47

Molecular and Biochemical Parasitology, 48 (1991) 4%58 ~(- 1991 Elsevier Science Publishers B.V. 0166-6851/91/$03.50 ADONIS 0166685191002823 MOLBIO 01576

Induction and localization of Plasmodium falciparum stress proteins related to the heat shock protein 70 family N i r b h a y K u m a r l, Gary Koski l, M a s a k a d u Harada 2, M a s a m i c h i A i k a w a 2

and Hong Zheng 1 IDepartment g/Immunology and Infectious Diseases, School (~f Itygiene and Public Health, Johns Hopkins University. Baltimore, MD, U.S.A., and 2Institute of Pathology, Case Wes.tern Reserve University, Cleveland. Oil, U.S.A. (Received 11 February 1991; accepted 3 April 1991)

Induction of heat shock-related stress proteins Pfhsp and Pfgrp, similar in sequence to hsp70 (heat shock protein) and grp78 (glucose-regulated protein), respectively, was studied in culture-derived parasite Plasmodium jalciparum. Elevation in temperature from 2 6 C to 37~'C and higher caused significant induction of Pfhsp with a moderate effect on the synthesis of Pfgrp also. Synthesis of Pfgrp, however, was not induced by partial glucose deprivation. On the contrary, lack of glucose in the medium resulted in cessation of protein synthesis in the parasites. Other known inducers of grp synthesis in mammalian cells, i.e., calcium ionophore A23187 and inhibitors of glycosylation (tunicamycin, 2-deoxy glucose) were also without any apparent effect on the synthesis of Pfgrp. Heat shock-induced responses were transient in nature: removal of stress caused repression of these responses. The effect of glucose deprivation was only partially reversible with better recovery if parasites were subjected to glucose starvation at 2 6 C than at 3TC. Northern blot analysis and in vitro translation of mRNA revealed a parallel increase in the levels of m R N A for Pfhsp upon heat shock, lmmuno-gold electron microscopy with cultured parasites revealed nuclear location of Pfhsp and primarily cytoplasmic (probably endoplasmic reticulum) location of Pfgrp. These findings suggest that SDEL (carboxy terminal sequence of Pfgrp) might play a similar role in the cellular localization of Pfgrp as does the sequence KDEL in mammalian cells and HDEL in yeast. Key words: Heat shock protein; Stress protein; Malaria: Plasmodium [ah'iparum: Endoplasmic reticulum, Host-parasite adaptation

Introduction Heat shock-related stress proteins (hsp) are produced when cells (prokaryotic and eukaryotic) are subjected to environmental stress. Members of the hsp70 family include proteins synthesized in response to elevated temperature or heavy metals known as hsp70 and others that exhibit increased synthesis when cells are deprived of glucose, oxygen or Correspondence address." Nirbhay Kumar, DllD-SHPH-JHU, 615 N. Wolfe St., Baltimore, MD 21205, U.S.A. Abbreviations." hsp, heat shock protein; grp, glucose-regulated protein; Pfhsp, Plasmodium falciparum hsp70; Pfgrp, P. fidciparum grp78; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; EM, electron microscopy: ER, endop[asmic reticulum; TCA, trichloroacetic acid.

exposed to chemicals causing perturbation of glycosylation and calcium homeostasis, known as glucose-regulated proteins (grp) (reviewed in refs. 1-3). Proteins like hsp70 are found in the cytoplasm and nucleus and the grp78 in the lumen of endoplasmic reticulum (ER). A variety of functions have been postulated for hsp70 family stress proteins. Based on the initial finding that these stress proteins bind ATP with high affinity [3 5] and studies following this observation have led to the concept of their potential involvement in folding and unfolding of proteins and in post-translational protein translocation in the mitochondria and ER [6,7]. Proteins like grp78 in mammalian cells have a hydrophobic signal sequence at the amino terminus and a tetrapeptide sequence of Lys-Asp-Glu-Leu (KDEL)

48 at the carboxy terminus, which direct them and are essential for their retention in the lumen of ER, respectively [8]. It has been suggested that grp78 by regulating the processing of proteins inside the ER may play a role in the normal assembly of membrane-bound and secreted proteins [9]. Expression of aberrant polypeptides in the cells, whose transport from the ER is defective, results in the induction of synthesis of grp78 [10]. On the other hand, grp78 forms a tight complex with misfolded or aberrantly glycosylated proteins and probably assists in proper assembly and trafficking of membrane or secreted pioteins, as shown using a coupled translation-translocation system [11]. In recent years, genes for similar stress proteins in many parasitic systems have been identified (see ref. 12 for a review). It is of course not surprising to find heat shock related genes in parasites which spend a part of their life cycle in the poikilothermic environment in the invertebrate vector and the remainder in a warm blooded vertebrate host [13]. These proteins exhibit significant sequence similarities with those found in higher eukaryotes and mammals. Plasmodium falciparum hsp70 (Pfhsp) has an apparent molecular mass of 75000 and shows > 6 0 % sequence homology with Xenopus or Drosophila hsp70 [14~17]. Another P. falciparum gene for a protein designated as Pfgrp (72 kDa), similar in sequence to rat or hamster grp78 has also been cloned [16,18]. Sequence upstream of the 5' end of the cloned Pfgrp insert reported in [19] and additional sequencing has revealed an overall similarity of 67% in a 531-amino acid overlap with mammalian grp78 (Kumar and Zheng, unpublished). A significant difference in the Pfgrp sequence is that at the carboxy terminus: Ser-Asp-Glu-Leu (SDEL) as compared to KDEL in mammalian systems. These findings of stress proteins in the malaria parasite have raised a number of interesting questions. Are the genes for Pfhsp and Pfgrp in the parasite inducible by various stimuli of physiological significance? An equally important question is raised by the sequence SDEL, unlike K D E L in mammalian systems, at the carboxy terminus of the Pfgrp. If the

Pfgrp is a homologue of grp78, can the sequence of SDEL function as signal for its localization in the ER? If not, what is the function of Pfgrp in the parasite? Studies presented here address some of these questions.

Materials and Methods

Parasites and antibodies. Cloned parasites 7G8 (a clone of isolate IMTM22) and 3D7 (a clone of isolate NF54) were grown in culture as described [20]. Antibodies against fusion proteins of Pfhsp and Pfgrp have been described elsewhere [16]. A synthetic peptide corresponding to the last 11 amino acids (SGDEDVDSDEL) at the carboxy terminus of Pfgrp was conjugated to keyhole limpet hemocyanin for the production of Pfgrp specific antiserum in rabbits.

Induction of PJhsp and PJgrp. Cultures at 10% or higher parasitemia were washed twice at 26"C with cysteine and methionine-free medium RPMI-1640 (prepared using SelectA-Amine kit, Gibco) and incubated at various temperatures in 1 ml of the same medium for differing time periods. At the end of preincubation, 100/~Ci Trans 35S-label, (sp.act./> 1005 Ci mmol ~, ICN) was added and incubation continued for additional time periods at respective temperatures. To study the effect of glucose deprivation, parasites were washed and incubated at 26°C in RPMI-1640 (Cys and Met-free) media containing different amounts of glucose (5, 2 (normal), 1, 0.5, 0.25, 0.125, 0.06 and 0 g 1 1) prepared by serially diluting glucose in the cysteine and methionine free medium with medium lacking glucose, cysteine and methionine. At the end of incorporation period, cells were washed (2 x 1.0 ml, 1 min using 'Dade' immufuge) with medium RPMI1640 at 26>C and extracted in 1.0 ml of Triton X-100 buffer containing protease inhibitors as described [21].

Immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis. Extracts containing equal number of

49

trichloro acetic acid (TCA) precipitable counts were incubated in a total volume of 100-150 pl with l0 ~tl of antibodies at 4°C overnight. Protein A-Sepharose (100 #1 of 25% suspension, Pharmacia Fine Chemicals) was added and tubes rocked for 1 h at room temperature. The beads were washed two times with NETTS (10 mM Tris/1 mM EDTA/0.5 M NaC1/0.5% Triton X-100, pH 7.6), twice with N E T T (same as NETTS but containing only 0.15 M NaC1) and finally with NET (same as N E T T but without Triton X-100). Beads were extracted in 40 lal boiling SDS-sample buffer (62.5 m M Tris pH 6.8/10% glycerol/5% SDS/0.004% bromophenol blue/2.5% 2-mercaptoethanol) and analyzed by 5 15% gradient SDS-PAGE [22]. Gels were treated with Fluoro-Hance (Research Product International), dried and exposed with Kodak XAR-5 films at - 7 0 ° C for fluorography.

RNA analysis and in vitro translation. RNA was isolated using guanidine isothiocyanate/ phenol-chloroform extraction method [23]. Total cellular R N A (10 ktg per lane) was fractionated on formaldehyde-agarose gels, transferred to nitrocellulose membranes and probed with 32p-labeled Pfhsp-specific oligonucleotides [16]. Prehybridization and hybridization were carried out in the buffer containing 5 x SSC/5 x Denhardt's solution/0.1% SDS/ 0.1 mg ml-1 sonicated salmon sperm D N A at 37°C. Filters were washed at 35°C in 0.2 x SSC and 0.1% SDS and exposed with Kodak XAR-5 films at - 7 0 ° C [24]. /3-emission from Northern blots was imaged and quantitated directly using a Betascope 603 Blot Analyzer (Betagen). R N A samples (5 gg) were translated using in vitro rabbit reticulocyte lysate translation system (Promega) and Trans 35S-label. Translated products were analyzed by SDSPAGE after immunoprecipitation with rabbit antiserum against Pfhsp and Pfgrp peptide.

buffer. Fixed cells were dehydrated with ethanol and embedded in LR White resin as described previously [25]. For immuno-EM on cryo-fixed parasites, a small drop of the concentrated suspension of infected erythrocytes was placed on a 5 x 3 mm rectangle of 'Millipore' polycarbonate filter (Type RA, 1.2 gm) that had been presoaked in culture medium and placed on a 5 x 3-mm brass plate supported by a 1-mm-thick rectangle of 2.5% agar. These were positioned in the center of a 10-mm square of thermanox plastic and attached to a 10-mm cube of foam plastic. This unit was slam-frozen onto a liquid nitrogencooled metal mirror freezing device (Life Cell Corporation, The Woodlands, TX) [26]. The samples were transferred immediately to liquid nitrogen. The brass plate and attached filter were pried from the thermanox/foam support with the tip of a sharp scalpel and transferred under liquid nitrogen to a solid copper specimen holder. The specimen holder was transferred in liquid nitrogen to a molecular distillation dryer (Life Cell Corp.) for removal of amorphous phase tissue water without ice crystal damage [27]. The sample was stabilized with a 10 min exposure to para-formaldehyde vapor, dried under vacuum and brought to room temperature. The polycarbonate filter with attached cryo-fixed parasites was subsequently infiltrated with LR White resin for 4 h at room temperature and polymerized overnight at 50~'C in a gelatin capsule. Ultrathin sections were cut with a diamond knife, mounted on nickel grids and etched. Sections were blocked and incubated with dilutions of primary antibodies and protein A-gold as described elsewhere [28]. Control grids were incubated with pre-immune rabbit serum and protein A-gold or with protein A-gold alone to confirm the specificity of reactivity of antisera.

Results

Immuno-electron microscopy. Cultures oferythrocytic parasites (7G8 clone) were fixed for 10 min with 1% paraformaldehyde/0.2% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4 and washed with 0.1 M phosphate

Induction of PJhsp and PJgrp. Initially bloodstage parasites (containing asexual stages and a few sexual stages also) cultured in vitro were preincubated at 26, 37, 39 and 41°C for 0, 1,

50 N

H

G

H

G

H

G

H

30000-

G

A

37 c

39 c

116 97

20000

:E

D.. O

68

10000.

45

i 20

i 40

l 60

80

TIME (rain) Temp.

26

37

39

41

Fig. l. Effect of temperature on the induction of Pf'hsp and Pfgrp. Cultured parasites preincubated for 1 h were labeled at indicated temperatures for 1 h. Extracts containing equal number of TCA-precipitable counts were immunoprecipitated with normal rabbit sera (lanes N), rabbit antiserum against Pfhsp fusion protein (lanes H) and rabbit antiserum against Pfgrp peptide (lanes G). Positions of Pfhsp (75 kDa) and Pfgrp (72 kDa) are indicated by arrowheads. Numbers on the right show the positions of molecular weight standards in kDa.

and 2 h following which they were labeled at respective temperatures for 1 h. Triton X-100 extracts of labeled cells containing equal number of TCA-precipitable counts were immunoprecipitated using antisera against Pfhsp and Pfgrp peptide. Fig. 1 shows the results from parasites with 1 h preincubation and labeling for one h at various temperatures. Incubation of parasites at increasing temperature caused induction of Pfhsp (lanes H). Synthesis of Pfgrp (lanes G) was also induced by increase in temperature from 26 to 37~C without any further apparent induction thereafter. In order to address early time kinetics of induction of Pfhsp, parasites were either labeled continuously for 15, 30, 45 and 60 min at 26, 37 and 3WC (Fig. 2) or preincubated for 0, 15, 30 and 45 min at various temperatures and then labeled for 15 min at respective temperatures (Fig. 3). Extracts containing equal number of TCA precipitable counts were immunoprecipitated using antiserum

B

1

2

3

26 C

1

2

3

~

~

37 C

1

2

3

O Q Q 39 C

Fig. 2. Early time kinetics of incorporation of Trans 35S-label at various temperatures. Cultured parasites were labeled continuously for 15, 30, 45 and 60 min at 26, 37 and 3 9 C and incorporation measured by TCA precipitation (A). Equal number of TCA precipitable counts from 15, 30 and 60 min samples, (lanes 1,2,3 respectively, panel B), were used for immunoprecipitation with antisera against Pfhsp.

against Pfhsp. Combined results in Figs. 2 and 3 demonstrate that maximum induction of Pfhsp occurs at temperatures > 37°C between 30-60 min of heat shock. In 9 separate observations on induction at 1 h, densitomettic scanning of autoradiograms (LKB Ultroscan XL-Enhanced laser densitometer), suggested 2.5-fold (1.4-6.9) induction of Pfhsp at 37°C and 7.2-fold (3.7 14,2) induction at 39°C as compared to synthesis at 26°C.

EJfect of glucose deprivation. To test if glucose deprivation of parasites can induce the synthesis of Pfgrp, parasites were preincubated for one h in the medium containing different concentrations of glucose before the addition of Trans 35S-label for an additional 1 h. Fig. 4 shows the effect of glucose deprivation on total protein synthesis and specific Pfhsp and Pfgrp after immunoprecipitation. Reduction in the amount of glucose from 2 g

51

120

I

A 15000

A

100

o= ,,,,,I

80

I-Z

10000

0

60

O

13.

o

• 40 5000

1

20

0

26

37

0.06

1 2

a

4

1

2

3

4

0.25

0.5

1

B 1

2

2

5

G L U C O S E (g I"t )

39

TEMPERATURE (C) B

0.125

3

4

4SS

1

2 ~

C

3 ~

4

GRP

Fig. 3. Effect of preincUbation at various temperature~ Cultured parasites were preincubated at 26, 37 and 39~C fol 0 (open bars), 15 rain (stippled bars), 30 rain (hatched bars) and 45 min (filled bars). Trans 35S-label was then added to each culture for 15 min at the reSpeCtive temperatures and incorporation measured by TCA ptecipitlation (A). Equal number of TCA precipitable counts from 0 rain (lanes 1), 15 min (lanes 2), 30 min danes 3) and 45 rain preincubation (lanes 4) were used for immunoprecipitation with antisera against Pfhsp (B).

1- t (normal RPMl-1640 medium concentration) to 0.25 g 1-z had no apparent effect on the incorporation of Trans 35S-label. A further reduction in glucose concentration, however, resulted in inhibition of total protein synthesis with cessation if there was no glucose in the medium. Immunoprecipitation analysis of labeled extracts also revealed that Pfgrp was not induced under conditions of partial glucose deprivation. Other inducers of expression of grp78 in higher eukaryotes namely inhibitors of glycosylation such as 2-deoxyglucose (10 and 20 mM) and tunicamycin (0.5 and 1 /~g ml-z), and calcium ionophore A23187 (5 and 10 #M) were also without any significant effect, either at 26°C or at 37°C, in the parasite (data not shown)

Recovery from stress of heat shock and glucose

5

qNIl~

6

1 ~

3

4

9up i p

2

IP

5

6

tnR, O

HSP

Fig, 4. Effect of glucose deprivation. Parasites were preincubated at 26cC in the medium containing 0, 0.06, 0.125, 0.25.0.5, 1.0, 2.0 and 5.0 g 1 i D-glucose for I h followed by additional h in the presence of Trans 3SS-label. Incorporation was measured by TCA precipitation and results (A) are expressed as % of incorporation at 2 g I i (100%). Extracts from parasites containing equal number o f T C A precipitable counts at () (lanes 1), 0.25 (lanes 2), 0.5 (lanes 3), 1.0 (lanes 4), 2.0 (lanes 5) and 5,0 g I - L (lanes 6) ~-glucose were immunoprecipitated using rabbit antisera against Pfgrp. peptide (B) and Pfhsp fusion protein (C).

deprivation. After incubation of parasites at 39°C (condition for induction) for one h they were quickly equilibrated to 26°C and allowed to recover for different periods (15, 30 and 60 min) prior to labeling for 1 h at 26":C. As shown in Fig. 5, return of parasites to lower temperature (from 39°C to 26~'C) resulted in repression of the synthesis of Pfhsp. Effect of glucose starvation and reversal was temperature dependent. As mentioned before, lack of any glucose in the medium resulted in cessation of protein synthesis at all temperatures. However, if parasites after 1 h incubation in the glucose-free medium were shifted to glucose (2 g 1 ~) containing medium, total protein synthesis recovered to approx. 25% and approx. 16% in parasites maintained during deprivation and recovery at 2 6 C and 37°C, respectively. In some experiments suppression of protein synthesis caused by incuba-

52

A

B

C

D E

A

B

C

D

E

200

116 97.4"-" ~ t

66-43--

HSP

GRP

Fig. 5. Removal of heat stress results in the repression of synthesis of Pfhsp. Parasites were preincubated at 3 9 C for I h. Temperature was then quickly equilibrated to 2 6 C and preincubation continued for additional 15 min (C), 30 rain (D) and 60 min (E). At the end of each recovery period Trans 35S-label was added and parasites were labeled at 2 6 C ['or 1 h. For comparison parasites without any preincubation were also labeled at 26"C (A) or 39'C (B) for 1 h. Total incorporation as determined by TCA precipitation was 155% in (B), 83% in (C), 93% in (D) and 66% in (E) as c o m p a r e d to 100% incorporation (60140 cpm) in (A). Extracts containing equal number of TCA-precipitable counts were immunoprecipitated using antisera against Pfhsp fusion protein and Pfgrp peptide.

tion of parasites at 37°C in the glucose-free medium could not be reversed or restored after return of parasites to glucose-containing medium (not shown).

Effect of heat shock on the levels of mRNA and in vitro translation. Northern blot analysis of RNA samples from parasites incubated at various temperatures (26, 32, 37, 39°C) for 30, and 60 rain using Pfhsp-oligonucleotide as a hybridization probe in two separate experiments, showed an increase in the amount of Pfhsp-specific m R N A in heat shocked parasites (Fig. 6 A,B). Quantitation of images of fl-emission (corrected for the background in corresponding lanes) from these Northern blots using Betascope Blot Analyzer revealed 2.3- and 5-fold increases (in 2 separate experiments) in the levels of transcripts in the parasites exposed to 39°C for 60 min as compared to parasites incubated at 26°C. RNA samples from parasites incubated at various temperatures for 60 rain were translated using rabbit reticulocyte lysate. Translated products either in equal volume (panels a) or in equal TCA-precipitable counts (panels b) were analyzed by SDS-PAGE after immunoprecipitation using antisera against Pfhsp and Pfgrp peptide. Data in Fig. 6C show increased amounts of Pfhsp in the translation products from m R N A from heat shocked parasites. The amount of Pfgrp in the translation products of mRNA from heat-shocked parasites was only slightly higher as compared to amounts of Pfhsp. Localization of PJhsp and PJgrp in parasites. Another question investigated in this study pertains to the difference in the amino acid sequence at the carboxy terminus of Pfgrp: SDEL [16,18] as compared to KDEL [8] in mammalian cells. Antiserum against Pfgrp peptide was employed in immuno-EM with asexual and sexual (gametocyte) stage parasites of P. falciparum. For comparison, antisera against Pfhsp and Pfgrp fusion proteins were also used in these studies. The

Fig. 6. Effect of heat shock on relative transcript levels in parasites and in vitro translation. Parasites (2 separate experiments, A, B) were incubated at 26, 32, 37, 39~C for 30 rain (lanes 1,3,5,7) and 60 min (lanes 2,4,6,8). Total R N A (10 #g per lane) was fractionated by formaldehyde-agarose gel electrophoresis, and transferred to nitrocellulose membranes. Filters were hybridized with end-labeled oligonucleotide specific for Pfhsp at 37'C and washed at 3 5 C . (C) R N A samples (5/lg) from 60 rain heat shocked parasites (26"C, lanes 1; 3 T C , lanes 2 and 3 9 C , lanes 3) were translated in vitro in the rabbit reticulocyte lysate system. Translation products in equal volume (panels a) and equal TCA precipitable counts (panels b) were analyzed by SDS-PAGE after immunoprecipitation with rabbit antisera against Pfhsp and Pfgrp peptide. Position of the primary product is denoted by an arrow. -,

53

A 1

2

'~"~'~'~"~ "~

3

4

"~' ~

'~"~'"~'~'

5

B 6

7

2

3

4

5

32

!~!ii!~!~i~~!~

37

39

26

37

39

C 1

23

1 23

1 2

3

123

F 200--

11697.466-

qlD dip

~ Q O

43-

a

6

~ O ¸~!~u ~i~

26

8

(HSP)

b

a

(GRP)

b

54

55

anti-peptide antibodies are highly specific for the carboxy terminal sequence of Pfgrp and do not cross react in immunoprecipitation analysis with mouse and human grp (data not shown). Antibodies against Pfhsp fusion protein labeled predominantly nuclei of all developmental stages of erythrocytic parasites, including rings, trophozoites, schizonts and gametocytes (Fig. 7A). Gold label was scattered diffusely over the nucleoplasm of the parasites. Occasional localization in the cytoplasm was also observed (not shown). Antibodies against Pfgrp fusion protein and against Pfgrp peptide-labeled discrete regions in the cytoplasm of trophozoites, schizonts and gametocytes that are reminiscent of membranes of the ER (Fig. 7 B, C). In order to further improw," the localization of Pfgrp,

parasites were fixed by the technique of slamfreezing in which specimens are cryo-fixed and then stabilized in the vapors ofpara-formaldehyde after dehydration to minimize ice crystal damage. Results obtained with anti-Pfgrp peptide sera in Fig. 8 further demonstrate association of gold particles with the dilated membranous structures (probably ER) in the cytoplasm.

Discussion

Earlier studies have shown that malaria parasite (P. falciparum) expresses genes encoding proteins of 75 kDa (Pfhsp) and 72 kDa (Pfgrp), similar in sequence to hsp70 and grp78, respectively [14-19]. In this paper we have studied synthesis of these two stress

Fig. 8. Section o f a cryo-fixed schizont incubated with anti-Pfgrp-peptide antiserum as in Fig. 7B. Arrows denote decoration of dilated ER membranes by gold particles, x 44000.

+__

Fig. 7. lmmuno-gold localization of Pfhsp and Pfgrp. (A) Sections of a mature segmenter incubated with rabbit antiserum against Pfhsp fusion protein and protein A-gold. The nuclei (N) of budding merozoites (m) are labeled diffusely with colloidal gold particles (arrows). (B) Sections of a mature gametocyte incubated with antiserum against Pfgrp peptide and protein A-gold. Label (arrows) is associated with membranes of the ER. (C) Sections of a schizont incubated as in (B). Label (arrows) is associated with discrete areas of the schizont cytoplasm that correspond to ER. Note absence of label on the nucleus (N) and parasitophorous vacuole (PV). Bar 0.5 ~m.

56

proteins in parasites subjected to elevation in temperature and partial glucose deprivation. While both the proteins were induced, the effect of temperature was more noticeable on Pfhsp. Unlike the effect of heat shock, partial glucose deprivation had no significant effect on the induction of either of two stress proteins. In mammalian cells the synthesis of grp78 induced by glucose deprivation is only slightly affected by shift in temperature [29] and often, the two groups of proteins are regulated in inverse fashion (see ref. 30 for a review). In yeast, on the other hand, synthesis of grp78like proteins (a product of KAR2 gene) has been shown to be stimulated by heat shock [31]. Other inducers such as tunicamycin, 2deoxyglucose, and calcium ionophore A23187 which affect grp78 in mammalian cells were also without any effect on Pfgrp and Pfhsp synthesis. Mechanisms of regulation of Pfgrp in the parasites thus remain unclear. Stress responses in the parasite are transient in nature as removal of heat stress resulted in reversal to levels of synthesis observed at lower temperature. Effects of glucose starvation are particularly interesting as it was found to be a temperature-dependent phenomenon. Within the vertebrate host, malaria parasites infect red blood cells and spend a major part of their life as the intracellular parasites. Red blood cells and parasites do not have any reserve of glucose storage and are dependent upon availability of glucose in the circulation for their energy requirement [32]. In view of total dependence on glucose supply from the circulation, it is not surprising to observe an immediate effect of lack of glucose in the medium on cellular function. First of all, there was cessation of protein synthesis in the parasites with no induction of Pfgrp in the absence of glucose. Secondly, the recovery of parasites from glucose stress was temperaturedependent. A combination of elevated temperature and depletion of glucose in the medium appears to cause irreversible inactivation of metabolic activity in the parasites. Induction of Pfhsp is at least partially due to increased levels of specific m R N A in the parasites. In earlier studies, different regula-

tory mechanisms during heat shock respons e s - p r i m a r i l y transcriptional in yeast and both transcriptional and translational in Drosophila--have been described (see ref. 1 for a review). Theodorakis et al. have also suggested that translational control of heat shock responses in chicken reticulocytes is regulated at the level of elongation of nascent chains during protein synthesis [33]. Localization of Pfhsp and Pfgrp in the parasite was investigated by immuno-gold EM using ultra-thin sections prepared from 37°C cultured parasites. Pfhsp was predominantly localized in the nucleus with no surface and occasional cytoplasmic localization. Since a change in temperature from 26°C to 37~'C caused induction of Pfhsp in the parasites, these observations are similar to those reported for Drosophila wherein heat shock caused nuclear concentration of hsp70 [34]. Using a highly specific antiserum (anti-peptide) Pfgrp was localized in the cytoplasm, corresponding to ER-like membranous structures. The antipeptide serum used in these studies does appear to cross react with a few other highmolecular-weight proteins (Fig. 5). However, when these antibodies were used for immunoaffinity purification of Pfgrp from parasites, only a single band was observed in SDS-PAGE (Kumar, unpublished observation). Recently, in experiments with mouse malaria parasites (P. berghei) and a different stage during life cycle (sporozoites and exoerythrocytic hepatic stage), we have found a similar ER reactivity of anti-Pfgrp antisera (Kumar et al., 1989, 39th Annual Meeting of the Am. Soc. Trop. Med. Hyg., Hawaii). This protein is neither localized in the nucleus nor secreted from the cells expressing it. The sequence at the carboxyterminus of Pfgrp is SDEL as compared to KDEL in mammalian cells [8] and H D E L in yeast [31,35]. H D E L serves as a signal sequence for ER retention in the yeast and not in the mammalian cells [35]. Similarly mammalian sequence K D E L functions only in the homologous systems and not in the yeast [8,35]. Our observations suggest that the carboxy terminal sequence SDEL in malaria parasite could also function as a signal for

57

cellular retention and ER localization in the parasite. Malaria parasites cycle through reduced temperatures in the mosquito vector to elevated temperatures in the vertebrate hosts. Induction of Pfhsp upon heat shock and repression after removal of heat stress clearly demonstrate that stress responses, in particular heat shock, might play a key role in the hostparasite adaptation. Febrile response during blood-stage infection further emphasizes significance of heat shock responses in the adaptation of parasites.

Acknowledgements These studies were supported in part by the funds from Public Health Service award No. AI24704 (NK), John D. and Catherine T. MacArthur foundation (NK), USAID# DPE0453-A-00-4027-00-US (MA) and UNDP/ World Bank/WHO Special program for Research in Tropical Diseases (NK, MA). Authors also thank Dr. Hoovler of Life Cell Corp. for assistance with cryo-fixation of parasites for immuno-EM and Mr. Jim Bahre of Betagen for assistance with [3-emission imaging of Northern blots.

References 1 Lindquist, S. (1986) The heat-shock responses. Annu. Rev. Biochem. 55, 1151 1191. 2 Lee, A.S. (1987) Coordinated regulation of a set of genes by glucose and calcium ionophores in mammalian cells. Trends Biochem. Sci. 12, 20 23. 3 Pelham, H.R.B. (1986) Speculations on the functions of the major heat shock and glucose-regulated proteins. Cell 46, 959 961. 4 Chappel, T.G., Welch, W.J., Schlossman, D.M., Palter, K.B., Schlesinger, M.J. and Rothman, J.E. (1986) Uncoating ATPase is a member of the 70 kilodalton family of stress proteins. Cell 45, 3 13. 5 Welch, W.J. and Feramisco, J.R. (1985) Rapid purification of mammalian 70000-dalton stress protein: affinity of the proteins for nucleotides. Mol. Cell. Biol. 5, 1229 1237. 6 Chirico, W.J., Waters, M.G. and Blobel, G. (1988) 70K heat shock related proteins stimulate protein translocation into microsomes. Nature 332, 805 810.

7 Deshaies, R.J., Koch, B.D., Werner-Washburne, M., Craig, E. and Schekman, R. (1988) A subfamily of stress proteins facilitates translocation of secretory and mitochondrial precursor polypeptides. Nature 332, 800 805. 8 Munro, S. and Pelham, H.R.B. (1987) A C-terminal signal prevents secretion of luminal ER proteins. Cell 48, 899 907. 9 Bole, D.G., Hendershot, L.M. and Kearney, J.F. (1986) Posttranslational association of immunoglobulin heavy chain binding protein with nascent heavy chains in nonsecreting and secreting hybridomas. J. Cell Biol. 102, 1558 1566. 10 Kozutsumi, Y., Segal, M., Normington, K., Gething, M-J. and Sambrook, J. (1988) The presence of malfolded proteins in the endoplasmic reticulum signals the induction of glucose-regulated proteins. Nature 332, 462 464. 11 Kassenbrock, C.R., Garcia, P.D., Walter, P. and Kelly, R.B. (1988) Heavy-chain binding protein recognizes aberrant polypeptides translocated in vivo. Nature 333, 90 93. 12 Newport, G., Culpepper, J. and Agabian, N. (1988) Parasite heat-shock proteins. Parasitol. Today 4, 306 312. 13 Van der Ploeg, L.H.T., Giannini, S.H. and Cantor, C.I