 2005;17:575–580.
30. Sanchez C. The pharmacology of citalopram enanti-omers: the antagonism by R-citalopram on the effect ofS-citalopram. Basic Clin Pharmacol Toxicol. 2006;99:91–95.
31. Frazer A. Serotonergic and noradrenergic reuptakeinhibitors: prediction of clinical effects from in vitro potencies.J Clin Psychiatry. 2001;62(Suppl 12):16–23.
32. Marazziti D, Rossi A, Giannaccini G, Baroni S,Lucacchini A, Cassano GB. Presence and characterization ofthe serotonin transporter in human resting lymphocytes. Neu-ropsychopharmacology. 1998;19:154–159.
33. Jackson JC, Walker RF, Brooks WH, Roszman TL.Specific uptake of serotonin by murine macrophages. Life Sci.1988;42:1641–1650.
34. Cedeno N, Urbina M, Obregon F, Lima L. Charac-terization of serotonin transporter in blood lymphocytes ofrats. Modulation by in vivo administration of mitogens. JNeuroimmunol. 2005;159:31–40.
35. Nunez MJ, Balboa J, Rodrigo E, Brenlla J, Gonzalez-Peteiro M, Freire-Garabal M. Effects of fluoxetine on cellularimmune response in stressed mice. Neurosci Lett. 2006;396:247–251.
36. Hellstrand K, Hermodsson S. Enhancement ofhuman natural killer cell cytotoxicity by serotonin: role ofnon-T/CD16+ NK cells, accessory monocytes, and 5-HT1Areceptors. Cell Immunol. 1990;127:199–214.
37. Evans DL, Lynch KG, Benton T, et al. Selective sero-tonin reuptake inhibitor and substance P antagonist enhance-ment of natural killer cell innate immunity in humanimmunodeficiency virus/acquired immunodeficiency syn-drome. Biol Psychiatry. 2008;63:899–905.
38. Iken K, Chheng S, Fargin A, Goulet AC, Kouassi E.Serotonin upregulates mitogen-stimulated B lymphocyte pro-liferation through 5-HT1A receptors. Cell Immunol. 1995;163:1–9.
39. O’Connell PJ, Wang X, Leon-Ponte M, Griffiths C,Pingle SC, Ahern GP. A novel form of immune signalingrevealed by transmission of the inflammatory mediator sero-tonin between dendritic cells and T cells. Blood. 2006;107:1010–1017.
204 • nguyen et al.
40. Yin J, Albert RH, Tretiakova AP, Jameson BA. 5-HT(1B) receptors play a prominent role in the proliferation ofT-lymphocytes. J Neuroimmunol. 2006;181:68–81.
41. Katoh N, Soga F, Nara T, et al. Effect of serotoninon the differentiation of human monocytes into dendritic cells.Clin Exp Immunol. 2006;146:354–361.
42. Kubera M, Simbirtsev A, Mathison R, Maes M.Effects of repeated fluoxetine and citalopram administrationon cytokine release in C57BL/6 mice. Psychiatry Res.2000;96:255–266.
43. Pellegrino TC, Bayer BM. Modulation of immunecell function following fluoxetine administration in rats. Phar-macol Biochem Behav. 1998;59:151–157.
44. Stefulj J, Cicin-Sain L, Schauenstein K, Jernej B. Sero-tonin and immune response: effect of the amine on in vitroproliferation of rat lymphocytes. Neuroimmunomodulation.2001;9:103–108.
